CN1110779C - Intelligent management system for video frequency information - Google Patents

Intelligent management system for video frequency information Download PDF

Info

Publication number
CN1110779C
CN1110779C CN99123164A CN99123164A CN1110779C CN 1110779 C CN1110779 C CN 1110779C CN 99123164 A CN99123164 A CN 99123164A CN 99123164 A CN99123164 A CN 99123164A CN 1110779 C CN1110779 C CN 1110779C
Authority
CN
China
Prior art keywords
video
data
image
video data
module
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN99123164A
Other languages
Chinese (zh)
Other versions
CN1271146A (en
Inventor
戴维·R·马克迈克
帕特里克·O·努纳里
查尔斯·P·威尔逊
哥尔哈德·J·温特尔
哈里·E·克莱恩
威廉·T·恩古阳
森·林-柳
林恩·恩古阳
亚历克斯·K·奥耶昂
小克里斯·H·帕德森
戈登·W·史密斯
戴维·J·奥斯里
舍温·圣-舒·王
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sensormatic Electronics Corp
Original Assignee
Sensormatic Electronics Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from US08/741,983 external-priority patent/US5854902A/en
Priority claimed from US08/740,651 external-priority patent/US5862342A/en
Priority claimed from US08/741,715 external-priority patent/US5875304A/en
Priority claimed from US08/741,982 external-priority patent/US5875305A/en
Priority claimed from US08/741,650 external-priority patent/US6031573A/en
Priority claimed from US08/729,620 external-priority patent/US5822542A/en
Priority claimed from US08/740,628 external-priority patent/US5917958A/en
Priority claimed from US08/741,914 external-priority patent/US5828848A/en
Priority claimed from US08/740,627 external-priority patent/US5884042A/en
Priority claimed from US08/742,015 external-priority patent/US5909548A/en
Application filed by Sensormatic Electronics Corp filed Critical Sensormatic Electronics Corp
Publication of CN1271146A publication Critical patent/CN1271146A/en
Application granted granted Critical
Publication of CN1110779C publication Critical patent/CN1110779C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Abstract

A flexible video information storage and analysis apparatus stores a video information data base and a plurality of moving image content analysis algorithms for analyzing the video information in the data base. The selected algorithm is used to analyze video information in the data base. The apparatus records and displays plural streams of video information. Content analysis algorithms may be downloaded from a remote location. The apparatus takes appropriate actions in response to detected features in incoming video streams. Tile-based conditional-refresh video data compression is employed. A support structure efficiently supports several record-medium drive units. A tape back-up drive serves as an archiving mechanism for video data stored on hard disks. Recording of incoming video, and archiving and displaying of stored video, all proceed simultaneously.

Description

Intelligent management system for video and audio information
(the application is dividing an application of application 97199221.5)
Technical field
The present invention relates to closed circuit video safety monitoring system, especially, relate to and use machine intelligence to manage, explain and use the application of video information.So the system that is disclosed can be known as one " intelligent video information management " (IVIM) system here.Up to now, also IVIM field and other association areas are not made a distinction significantly, so the ultimate principle of the general application of the appearance technology that one of basic purpose of the present invention is the instruction intelligent video information manages.
Background technology
As everyone knows, when the video security system is provided, need to use a video camera to produce the vision signal that expression needs the position of security monitoring.In an exemplary systems, a part or all these vision signals are displayed on the video screen, so that the Security Officer monitors.In addition, as everyone knows, a part or all these video signal recordings on video tape, or are Security Officer's the statement evidence that provides support, or are applied to realize that " in real time " manually monitors or do not need " in real time " artificial occasion that monitors.
But as a storage media, tape has important disadvantages, when particularly the video security system has produced multitude of video information.A main problem is the tape that needs to preserve enormous quantity, particularly when needing the record signal that a large amount of surveillance camera produced.In addition, in a big system, need many video tape recorders, this has caused a large amount of capital consumption, also needs to be simultaneously these video tape recorder allocation space.Another problem is to change tape continually.
Retrieval institute information of interest also is the problem that needs are considered from the tape that is write down.The characteristics of video monitor are that the Video Supervisory Signal that most of tape write down less needs, because a still image in its general aisle of expression or similar place.A special sequence finding a critical event of expression is unusual difficulty, also is time-consuming, and it needs the staff to spend several hours or the long-time over these days dull signal of observing tape dullly and being write down.
When interested vision signal has just been recorded soon, thus current recording operation still at service recorder during the tape of this signal, just produced another problem.In this case, need to end recording operation, tape is shifted out and inserts another tape from tape cassete.Then, use an independent tape cassete to come this interested signal of playback.Clearly, this a process need tape cassete inconvenience and consuming time is handled, and a danger is arranged is can not write down interested new signal when the replacing tape cassete.
Attempt certain methods and overcome these shortcomings, but only obtained limited success up to now, perhaps produced other shortcoming.For example, well-known, use one " four multiplexers " will become a single dynamic image from the signal fused of four video cameras, it has four 1/4th part, and wherein each is exclusively used in an appropriate section of video camera.Write down the spatial reuse signal of this generation then, thereby on required memory capacity, realize 4 to 1 compression factor.But, this multiplexing image loss corresponding spatial resolution, and might weaken as the value of this recording picture of evidence or can influence subsequently observation.In addition, the record of multiplexing image can not solve and find the problem that interested sequence produced from recording tape.
In addition, as everyone knows, be that one is just being responded in the input of a human operator may of supervisory signal, perhaps respond, write down the monitoring video signal selectively being arranged in the signal that sensor device produced that detects incidents such as opening the door or window.This technology can reduce the total quantity of information that needs record, and need not preserve uninterested information in a large number, but have a kind of danger be may not write down those can not be easily or immediately detected critical event by sensor or human operator may institute.In addition, may produce unreliability and increase expense, when particularly starting record by human operator may to the dependence of external unit.
The application's the OPTIMA II video monitor multiplexer that the assignee introduced uses a more complicated technology, to pick out uninterested information before preserving.In OPTIMA II multiplexer, can form a pool image stream from the image of video camera from the respective streams of a plurality of video camera receiver, video picture information and by time division multiplex.The visual stream that will make up then outputs to a traditional video tape recorder to be recorded on the tape.This OPTIMA II multiplexer can carry out self-adapting time slot and distribute by corresponding input imagery stream is carried out the motion detection analysis in output stream, those are distributed more time slot from the image that the inlet flow that detects motion comes.Make in this way, the more relatively part in the system memory size is assigned to the image stream that comprises moving target, more may comprise important information thus.
With respect to traditional, based on for the tape monitoring video memory technology, OPTIMA II multiplexer has had a big improvement, but still needs to raise the efficiency and improve adaptability.
" MultiScop " video disk recording device that Geutebruck GmbH sells is a product that digital recording is applied to the problem of preserving video monitor information.This MultiScop system has used above-mentioned selectivity recording technique to make the uninterested information capacity minimum of record.In addition, by using a kind of traditional digital image compress technique, from record, some redundant informations have been got rid of.According to the date and time index, the perhaps information of coming random access to preserve according to the index of outside responsive alarm conditions of expression, from the video image that retrieval is preserved, there is the improvement of appropriateness in this system with respect to traditional system based on tape.
Although the MultiScop register has advantage in that there is no question about aspect the random access retrieval, a less-than-ideal characteristic of this register is when retrieving, and can end all recording operations.In order to reduce writing down and the possibility of unwritten new video information, MultiScop need to be programmed for alarm conditions are responded, thereby can automatically end retrieval, to turn back in the logging mode immediately.When not introducing this automatic discontinuation feature, or in default of alarm conditions, or, will not write down this new video information because do not select this characteristic.Although MultiScop has avoided the replacing of tape, but during search operaqtion whole, the record of MultiScop is ended, its end time ratio the independent VCR that a playback is arranged, to change the needed time of tape in based on the system of tape longer relatively.
Therefore people urgently wish to have an intelligent management system for video and audio information, and it can avoid the system of front peculiar, switching back and forth between record and playback mode of operation, and there are higher efficient and greater flexibility the while than the system known to the front again.
Higher efficient that the MultiScop system is provided and greater flexibility are very desirable.Especially, if uninterested information can be excluded, make the possibility minimum of missing important information again simultaneously from record, this is the most useful.In addition, also need more effective information retrieval technique.
Can believe that fully the MultiScop system has adopted the compression coding technology based on conversion, for example well-known Joint Photographic Experts Group, it can provide the compressibility factor of 40: 1 or 50: 1.For data storage capacity is that 2,000 megabyte and recording rate are 50 of per seconds, and the MultiScop system can write down the vision signal of a hour more point.A preferable method is for a given data storage capacity, to provide one to have the more technology of high compression degree, so that higher recording rate and/or longer writing time to be provided.
To being used for a video data compression technology of IVIM and video monitor operation, if compression and decompression operand separately is roughly the same, this video data compression technology can comprise an embodiment, it adopts a general processor device to realize the compression and decompression computing simultaneously, and this characteristic of so this video data compression technology is very desirable.In addition, if this compress technique can make the machine analysis and the reverse playback of video image stream talk about very easily, just more helpful.
Inventor of the present invention is also noted that current widely used video data compression technology do not utilize the special characteristics of monitoring video signal, and this special characteristics can realize higher compression factor, and enough image qualities that is used to monitor are arranged again simultaneously.For example, well-known MPEG monitors can be provided higher compression factor and provide users broadcasting to use required high picture quality, but its complicacy has increased.In addition, the technology of MPEG type is perhaps compressed these characteristics of a plurality of input video streams or the like simultaneously to reverse playback, is immalleable.
Not only memory storage has proposed very high requirement to the bulk information that occurs in a video signal flow (useful or other) to digital data, and to those before being saved on the memory storage, perhaps very high requirement has also been proposed from the circuit structure that sends and handle video information after a memory storage is retrieved out.When needs write down many video signal flows simultaneously, this was a kind of situation of frequent appearance, and these require just higher.In addition, video information is compressed the effective use that has promoted memory capacity, but increased the burden of transmission and treatment circuit.
Another shortcoming of existed system is this needs: for the video data that the Visitor Logs device is preserved, the user must actually arrive register.In addition, user's selection of searching video data is limited on the ability that this register provides.
Another design problem relevant with the video data storage is the total memory capacity that need provide.Existing system has only comprised a single internal hard disk drive, and the capacity that the volume ratio internal hard disk drive that needs as the user is provided just needs to use exterior storage medium driver for a long time.Suppose that a register that comprises several inner drives unit can allow bigger memory capacity and other advantages are provided, it also is desirable being installed in the inner drive unit in the structure of making easily and assembling safely.Further, should arrange other internal electrical components of a digital video recorder properly, so that the manufacturing expense minimum can make things convenient for catching and handling of video information simultaneously again.
Suppose that an intelligent management system for video and audio information has the effective hard disc storage capacity bigger than existing device, when the time past tense with need catch and write down more current video information the time, even such disk, its memory capacity also are limited, and need reuse this disk.So even used fixed disk rather than the basic storage media of video tape conduct, permission is forever preserved on dismountable medium or " file " preservation video information also is desirable.The user friendly management of ability energy that system files, and can coincide mutually with the other system feature effectively.
Though unavoidablely probably be, the increase of the dirigibility of a register system will be attended by the increase of complicacy, and it is difficult that the use of system but should not become, and this point also is important.When judging whether an IVIM system has carried out its desired function satisfactorily, the comfort level of use may become a vital factor.
Intelligent management system for video and audio information disclosed herein has been represented videograph and personal computer technology's a kind of brand-new fusion, and it can improve the ability of catching with retrieve video information comprehensively.The option that provides certain user to be provided with, this has strengthened the dirigibility of system and has allowed the user that system is adjusted to the special application of adaptation.
Proposed many technology and come the filtering video image data, to detect the key character of the represented image of pictorial data.The technology of delivering about this field comprise following these: M.P.Cagigal, or the like in August, 1994, the 33rd the volume, the 8th phase Optical EngineeringThe 2810-2812 page or leaf on " the target travel character representation of low-light level rank image " delivered; S.J.Nowlan or the like is in Dec, 1994, and the 11st rolls up, the 12nd phase J.Opt.Soc.Am. AThe 3177-3200 page or leaf on " the wave filter preference pattern that is used for motion segmentation and gait synthesis " delivered; T.G.Allen or the like is in July, 1994, and the 33rd rolls up, the 7th phase Optical EngineeringThe 2248-2254 page or leaf on " in the image sequence the multiple dimensioned of moving object detection being approached " of delivering; M.-P.Dubuisson or the like is in nineteen ninety-five, the 14th volume International Journal of ComputerVisionThe 83-105 page or leaf on " in the complicated outdoor environment profile of motion video extract " delivered; M.Bichsel is in November, 1994, and the 16th rolls up, o. 11th IEEE Transactions on Pattern Analysis and Machine IntelligenceThe 1138-1142 page or leaf on " in a static environment, the cutting apart the moving object of simple connection " of delivering; M.Irani or the like is 12:1's in 1994 International Journal of ComputerVisionThe 5-16 page or leaf on " the closing calculating " of delivering with transparent motion; I.J.Cox is 10:1's in 1993 International Journal of ComDuterVisionThe 53-66 page or leaf on " review of the corresponding statistics correlation technique that is used to move " delivered; D.Koller or the like is 10:3's in 1993 International Journal of ComputerVisionThe 257-281 page or leaf on " the target following in the monocular image sequence of road traffic environment " delivered based on model; J.H.Lee or the like is in May, 1994, and the 40th rolls up, the 2nd phase IEEE Transactions on Consumer ElectronicsThe 2154-2160 page or leaf on " a HDTV signal being carried out the VLSI chip of estimation " of delivering; T.Koivunen is in August, 1994, and the 40th rolls up, the 3rd phase IEEE Transactions on Consumer ElectronicsThe 753-760 page or leaf on " motion detection of an interleaved vision signal " delivered; S.-I.Jang or the like is in May, 1994, and the 40th rolls up, the 2nd phase IEEE Transactions on Consumer ElectronicsThe 145-153 page or leaf on " to the motion of a video camera and focus on a real-time identification method of the bad image blurring that causes " delivered; But the filtering technique major part of these recommendations only limits to analyze still image, rather than attempts to obtain information by analyzing a dynamic image stream, still is fewer so the commerce of the filtering technique of these recommendations is used.
Wish to improve to being included in management of information and the use in the single or concurrent video image stream.Also special hope can allow a user of a video monitoring system, and perhaps other users that preserve dynamic image information can manage enter video information more neatly and use input and the video flowing of being preserved more neatly.
Except above-described OPTIMA II multiplexer, other are exclusively used in the fundamental purpose that the detected characteristics of video information is made from the device of dynamic response is to detect motion, and just excites alarm conditions when moving when detecting.At U.S. Patent number 4,737,847 (Araki or the like); 4,160,998 (Kamin); 4,198,653 (Kamin); 3,988,533 (Mick or the like); 4,081,830 (Mick or the like); 3,743,768 (Copeland); With 4,249, a kind of like this device is disclosed among 207 (Harmon or the like).
Except that in the prior art the expectation those, the detection and the finishing automatically of function that machine intelligence are expanded to the video information feature are very desirable.
Summary of the invention
An object of the present invention is to provide a device, it can allow the user to select one or several required options from a plurality of available video data options by using this device.
To be the user that can allow a kind of like this device be adjusted to the feature that can adapt to by the represented known or expectation of information to be analyzed with the video information analytical algorithm of a selection to another object of the present invention.
Another object of the present invention provides an intelligent management system for video and audio information with mode of operation that some users can select, and it has improved the convenience of the dirigibility and the use of this system.
Another object of the present invention provides a distributed video information database system.A further purpose provides so a kind of ability, even the position that the user who wishes retrieve video information stores away from video information, he also can carry out content-based retrieval to these video informations.
Another object of the present invention provides an intelligent management system for video and audio information, and its detects the key character of an enter video information stream, and during interested feature, takes suitable action to come this is responded detecting.
Another object of the present invention provides a method and a device, highly effectively compresses with the video data to the stream of expression dynamic video image.
Another object of the present invention provides a video data compression technology, and it can be good at adaptive video and monitors Secure Application.
Another object of the present invention is to use a kind of form that is suitable for the machine analysis of data that compressed video data is provided.
Another object of the present invention is to use a kind of form that can be convenient visual stream be carried out reverse-play compress a video image stream.
Another object of the present invention provides a kind of catching of video information that be used for, and form and interim technology of preserving are to simplify the processing and the route of video information in an intelligent management system for video and audio information.
Another object of the present invention is will be installed in a recording medium actuator unit in the digital video recorder supporting structure is provided for several.
Another object of the present invention is to use a structure that is convenient to make and assemble that a kind of like this supporting structure is provided.
Another object of the present invention is to provide an effective and good structure of cost performance for being included in a data processing circuit structure in the digital video recording device.
Another object of the present invention provides the intelligent management system for video and audio information with video information archival capabilities.
Another object of the present invention provides an IVIM system, and wherein its archival capabilities is flexibly and is user-friendly.
Another object of the present invention provides an intelligent management system for video and audio information, wherein uses a kind of intelligence and useful form to come user to this system that information about the state of this system is provided.
Another object of the present invention is sampling parallel processing and a multi-tasking in an intelligent management system for video and audio information, so that can also eliminate the interference between the function or make its minimum when this system carries out basic function.
Another object of the present invention provides identification, and the technology of the appropriate section of sign and index video information is to support neatly and to preserve effectively and retrieve video information.
To achieve these goals, the invention provides one and analyze the method that a series of frames of video information detect the change of described frames of video information, the method comprising the steps of: show the step by an image of a represented target of described frames of video information in an image plane; Operate a graphic image element with the zone of selection, thereby specify the step of monitored object corresponding to the image plane of this target; Select the step of the part of described appointed target; And in the described frames of video information of continuous review, judge with the corresponding part of the described chosen part of described designated target whether described designated target rests on the step in a precalculated position in the described frames of video information.
The present invention also provides a video data save set, comprising: the device of video data that is used for receiving an image of an image plane of expression; Be used to select not need to preserve the drop device of a part of the image plane of its data; And save set, described drop device is responded, be used for preserving the video data of described reception, not with the corresponding so a part of video data of the described chosen part of described image plane; Display device is used to show and the corresponding image of being represented by the video data of described reception of an image; In the video data of described reception, be not stored in described this save set with the corresponding part of the described chosen part of described image plane.
The present invention also provides the method for preserving video data, comprises step: the video data that receives an image in image plane of expression; Not need to select to preserve the part of the image plane of its data; With on a recording medium, only preserve those not with the part of the corresponding receiving video data of described chosen part of described image plane.
The present invention also provides an apparatus for processing of video signals, and this apparatus for processing of video signals comprises: the device that is used for receiving a vision signal of an image that is illustrated in an image plane; Be used to select the part of image plane to improve the selecting arrangement of contrast; With improve device, link to each other with described receiving trap, described selecting arrangement is responded, be used to expand a dynamic range with the part of the corresponding described receiving video signals of described chosen part of image plane, and be used to export one and optionally improve vision signal.
Description of drawings
Figure 1A and 1B use the form of block diagram to represent according to a distribution type intelligent video information management system provided by the present invention.
The form of Fig. 2 functions of use module has been represented to form as video analysis of a funtion part of the system of Figure 1A and 1B and save set.
Fig. 3 is according to a block diagram provided by the present invention and electronic unit of an employed videograph/personal computer (VR/PC) unit in the system of Figure 1A and 1B.
Fig. 4 is an image representation of a board structure of circuit being adopted of the VR/PC unit of Fig. 3.
Fig. 5 is the frame of this VR/PC unit of explanation and a skeleton view of front panel.
Fig. 6 is a front view (FV) of the front panel of this VR/PC unit.
Fig. 7 is the rear view of this VR/PC unit.
Fig. 8 is an isometric projection view that is provided in this VR/PC unit, be used to support a disk drive unit and a similar structural support component.
Fig. 9 A and 9B are exploded views of this disc driver supporting structure of Fig. 8.
Figure 10 is an isometric projection view of this driver supporting structure of Fig. 8, as assembled and be used to support hard disk drive units and other medium actuator units.
Figure 11 be when this VR/PC cell operation under an on-the-spot display mode, and comprise when cursor excites the control area printing of a screen display that is produced by this VR/PC unit.
Figure 12 be when this VR/PC cell operation under a playback mode, and comprise when cursor excites the control area printing of a screen display that is produced by this VR/PC unit.
Figure 13 A and 13B are theory diagrams that is presented at a simulation process/multiplex circuit plate among Fig. 4.
Figure 14 is a block diagram of a numeral/compression/analysis circuit plate of showing in Fig. 4.
Figure 15 is illustrated in an analog to digital conversion that shows among Figure 14, the functional block diagram of field buffer and video bus control module.
Figure 16 has illustrated some details of this analog to digital translation function that shows in Figure 15.
Figure 17 has illustrated some details of a video digitizer control function that shows in Figure 15.
Figure 17 A uses the form of flow graph to illustrate the processing of carrying out in the controller logic block diagram that shows in Figure 17.
Figure 18 has illustrated some details of a field buffer that shows in Figure 15.
Figure 19 has illustrated some details of a video output control function that shows in Figure 15.
Figure 20 A and 20B are illustrated respectively in the raster scanning form and the rectangle data tile form of an image plane (field) of the video data that forms in this VR/PC unit.
Figure 21 illustrates the form of a rectangle data tile that is adopted this VR/PC unit.
Figure 22 represents the form of the minimum code unit (MCU) that this VR/PC unit is adopted.
Figure 23 represents 4: 1: 1 color video data format that this VR/PC unit is adopted.
Figure 24 represents the employed color data form of each MCU.
Figure 25 is an image representation that sends one first pattern of video data on a shown video bus of Figure 14.
Figure 26 uses the form of image to represent to send one second pattern of video data on this video bus.
Figure 27 uses the form of image to represent to send a three-mode of video data on this video bus.
Figure 28 uses the form of image to represent to be used for a data grating of the raster scanning line of the video data that sends on this video bus.
Figure 29 is an image representation of a form of the video data data tile that is sent out on this video bus.
Figure 30 represents to be included in a form of the data bit in the head of the video data field that is sent out on this video data bus.
The form of Figure 31 functions of use block diagram is illustrated in a processing that the output control logic module is performed that is shown among Figure 19.
Figure 31 A uses the form of a flow graph, and expression is by the processing that carry out, that be used to form a data of the output control logic module of Figure 31.
Figure 31 B uses the form of flow graph, illustrate carry out by the output control logic module of Figure 31, be used to use line and data tile header data to come the processing of control line and the output function of data tile.
Figure 32 uses the form of flow graph, illustrates the processing of the raster scanning line of being carried out by the output control logic module of Figure 31, be used to produce video data.
Figure 33 uses the form of flow graph, illustrates the processing by the output control logic module rectangle data tile that carry out, that be used to form video data of Figure 31.
Figure 34 uses the form of flow graph, illustrates the processing of sequence of a cross occurrence of the raster scanning line carried out by the output control logic module of Figure 31, be used to produce video data and rectangle data tile.
Figure 35 is the state machine diagram of operation of the output control logic module of an expression Figure 31.
Figure 36 is a theory diagram of digital signal processing chip of expression and related circuit structure, and it is the control function of video front processing and initial treatment that this digital signal processing chip and related circuit structure are carried out together for this video data compression.
Figure 37 uses the form of block diagram, the specific detail of the live video data analysis processing module of expression Figure 14.
Figure 38 uses the form of block diagram, illustrates the specific detail of the on-the-spot display process module of Figure 14.
Figure 39 is a schematic diagram of main software module that the operation of the control dsp chip of Figure 36 and 14 is programmed.
Figure 39 A uses the form of flow graph, represents the map generator module of the software that this is used to control DSP " changing the data tile ".
Figure 40 is the flow graph of the pci interface Drive Module of the expression software that is used to control DSP.
Figure 41 A and 41B form the flow graph of compressor reducer administration module that an expression is used to control this software of DSP together.
Figure 42 uses the form of flow graph, and expression is by the performed video data data tile comparing function of this control DSP.
Figure 43 A is an image of representing an employed in the present invention video data data tile form.
Figure 43 B and 43C use image to represent this video data data tile form of Figure 42 A is divided into 1/4th, and represent a checker plank sampling template that is applied to each data element of 1/4th.
Figure 44 represents to be used for a data form of the compressed video data that the front-end processing plate by Figure 14 produced.
Figure 45 represents to control the skeleton diagram of main software parts of the work of the mainboard microprocessor that is presented among Fig. 3.
Figure 46 A is a COM (component object model) of the systematic direction device module of this mainboard of expression CPU software.
Figure 46 B is the COM of general type of the miscellaneous part of this CPU software of expression.
Figure 47 is the flow graph that an expression is used for safety management person's software module of this mainboard CPU.
Figure 48 represent one can be constantly by the typical remote object that illustrates with concrete example in the processing of carrying out by this mainboard CPU.
Figure 49 represents to handle an interactive software object between the physics front panel of VR/PC unit of this mainboard CPU and Fig. 5.
Figure 50 uses the form of flow graph, has illustrated that of being used for mainboard CPU sets up the manager software module.
Employed index data form when Figure 51 illustrates video data and this VR/PC unit is on the hard disk that video data is kept at wherein.
Figure 52 uses the form of flow graph, illustrates a video that is used for mainboard CPU and preserves the process software module.
Figure 53 uses the form of flow graph, illustrates a video that is used for this mainboard CPU and searches the process software module.
Figure 54 uses the form of flow graph, illustrates " carry out and search " module of Figure 53.
Figure 55 uses the form of flow graph, illustrates the performed processing when carrying out video playback operation by this mainboard CPU.
Figure 56 uses the form of flow graph, illustrates " decompressor again " module of Figure 55.
Figure 57 uses the form of flow graph, illustrates " decompression reference picture " module of Figure 56.
Figure 58 uses the form of flow graph, illustrates " stopping decompressor " module of Figure 55.
Figure 59 uses the form of flow graph, illustrates " decompression difference image " module of Figure 56.
Figure 60 uses the form of flow graph, illustrates the part of the processing of Figure 59 relevant with the particular video frequency data block.
Figure 61 uses in this VR/PC unit, be used to reformulate an image representation of a technology of a compressed video data field.
Figure 62 A and 62B illustrate the substitute technology that is used to refresh a video display buffer and VR/PC unit respectively.
Figure 63 A and 63B illustrate the low resolution and the high-resolution video data that use possibly respectively and preserve and playback technology in this VR/PC unit.
Figure 64 A and 64B are when the effect of this VR/PC unit provides a reverse playback function, a picture specification that is used to reformulate a technology of compressed video data.
Figure 65 A and 65B are the high level flow graph representations of this reverse playback decompression technique of Figure 64 A and 64B.
Figure 66 uses the form of flow graph, illustrates " setting up the back to historic buffer " module of Figure 65 A and 65B.
Figure 67 is as the part of this reverse decompression, is used and decompresses and a flow graph representation of this process of recompile reference picture data.
Figure 68 uses the form of flow graph, illustrates " the " ﹠amp of Figure 67; Recompile data tile " module.
Figure 69 A and 69B use the form of flow graph, illustrate this mainboard CPU is programmed a command process part of carrying out an interactive software module with this front-end processing electronic installation.
Figure 70 A and 70B programme a flow graph representation that carries out the status message processing section of an interactive software module with this front-end processing electronic installation to this mainboard CPU.
Figure 71 is a flow graph representation of being handled by the video data that this mainboard CPU carries out the video data that sends to from this front-end processing electronic installation on it.
Figure 72 illustrates when this mainboard CPU begins inputting video data is recorded on the hard disk of this VR/PC unit, performed processing.
Figure 73 illustrates when archive management person software module is handled the search request, performed processing.
Figure 74 A and 74B use the form of flow graph, illustrate at this mainboard CPU with video data recording at a dismountable file record medium and/or from this dismountable file record medium during the playback video data performed processing.
Figure 75 illustrates, and this mainboard CPU responds to an expression of alarm conditions, by the performed processing of this mainboard CPU.
Figure 76 be provided as a normal response that alarm conditions are made, a video data script holder records an image representation of a distribution of time slots sequence.
Figure 77 uses the form of flow graph, illustrates the image processing facility feature with this VR/PC unit, by the processing of this mainboard CPU execution.
Illustrate to Figure 78 principle operation with this mainboard CPU of the operation of this graphical user interface.
Figure 79 uses the form of flow graph, and illustrating programmes to this mainboard CPU selects mainly to plan the software of option.
Figure 80 uses the form of flow graph, illustrates at this mainboard CPU to set up when selecting a vacation that is used to plan the operation of this system the operation of being carried out by this mainboard CPU.
Figure 81 is one and is used to define the diagrammatic representation of " daytime " and the screen display in " night " period at interval of 24 hours calendar day.
Figure 82 uses the form of flow graph, illustrates the day/night distribution function that this mainboard CPU is programmed and carries out graphic presentation in Figure 81.
Figure 83 uses the form of flow graph, illustrates when this mainboard CPU planning system is operated, and this mainboard CPU is programmed carry out the option selection.
Figure 84 uses the form of flow graph, illustrates when this mainboard CPU produces the description of the system operation be used to control pre-planning the programming that this mainboard CPU is carried out.
Figure 85 illustrates control by a data form of a formation of the system operation of pre-planning.
Figure 86 is the type that illustrates by the planning description list that processing produced of Figure 84.
Figure 87 uses the form of flow graph, illustrates this mainboard CPU is programmed, to come the control system operation according to the projecting parameter that pre-sets.
Figure 88 uses the form of summary, illustrates at this mainboard CPU to carry out in the typical image analysis algorithm, by the performed operation of this mainboard CPU.
Figure 89 uses the form of flow graph, illustrate Figure 88 image analysis tool operation selection and set up part.
Figure 90 A-D is that a girth that is provided in expression parameter setting operation and disclosed here this system is violated the detection image analysis tool.
Figure 91 uses the form of a flow graph, illustrates carrying out the previously needed preparation process of image analysis.
Figure 92 uses the form of flow graph, illustrates an initial phase to a performed image analysis algorithm of input video image.
Figure 93 A illustrates the performed processing of scene photo picture analysis processor of the front end video board of Figure 14, and Figure 93 B and 93C are the relevant treatment steps that is continued execution by the video board controller.
Figure 93 D and 93E use the form of flow graph, further illustrate the processing that is continued execution by this scene photo picture analysis processor.
Figure 94 is the flow graph of the image analysis of a video image stream that illustrates the initialization step that is continued to carry out and be applied to a regeneration.
Figure 95 A-95C illustrates the flow graph that carry out, that be used for an image analysis algorithm is applied to the processing of a video flowing that is reproduced by this mainboard CPU.
Figure 96 is an employed picture specification in explaining a curve fitting analysis performed to video flowing that is reproduced.
Figure 97 uses the form of flow graph, illustrates a girth is violated the process that image analysis algorithm is applied to a video flowing that is reproduced that detects.
Figure 98 uses the form of flow graph, illustrates the responding an of incident, and excites the process of a live video analytical algorithm.
Figure 99 uses the form of flow graph, and a video data compression technology that is provided in an alternate embodiments of the present invention is provided.
Figure 100 uses the form of flow graph, illustrates a decompression technique performed according to the compressed data of the processing of Figure 99.
Figure 101 A and 101B use the form of flow graph, illustrate to allow a user to regulate the system operation that video shows the setting of temporal resolution and/or spatial resolution.
Figure 102 uses the demonstration of flow graph, illustrates to allow a user to regulate the operation of this system of video data compression parameter.
Figure 103 A-103C is the flow graph that illustrates system operation and image analysis algorithm is downloaded to another device from a system and device.
Figure 104 illustrates being responded by a detected condition of live video analytical algorithm, optionally sends a flow graph of an input video stream.
Figure 105 illustrates being responded by a detected predetermined characteristic of live video analytical algorithm, changes a flow graph of video camera stream saving sequence.
Figure 106 uses the form of flow graph, illustrates being responded by the detected predetermined characteristic of a primary scene video analysis algorithm, excites the process of a secondary scene video flowing analytical algorithm.
Figure 107 responds to detecting alarm conditions, illustrates a flow graph of a video data compression parameter of automatic change.
Figure 108 is to being responded by a detected predetermined condition of live video analytical algorithm, illustrating a flow graph that is dwindled the operation of view by self-actuating software.
Figure 109 responds to analyzed a detected predetermined condition by live video, illustrates the flow graph of operation that this system excites a motion of a video camera automatically.
Figure 110 is when two different live video analytical algorithms detect the corresponding mobile condition, illustrates the flow graph that this system takes the operation of a predetermined action.
Figure 111 is when this video camera that this video flowing is provided when mobile or when detecting alarm conditions, illustrates the flow graph that this system forbids the operation of live video flow analysis.
Figure 112 illustrates this system to preserve the flow graph of a video camera of an input video stream of expression generation in the operation of a data that moves.
Figure 112 A be illustrate this system, forbid an image analysis algorithm is applied to a flow graph that is reproduced the operation of video flowing by that mobile camera produced.
Figure 113 is when detecting alarm conditions, illustrates the flow graph that this system increases the operation of a set throughput rate.
Figure 114 is that to illustrate that this system provides different inputting video data stream be a flow graph of the operation of different compression parameters.
Figure 115 illustrates according to whether detecting alarm conditions, and this system changes a flow graph of the operation of a display buffer renewal technology.
Figure 116 is the record that illustrates about video data stream, a flow graph of the priority of regeneration and file.
Figure 117 A and 117B are the flow graphs that illustrates the alternative archive operations pattern of this system respectively.
Figure 118 illustrates this system to reduce the flow graph that speed that incoming video signal is saved prevents from excessively to move the operation of an archive operations.
Figure 119 illustrates the flow graph that this system preserves the operation of the relevant data of the vision signal of being filed with the front.
Figure 120 is a flow graph that illustrates the operation of an automatic camera diagnostic function.
Figure 121 illustrates the flow graph that this system provides the operation of a system state that is shown to this user.
Figure 122 illustrates this system automatically to search and show a flow graph with the operation of the produced simultaneously regeneration video flowing of current regeneration video flowing that is shown.
Figure 123 uses the form of a flow graph, illustrates the operation that this system writes down incoming video signal simultaneously and shows the video image stream of a regeneration.
Figure 124 illustrates the flow graph that this system shows the operation of two video image streams that are reproduced simultaneously.
Figure 125 illustrates the flow graph that this system is recorded in the video data index information two operations on the different hard disks simultaneously.
Figure 126 illustrates this system provides the operation of pre-alarm buffering for a video image stream that is recorded with its speed that descends in time a flow graph.
Figure 127 is the machine analysis of this system of explanation according to the video information that is reproduced, and changes and/or a flow graph of the operation of complementing video data head information.
Figure 128 (a) and (b) illustrate the image enhancement facility of a user-operable together.
Figure 129 be provided for a user, with search the relevant screen display of video information be recorded.
Figure 130 be provided for a user, with search the relevant screen display of video information be recorded.
Figure 131 be provided for a user, with search the relevant screen display of video information be recorded.
Figure 132 be provided for a user, with search the relevant screen display of video information be recorded.
Figure 133 be provided for a user, with search the relevant screen display of video information be recorded.
Figure 134 is provided for this user, allows this user to select a screen display setting up operation.
Figure 135 is provided for this user, allows this user to select a screen display of a video data recording pattern.
Figure 136 is provided for this user, allows this user to select a screen display of a video data recording pattern.
Figure 137 is provided for this user, allows this user to sign in to a screen display of this system.
Figure 138 is provided for a user, allows this user an authorized users to be added to a screen display of this system.
Figure 139 is provided for a user, and allowing to select will be by a screen display that is authorized to the systemic-function of user capture.
Figure 140 is provided for a user, and allowing to select will be by a screen display that is authorized to the systemic-function of user capture.
Figure 141 is provided for a user, and allowing to select will be by a screen display that is authorized to the systemic-function of user capture.
Figure 142 is provided for a user, and allowing to select will be by a screen display that is authorized to the systemic-function of user capture.
Figure 143 is provided for a user, and allowing to select will be by a screen display that is authorized to the systemic-function of user capture.
Figure 144 is provided for this user, constructs a screen display of the security function of this system.
Figure 145 is provided for a user, a screen display of the video camera that allows this user that name is assigned to link to each other with this system.
Figure 146 is provided for this user, allows this user name to be assigned to a screen display of the alarm signal output that is provided by this system.
Figure 147 is provided for a user, allows this user name to be assigned to a screen display of the alarm signal input that can be received by this system.
Figure 148 is provided for this user, a screen display that allows this user to select in the facility function that some these systems can carry out.
Figure 149 is the screen display that system status information is provided to this user.
Figure 150 and Figure 151 are provided for this user, allow this user video display window to be assigned to the screen display of a plurality of video cameras that link to each other with this system.
Figure 152 is provided for this user, allows this user to specify a screen display of calendar date as the vacation that is used for the planning system operator scheme.
Figure 153 is provided for this user, allows the screen display of this user's adjusting about the parameter of an image analysis algorithm, and wherein this image analysis algorithm is used to detect the variation of being flowed the brightness of a represented scene by a video image.
Figure 154 is provided for this user, allows the screen display of this user's adjusting about the parameter of an image analysis algorithm, and wherein this image analysis algorithm is used for detecting the motion flowed a represented scene by a video image.
Figure 155 is provided for this user, allows this user to regulate a screen display about the parameter of an image analysis algorithm, and wherein this image analysis algorithm is used to detect a moving object and whether has passed through a perimeter border.
Figure 156 is provided for a user, allows the screen display of this user's adjusting about the parameter of an image analysis algorithm, and wherein this image analysis algorithm is used to detect the migration of an interested object.
Figure 157 is provided for a user, allows this user definition to be used for a screen display of tabulation vacation of pre-planning system operation modes.
Figure 158 is provided for this user, allows the screen display of this user for each the plan record pattern in a plurality of video cameras.
Figure 159 is provided for a user, allows this user definition about by the screen display of the planning horizon of the system operation modes of pre-planning.
Figure 160 is provided for a user, allows this user definition about by the screen display of the planning horizon of the system operation modes of pre-planning.
Figure 161 is provided for a user, with a screen display relevant with the operation of this image enhancement facility that illustrates (b) at Figure 128 (a).
Figure 162 is provided for this user, allow this system of this user command abandon with an image plane in a screen display of a corresponding video data of being selected by the user of part.
Figure 163 be one with the similar on-the-spot display screen format of Figure 11, except a live video zone of this screen is divided into one 1/4th structure, and wherein each display window has the record expression.
Figure 164 be provided for a user, with select the screen display that the archive operations pattern is relevant.
Figure 165 is an alternate formats of the demonstration of Figure 164.
Figure 166 be provided for this user, be stored in a screen display that the video data on the detachable file record medium is relevant with searching.
Figure 167 is provided for this user, allows to visit the screen display that the front is recorded in an index information database on the detachable file record medium.
Figure 168 is provided for this user, comes update contruction that a screen display of a file medium drive assembly tabulation of this VR/PC unit is arranged.
Figure 169 is provided for this user, allows to select a screen display that will be used for searching the image analysis algorithm of this video database that is stored in this VR/PC unit.
Figure 170 is and the similar screen display of a screen display shown in Figure 169, but pull-down menu tabulation image analysis algorithm can being selected by this user also has been described.
Figure 171 and 172 allows this user to select, and with relevant by the system operation modes of pre-planning, be applied to the screen display of an image analysis algorithm of an on-the-spot input video stream.
Figure 173 is to exciting " printing " option that provides in Figure 161 to respond an example of a hard copy print screen that is produced.
Embodiment
The IVIM systematic review
Figure 1A and 1B have represented one by total label 500 general surveys represented, intelligent video information management (IVIM) system.A plurality of positions and a plurality of commercial company have been crossed over by this system 500.For example, this commercial company may comprise a multiple-limb bank 502 and a multipoint retailer 504.This bank 502 and this retailer 504 provide service by IVIM system 506 and 508 respectively, these two IVIM systems 506 and 508 be each other standard independently, but optionally interact with a host node facility 510 that is provided by a security service tissue 512 at least.For example, this service organization 512 may provide and install and service intelligent management system for video and audio information and other video security systems.
Preferably, the form of this host node 510 is one and can be a provide support principal computer of function and Downloadable software resource of IVIM system 506 and 508.Although Figure 1A and 1B have only shown two IVIM of commercial company, wish that still this host node 510 can support that a large amount of wherein each maintains the commercial company of its IVIM system.This commercial company may cross over the whole U.S. (for example) and other countries and in all places and the other countries of the whole U.S. branch offices be arranged all.Can imagine that this system disclosed herein also is used for many other types company except being used for bank and retailer.
Shown in Figure 1B, each in the IVIM system 506 and 508 comprises a local node 514 that the supervision and management function is provided for corresponding IVIM system.Each local node 514 links to each other with this host node 510 through a corresponding data communication channel 516.For example, each data communication channel 516 may be made up of a private communication channel, and perhaps this channel 516 may be set up according to a request based on dialing.Preferably, this local node 514 uses the pc hardware and the software of standard to realize, and has increased a lot of new softwares, will discuss below about the ability of new software.
The critical component of each IVIM system is video analysis and the preservation unit 518 that links to each other with corresponding local node 514 by data communication path 519.Be connected with one or more above each unit 518, be represented as the video camera of video camera 520-1 to 520-N.Each video analysis and preservation unit 518 provide the preservation to the video stream that is produced by the video camera 520 that is connected thereto, and analyze and optionally retrieval.
The number that is connected to the video camera of each unit 518 is different for each analysis with preserving the unit.For example, the number of this video camera can be from 1 to 12 even more.As will be descried, a preferred implementation of this unit 518 will support number many to 16 video camera.
What also should be realized is that the analysis in each IVIM system also is different with the number of preserving unit 518.Although in Figure 1A and 1B, only shown that each IVIM system comprises the situation of 2 or 3 unit 518, the actual number of analysis in each commercial company and preservation unit can be tens or higher.
With each analysis and preserve unit 518 relevant be as by dotted box 522 typical miscellaneous parts represented, that in the video security system, found.Other safety components 522 that should be correlated with each unit 518 may be different to each unit.The example of a this miscellaneous part is by in the corner, lower-left of example and Figure 1A, and comprises two video display monitor 524, a plurality of can be by mechanically excited alarm sensor 526 and a camera control unit 528.
Optionally from this unit 518, guided by these video camera 520 video signal flows that produce and that in this analysis and preservation unit 518, received respectively to be presented on this monitor 524.As will be descried, can use multiple form, comprise full screen, the window that perhaps only occupies the part of this image plane shows this signal.A plurality of display windows can be formed on one or two displays 524, so that a plurality of video flowing is simultaneously displayed on the single video display 524.Preferably, display 524 is traditional displays, for example from the NTSC monitor model JC-1215MA of NEC Corporation and/or from the SVGA monitor MODEL C 1591E of PANASONIC.The advantage that this analysis of new system disclosed herein and preservation unit 518 are provided is that this unit 518 is carried out in the conventional video surveillance system by a video flowing distributed function that video switch is performed.So, in an actual embodiment of this IVIM system, in " other safety system components " 522, do not need to comprise video switch.
Preferably, this alarm sensor 526 be traditional, be used for test example as opening or close door, window, display casing or the like incident, and be used to produce the sensor of the signal of this incident of expression and alarm signal.This alarm signal is provided to this analysis and preserves unit 518 and be provided to camera control unit 528.
For example, this camera control unit 528 can be a traditional device, for example can buy from the assignee of this application " TOUCHTRACK ' R " video camera control pad.This camera control unit 528 and this analysis and preservation unit 518 are connected, and are used for exchange data messages between them.
Here suppose that some video cameras 520 are transportable.That is, some video cameras 520 are traditional video cameras, and for example by the dome that the assignee sold (dome) video camera of this application, it has the direction in a visual field of regulating with control signal.Movable camera in the video camera 520 also comprises the video camera that can move along a track.Typically, in video camera movably, its convergent-divergent and focus on to be provided with and also control by control signal.As 530 represented, this camera control unit 528 is organized in this video camera 520 movably that video camera provides control signal.
Also supposition, for the visual field, some video cameras in the video camera 520 are fixing.Should be appreciated that can imagine that all can be moved with the video camera that preservation unit 518 links to each other with a particular analysis, perhaps all video cameras can be fixed.
As 532 represented, this analysis and preserve unit 518 and this camera control unit 528 is joined together mutual swap data.Can organize this camera control unit 528 like this, so that or the output from alarm sensor 526 responded, perhaps a data message from this analysis and preservation unit 518 is responded, control signal automatically is sent to selecteed video camera in this transportable video camera 520 through control signal path 530, so that an incident that is detected by this alarm sensor 526 or an incident that is detected by this analysis and preservation unit 518 are responded, this movable camera is automatically positioned.
Above-described optional feature structure 522 only is an example, and can carry out many variations to it.For example, the number of monitor can be reduced to one or be increased to than 2 big numbers.As another possibility, can omit this alarm sensor 526 and this camera control unit 528.Especially, be appreciated that if all video cameras 520 are fixed, will be without any need for camera control unit 528.Other peripheral units, for example printer can be provided, and the alarm display device can be provided, for example Shan Shuo lamp, warning horn or similar device.Except those are included in this analysis and preserve the device in the unit 518, can also provide the auxiliary data save set.
In this optional feature 522, can also comprise by this application the assignee sold, trade mark is a point of sale exception monitoring (the Point of Sale ExceptionMonitoting) system of POS/EM.
Be described except the front, in this local node 514 and correlation analysis with preserve the communication link between the unit 518, between this host node 510 and this analysis and preservation unit 518, can also be just like 534 represented DCLs.This data link can be by traditional dedicated line, dial-up connection, satellite, LAN, WAN and/or form by the internet.If used this internet, preferably, these nodes and preservation unit are organized into support " stream " agreement, to carry out the active data transmission.
VR/PC unit summary
Fig. 2 provides this video analysis and has preserved a function summary of module 518.The master unit that will be described in detail below 550 provides 556, one file functions 558 of 554, one hold functions of 552, one analytic functions of a control function and a video processing function 566.
Comprise video input 562 and additional input 564, for example this alarm conditions detection signal that is provided by alarm sensor 526 (Fig. 1) to the input of this unit 550 from this video camera 520 (Fig. 1).
Continuation is with reference to figure 2, and the user control signal that is used for this master unit 550 can be provided by a cursor positioning and feature selecting device 566.Preferably, this device 566 is traditional mouses, those mouses that can buy from Microsoft for example, but this device can be with a tracking ball, touch screen, light pen, or the like substitute.A preferred implementation of this unit 550 also comprises a front panel (not being displayed among Fig. 2) that comprises some by the switch of this user's operation.
Output from this unit 550 comprises the live video data 568 that are provided to a display unit 524 by a video display buffer 570.Another output of this unit 550 is one as 572 represented regeneration vision signals.Although this regeneration video output 572 that shows is independently with this live video output 568, should be appreciated that this video that is reproduced can be sent out by video display buffer 570, to be displayed on this video display unit 524.Comprise control signal 574 and the video data that is reproduced and the index information of following from another output of this unit 550, as 576 represented, to be stored on the outside save set.This class device (not shown) can comprise numeral or analogue magnetic recorder, only writes once or read-write video disk recording device, and/or the DVD register, and these devices are connected to this master unit 550 by industrial siding or by a dial-up line.
Data link 578 provides the data communication between this master unit 550 and other computer installations, and comprises, for example communication channel 516,519 and 534 as shown in Figure 1.Although be not shown in the figure, traditional modem can be included in this VR/PC unit or be connected to this VR/PC unit.
Fig. 3 illustrates the hardware configuration of this master unit 550.This unit 550 has as shown in Figure 3 comprised the hardware and software feature of the uniqueness of the brand-new combination that PC and videograph ability are provided, so can be known as a VR/PC ' (unit of video recorder/PC).Except new video data compression and recording technique, data management, route and analytic function that this VR/PC unit 550 is not provided before carrying out.Unique user interface features has also been realized in this VR/PC unit 550, and it can make this user can select and operate the ability of this unit easily.
This VR/PC unit 550 comprises a mainboard 580,582, one rear panels 584 of front end Video processing and video data compression hardware and a front panel 586 (Fig. 5 and Fig. 6).
As Fig. 4, illustrating of a principle arranged, this front end Video processing and video data compression hardware 582 by two independently printed-wiring board (PWB) form: a simulation process/multiplex plate 588, it receives direct vision signal from this rear panel 584, and an a digitizing/compression/analysis plates 590 that is connected between this analog board 588 and this mainboard 580.
In an alternate embodiments of the present invention, this digitizing/compression/analysis component plate 590 is arranged at be connected between this analog board 588 and this mainboard 580 two independently on the PWB.
Again with reference to figure 3, preferably, the similar of this mainboard 580 is in the standard personal computer mainboard, and is whole by standardization and be commercial available parts.Like this, although new front end electronic unit is arranged, this VR/PC hardware is implemented as a Standard PC platform basically, will be described in detail as following.Parts on this mainboard 558 comprise a microprocessor 592, and its function is identical with a CPU.Preferably, this microprocessor 592 be one that Intel Company produced, be operated in 100 megahertzes and be operated in Pentium P5-120C under the Windows95 operating system.Can use other processors, comprise that those are operated in processor at a high speed.Bus 594 according to the PCI standard is carried out this CPU592 and the miscellaneous part on this mainboard 580 interconnected.As 596 represented, this pci bus 594 is extended to carries out this mainboard 580 and this front end electronic unit 582 interconnected.Miscellaneous part on this mainboard 580 comprises a program memory ROM 598 and a working storage 602.In a preferred implementation, working storage 602 is made up of the RAM of 16 megabyte.
Also being provided on this mainboard 580 is a SVGA chipset 604, and it can be sold by Cirrus Logic company " Alpine " chipset.A SVGA video data input path 606 directly is provided to this SVGA chipset 604 from this front end electronic unit 582.This SVGA chipset provides an output 608 to drive one or more SVGA monitor.(a NTSC output is directly provided driving N TSC monitor from this front end electronic unit.A NTSC monitor occurred if (by conventional apparatus, not being illustrated) detects, this SVGA output can be closed then.)
This mainboard 580 also comprises some serial ports 612, handles the data communication between this mainboard and attachment device.This attachment device may comprise above-mentioned alarm sensor, and alarm shows electronically controlled door lock, conventional P oSEM (point of sale exception monitoring point of sale exceptionmonitoring) device or the like.A mouse mouth 614 is included on this mainboard 580, to receive the control signal that the user excites from this mouse 566 (Fig. 6).Continuation provides a parallel port 616 with reference to figure 3 on this mainboard 580, as a data source that is used to drive a reporting of printer (not being shown).What also be connected with this mainboard 580 is a traditional floppy disk 618, and it preferably is organized into accepts 31/2 inch disk.
Also being provided on this mainboard 580 is an IDE (integrated drive electronic unit) controller 620, and it provides an interface and be installed in this VR/PC unit 550 to a plurality of IDE hard disk drives 622.This hard disk drive 622 provides video data, index information, and program or the like is carried out the ability that high capacity is preserved.The preferred implementation of this VR/PC unit comprises two, three or a plurality of hard disk drive 622.A suitable hard disk drive units that is used for this VR/PC unit 550 is " Caviar " 2 or the 2.5 gigabyte drivers that Western Digtal company is provided.Can also use hard disk drive from Seagate or other suppliers.
A scsi interface 624 also is provided on this mainboard 580.A DAT (digital audiotape) driver 626 links to each other with this mainboard 580 by this scsi interface 624, and forms the main file medium drive assembly that is used for this VR/PC unit.This DAT driver can, for example, be a Sony model SDT-700/BM who on one 4 millimeters * 90 meters tape, preserves 2 gigabyte data.Can also use other known DAT registers.Can imagine, except DAT driver 626, can use other file medium drive assemblies, perhaps substitute DAT driver 626.For example, can use a digital video disks (DVD) device or a linear digital tape drive.
Also had a SCSI delivery outlet 628 by what this scsi interface 624 provided.
Fig. 5 and Fig. 6 illustrate the physical appearance of this VR/PC unit 550.Fig. 5 has illustrated an injection moulding frame 630 with an anterior elevational view 632, and it comprises this front panel 586 and the retaining dust plate 634 that can pivot.Be displayed on Fig. 5 and Fig. 6, location status is in this closed retaining dust plate 634 and can be optionally opened by the user, to allow to take out and put into this floppy disk and the DAT driver that is comprised in this frame 630.
As providing the frame that the hardware that is suitable for frame dress has been installed to being displayed on an alternative structure of this independent rack structure among Fig. 5, can imagining.
Be provided at these switches 636 on this front panel 586, allow the user to control many functions of this VR/PC unit.These switches 636 comprise display format switch 638,640,642 and 644.Represented as the explanatory note on these four switches, this user can use these switches to come can be with 1,4, and 9 or 16 visual stream is presented in the display format on this monitor or a plurality of monitor and selects.Switch 646 allows these users to select to provide the display screen of information of the state of this VR/PC unit of expression, and switch 648 allows these users to select a kind of operator scheme, and this operator scheme allows once to occur a plurality of video flowings but occurs according to a predetermined sequence.
This front panel 586 also installed thereon be designated as " 1 " arrive " 16 ", video camera selector switch 650.When being excited, the vision signal that each video camera selector switch 650 will be produced by this corresponding video video camera at present is presented on this monitor.That relevant with each video camera selector switch 650 is a corresponding LED652, and when the on-site signal from this corresponding video camera was shown, this LED652 was just luminous.A LED652 also is installed on this front panel 586, and when this VR/PC unit was writing down one or more incoming video signal that is produced by this video camera 520, this LED652 was just luminous.
Another LED656 that is installed on this front panel 586 is used for expression to have detected alarm conditions (perhaps passing through this sensor 526, perhaps by performed image analysis in this VR/PC unit).Alarm switch 658 contiguous these alarm indicators 656 and can being excited to impel this system to show information by this user about this alarm that is detected and some alarms.
Another notable feature that is installed on this front panel 586 is one and comes and goes mobile switch 660 slowly.This comes and goes mobile switch 660 in appearance slowly, on operating aspect and the function be provided on traditional VCR Switch seemingly, and be what to be used to control to the playback that is stored in this hard disk drive in this VR/PC unit 550 or the stream of the video image on some drivers.
This round slowly mobile switch 660 is organized into and allows this user to control some well-known playback feature like this, for example forward direction playback, and the back is to playback, and time-out (still image) playback.Preferably, come and go slowly with this that mobile switch 660 is rotated by clockwise or the varying number that is rotated counterclockwise is corresponding, at least two forward direction playback are provided, the back is to the speed of playback.Preferably, after being discharged by this user, this comes and goes mobile switch 660 slowly and returns automatically to " neutrality " position, and the speed of playback or recoil to continue be to the selected speed of the operation of this switch 660 by last.
Integral layout to this front panel is summarized, and from left to right, has defined four zones:
1: two row in zone has the video camera selector switch 650 (8 switches of every row) of associated camera selector marker 652 (also forming two row), and comprises this recording indicator 650;
Zone 2: alarm indicator 656 and selector switch 658;
Zone 3: come and go mobile switch 660 slowly; With
Zone 4: display control switch 638-648.
As will be seen, excite on the screen display at the mouse of the part that forms the graphical user interface of being supported by this VR/PC unit 550 (GUI), provide simulate this front panel 586, substantially similar switch and indicator layout.Figure 11 and 12 has shown the example of this screen display and will further discuss below.
According to another embodiment, in this VR/PC unit, all are presented at the switch on this front panel 586, and LED and other features all can be omitted, so that only excite screen display to control this VR/PC unit by mouse.
This above rear panel 584 by middle zone in, 16 video input ports are arranged, they are aligned to two row, 8 of every row, and represent with label 662.Below this video input port 662 16 loop delivery outlets (they also are aligned to two row, 8 of every row), by label 664 expressions.All input ports 662 and delivery outlet 664 are used for analog video.In the delegation below mouth 662 and 664,666, one printers of a serial ports (walking abreast) mouthful 668, one SVGA (digital video) delivery outlet 670 and a SCSI mouth 672 are arranged from left to right.Adjacent with this input video mouth 662 have additional (simulation) video output 674.
In this part that keeps left above rear panel 584, a multiposition connector socket 676 is arranged, allow this VR/PC unit to be connected to carry out digital communication with other devices.Below this data communication mouth 676 power input socket 678 and a loop power accessory power outlet 680.With this supply socket 678 and 680 adjacent be a ventilating opening 682.On the right side of this rear panel 584 are three expansion slot 684.
Medium driver supporting structure
An effective and compact medium driver supporting structure is installed in the frame 630 of this VR/PC unit 550.Describe this medium driver supporting structure and generally represent this medium driver supporting structure referring now to Fig. 8-10 with reference to the label among the figure 8 700.
The critical piece of this supporting structure 700 is 702, one middle supporters 704 of an infrastructural support part and a top supporter 706.
As what can see from the analysis diagram of this Fig. 9 A and 9B, the xsect of this infrastructural support part 702 is the U-shaped shape basically, and this centre supporter 704 is flat basically, and this top supporter 706 is an anti-U-shaped shape basically.When this supporting structure 700 of assembling, place this centre supporter 704 on this infrastructural support part 702, in turn, place this top supporter 706 on this centre supporter 704.Preferably, bending and punching operation are applied to sheet metal and form all supporters 702,704,706.
This infrastructural support part 702 comprises a flat basically foundation plate part 708.Except an expansion 710 of expanding backward in the plane of this plate 708, from a part of a rear side 712 of this foundation plate 708, this foundation plate 708 is rectangle basically.At the opposite end of this plate 708, vertical sidewall 714 and 716 is formed and upwards expands from this foundation plate 708.Bolt 718 is installed in a zone on these foundation plate 708 right sides.This bolt 718 is to be used for fixing a DAT actuator unit to be mounted the position on this infrastructural support part 702.In a front face side of this foundation plate 708, formed the outshot 720 of a bending.By cutting out section 722, some additional outshots 720 have been formed in the corresponding aft section of this foundation plate 708.(in the view of Fig. 9, a relevant shear part 722 with it in these additional outshots 720 is sealed by this sidewall 716 and riser 726).The shape of this outshot 720 is adjusted in the bridge shape lance that can be inserted on the chassis (not shown) that is formed on this VR/PC unit.By these bridge shape lances and this outshot 720, the whole medium driver of result supporting structure (some drivers wherein have been installed) is safely installed in the frame 630 of this VR/PC unit.
The access port 724 that a projection is arranged in the left-half of the expansion 710 of this foundation plate 708.The effect of this protruding access port 724 be allow to insert a fixator for example bolt make this foundation plate 708 be connected to an aciculiform nut (not shown) on the chassis of this VR/PC unit safely.A riser 726 is fixed on a left-half of this foundation plate 708.A plurality of bolts 728 (wherein Fig. 8 has only shown a bolt) are provided on this riser 726, with the position of a floppy disk unit arranging to be supported by this riser 726.When having assembled this driver supporting structure 700, as Fig. 8 and 10, space between the lower surface of the upper surface of this riser 726 and this centre supporter 704 provides half the corresponding shape factor with a standard drive cell configuration factor, and is suitable for holding the floppy disk of a standard.
Come the 9B with reference to figure again, be formed with 3 grooves above the right side wall 714 of this infrastructural support part 702, they flatly extend to the position near the coboundary of this sidewall 714.The left side wall 716 of this infrastructural support part 702 has a coboundary 734, and from this coboundary, a plurality of short outshots 736 vertically extend upward from the front and back part of this coboundary 734 respectively.Fixator outshot 738 flatly stretches out from the center section of the coboundary 734 of this sidewall 716.A hole 740 is arranged, to allow inserting a fixator by this outshot 738 on this outshot 738.With reference to figure 9B, 8 bolts 742 are arranged on this centre supporter 704 again, these 8 bolts are aligned to two groups, and every group has 4 bolts, and 4 bolts of every group are to be used to fix a corresponding hard disk drive units.
It below this centre supporter 704 pressure pad 744 (representing) with dash area.Be positioned at when this supporting structure under the state of its assembling, as shown in figure 10, and when wherein also actuator unit being installed, this pressure pad 744 applies a downward pressure to a DAT actuator unit 746 and a floppy disk unit 748 respectively, so that these actuator units are remained in this driver supporting structure 700.Preferably, this pressure pad 744 is resilient elastomeric material.
The fixator outshot 749 that is stamped out extends upward from the position at 4 angles of this centre supporter 704.Hole on this outshot 749 allows to insert a plurality of bolts, and the hard disk drive units that will be installed on this supporter 704 is fixed on this centre supporter 704.A U-shaped carriage 751 is to be used for process to provide connection at a fixator 753 of the core of this centre supporter 704.This carriage 751 can help this hard disk drive units mounted thereto is fixed on this centre supporter 704.
Short outshot 750 flatly stretches out from a right hand edge 752 of this centre supporter 704.Two pivot outshots 754 from this edge 752 outwards and be bent upwards and lay respectively at the anterior locations and the back location at this edge 752.Position along the edge of this supporter 704 between this outshot 754 and 750 is arranged in order to be able to insert simultaneously by the groove 730 on the sidewall 714 of this infrastructural support part 702.When being inserted into this pivot outshot 754 in the external slot 730, when mounted, can be with this centre supporter 704 downwards towards these infrastructural support part 702 rotations.In the left side of this centre supporter 704, groove 756 is arranged, the position between them is arranged in order to be able to allow the short outshot 736 on the sidewall 716 of this infrastructural support part 702 is inserted among these grooves.The flap 758 and a fixator outshot 760 (similar with the fixator outshot 738 of this supporter 702) of a plurality of downward extensions are arranged on the left side wall of this supporter 704, and a hole 762 is arranged on it.
This supporter 704 also further has two pairs of grooves 764, and the distance between the respective edges of the every pair of groove 764 and this supporter 704 is short.On the front edge of this supporter 704 and edge, back, a downward short flap 766 is arranged.
With reference to figure 9A, similar with this centre supporter 704,8 bolts 742 are arranged on this top supporter 706, these 8 bolts are aligned to two groups, and every group has 4 bolts, and every group bolt is the position that is used for fixing a corresponding hard disk drive units.These bolts 742 be fixed on form most of this top supporter 706, on flat basically top board 768.Shown in Fig. 9 A, pressure pad 744 has been installed on the downside of this top board 768, the effect of this pressure pad 744 is to apply downward pressure to the hard disk drive of being installed on this centre supporter 704.
What extend downwards from the corresponding left hand edge of this top board 768 and right hand edge is sidewall 770 and 772.Some short outshots 774 vertically extend downwards from the lower limb of this sidewall 770 and 772.The position of these outshots 774 is aligned to those grooves 764 that can be inserted this centre supporter 704 simultaneously.(in the view of Fig. 9 A, an outshot in these outshots 774 on top supporter 706 is sealed by the top board that is provided 768).Fixator outshot 776 flatly stretches out from the center section of the lower limb of sidewall 772.This fixator outshot 776 is similar with said fixing device outshot 738 and 760, and a hole 778 is arranged thereon.This supporter 702,704 and 706 is general, this respective fixation device outshot 738 especially, 760 and 778 size is arranged to like this, so that work as by assembling this 3 supporters 702 as shown in Figure 8,704, with 706 o'clock, these corresponding holes 740,762 and 778 are assembled into and are in vertical position arranged side by side each other, therefore allow a single fixator 779 to fix this three supporters 702,704 and 706 by these three holes and this single fixator of permission simultaneously, to form this driver supporting structure 700.
Figure 10 has shown this medium driver supporting structure 700 that has assembled fully.Except aforementioned DAT driver 746 and floppy disk 748, hard disk drive units 786 and 788 is shown as being installed in side by side on this centre supporter 704, and a single hard disk drive units 790 is installed on the top supporter 706.Can see the keep right part on limit of the top board 768 that a space that is used for installing one the 4th hard disk drive units (not showing) is retained in this top supporter 706.
According to following step, installation of driver unit and this supporting structure can become fast and be effective: 1, DAT driver and floppy disk are installed on the infrastructural support part; 2, use the hinge outshot that middle supporter is installed to the infrastructural support part; 3, two hard disk drive units are installed on the middle supporter; 4, use the straight outshot that extends from sidewall downwards, the top supporter is installed on the middle supporter; 5, insert fixator (represented) and make it pass through all three holes on respective fixation device outshot by the label among Figure 10 779, make these three supporters 702,704 and 706 form a single rock-steady structure; 6, use carriage 782 and the fixator by outshot 780, one or two hard disk drives are installed on the supporter of top.This medium activation configuration that has installed just can be installed on the chassis of this VR/PC unit then.If only two hard disk drives need be provided, then can omit step 6 in this unit.
Should be appreciated that this medium driver supporting structure shown in Fig. 8-10 allows the high capacity preservation hardware components of this VR/PC unit can use a kind of like this convenience and price and all excellent mode of time to install.
Analog video hardware
With reference now to Figure 13 A and 13B, the analog front circuit structure (Fig. 4) that is provided on the plate 588 is described.
As can be seen, flow the multiplexer that is provided to thirty six to together by 16 that are connected to this VR/PC unit 16 analog video signals that video camera produced from Figure 13 A and 13B, this multiplexer of ten six to one is represented with label 802.This multiplexer 802 provides three outputs, and they are formed into the input of three field locking channels 804 respectively.Control each multiplexer 802 by a corresponding control signal (this control signal is represented by 806), select corresponding in the video camera 16, to carry out data acquisition by this corresponding lock channel 804.This control signal 806 is provided by this digital front-end plate 590 (Fig. 4).
Continue with reference to figure 13A and 13B, these three lock channels 804 are identical, so that only describe in these three channels one.Selecteed incoming video signal is provided to a colourity notch filter 812 and a chroma strip bandpass filter 814 respectively by amplifier 808 and 810.A luminance signal is exported from this colourity notch filter 812, and is provided to a synchronizing signal separator circuit 816, composite synchronizing signal of extracting from this luminance signal of these synchronizing signal separator circuit 816 outputs.This composite synchronizing signal from this sync separator 816 is provided to a vertical synchronizing signal separation circuit 818, and this vertical synchronizing signal separation circuit 818 is separated a vertical synchronizing signal from this composite synchronizing signal.This vertical synchronizing signal and this composite synchronizing signal all are provided to this digital front-end plate 590.This composite synchronizing signal from this sync separator 816 also is provided to a pulse gate testing circuit 820, pulse gate detection signal of these pulse gate testing circuit 820 outputs.This pulse gate detection signal and be used as input from the carrier chrominance signal of this chroma strip bandpass filter 814 outputs and be provided to a phaselocked loop (PLL) circuit 822.This PLL822 base band carrier chrominance signal of output and a reference signal.Another multiplexer module 824 is between this field locking channel 804 and two selective channels 826.This multiplexer module 820 is made up of 6 three to one multiplexer, wherein 3 multiplexers be used for two selective channels each.The control signal that is used for this multiplexer module 824 is represented by 828, and is provided from this digital front-end plate.
These two selective channels 826 are identical, so only describe a channel in two channels.Three inputs to each selective channel are luminance signals, a carrier chrominance signal and a reference signal, all these are corresponding with the vision signal that single camera provided that is selected for locking by one in three input camera signals, and this single camera is selected by one in three lock channels 804.From the output of corresponding three to one multiplexers of this module 824 is to be used for luminance signal, in carrier chrominance signal and the reference signal each, so that correspondingly, three inputs of this corresponding multiplexer are three brightness outputs, colourity output or reference output, depend upon circumstances, may be from this lock channel 804.This selecteed luminance signal is provided to a brightness and clamps down on circuit 830, and this brightness is clamped down on circuit 830 and is luminance signal that quilt is clamped down on of selected channel one output.This selecteed reference and carrier chrominance signal, they are corresponding with this selecteed luminance signal, are provided to a chrominance demodulation circuit 832 to a multiplexer 834 output R-Y and B-Y signal.Shown in 836 (control signal is provided by the front end digiboard), this multiplexer 834 is controlled to provides a R-Y who replaces and B-Y signal sequence, with as the carrier chrominance signal that is selecteed channel one.This continuous alternative colors difference signal of the luminance signal that this quilt is clamped down on and this carrier chrominance signal of composition is output to this digital front-end plate 590 then, to be further processed.
As described above, with channel two corresponding selected channels 826 and identical with this channel one corresponding selected channel.
The front end digital hardware
The form of Figure 14 functions of use module provides a summary of this digital front-end plate 590.Main functional modules on this front end-plate 590 comprises an analog-digital conversion and buffer module 840, control and compression processing module 842 for one, 844, one live video image analysiss of live video display process module module 846 and one " rear end " compression module 848.This front end digiboard 590 also comprises the interface 850 (Fig. 3) to pci bus expansion 596.
Continuation is with reference to Figure 14, this module 840 receive two by and by these AFE (analog front end) 588 selected analog video channels, and selecteed two analog channels are carried out digitizing, and with the video data field that is digitized be buffered in respectively with these two selecteed channel corresponding buffers 853 and 855 in.Will be output to the control signal of this AFE (analog front end) and represent the signal of the state of this AFE (analog front end), comprise synchronizing signal, be passed through the reception and the transmission of module 840.In addition, this module 840 is controlled a video data bus 852 and according to a form that will be described below, this video data field that is cushioned is assigned to module 842,844,846 and 848.The control module 842 of a control/status bus 854 and this digital front-end plate 590 and other modules are carried out interconnected, and allow this control module 842 to control the signal of the state of these other modules and these other modules of reception expression.Control that finally is sent to this AFE (analog front end) or is sent out from this AFE (analog front end) and status signal are also by in these control/status bus 854 transmission.
Except the integral body control of function that this front end-plate is provided, this module 842 is also carried out the primary data compression to the video data of output on video bus 852 and is handled.This vision signal of 844 pairs of deliveries on video bus 852 of module is carried out display process, and be provided to this mainboard 580 and be provided in the output signal of this display monitor, exporting a processed vision signal that comprises that coverage information and image plane distribute.According to the technology that describes below, this video data of 846 pairs of deliveries on video bus 852 of this module is carried out the motion video analysis, allows the feature of this VR/PC unit detection by the represented image of this inputting video data.
Preferably, this module 848 use a standard, the commercial integrated circuit that can obtain realizes that it can carry out data compression process to being carried out pretreated this video data by module 842.In a preferred embodiment of the present invention, handled consistently with well-known Joint Photographic Experts Group by 848 these the performed compressions of this module, and commerce can realize from the IC MODEL C L560 that C3, Inc. has bought.According to this embodiment, only used the code capacity of this JPEGIC, rather than its decoding capability.
This pci interface 850 is used to through direct memory visit (DMA) technology, under the control of module 842, provides this input, compressed encoded video signal to this mainboard 580.Also shifted by control signal that is received from this mainboard 580 and the status signal that is sent to from this mainboard 580 of this module 842 through this pci interface 850 from this module 842.
Digitizing and buffering video data
At first with reference to Figure 15, describe this digitizing and the buffer module 840 of Figure 14 in detail now.In Figure 15, shown to principle the main functional parts of this module 840, comprise analog-digital conversion (856), video digitizer control (858), field buffer (860), video output control (862) and control register visit (864).Provide this control register access function 864 to allow this control module 842 (Figure 14) to write control messages, and read inputting video data and about the status message of this module 840 and this AFE (analog front end) plate about this module 840 and this AFE (analog front end) plate.With reference now to accompanying drawing discussion subsequently, is displayed on other parts of this module 840 among Figure 15.
The details of this analog-digital conversion function 856 is displayed among Figure 16.Four discrete translated channel 866-874 are provided.Channel 866 and 868 is respectively what to be used in this luminance signal of channel one and two, and channel 870 and 872 is respectively what to be used in this carrier chrominance signal of channel one and two.In these four translated channels each comprises an impact damper amplifier 174 and simulation-to-digital conversion circuit 876.A datum shown in 878 and shown in 880, each channel is controlled according to an analog/digital conversion clock (Dot Clock).
As shown in figure 17, be digitized vision signal from this of this A/D translation function 856 outputs and be provided at that (channel one comprises brightness one and colourity one two channels, channel two comprises brightness two and colourity two), and be provided to this video digitizer controller 858.Carry out this digitizing, so that one 8 bit codewords is represented each point.This point data is provided to the steering logic 882 that execution will be got in touch the processing of Figure 17 A description.For each channel provides a FIFO storer 884, allow when vision signal will handling according to this PAL standard, to carry out required timing adjustment.
This controller logic 882 to each independently channel carry out this processing be displayed among Figure 17 A.To a specific channel, when this controller logic is waited until the beginning that detects a video field signal in this particular channel always (module 886), wait for a predetermined period of time (removing this vertical blank at interval) then, when waiting until the beginning that detects delegation then (module 888).When detecting the beginning of this row, this first point value is loaded into and corresponding this field buffer of this particular channel, and this logic is carried out a circulation of being made up of module 892 and 894 then, is written into the value of all points subsequently in this row thus, up to the end that detects this row.When detecting the end of this row, withdraw from this circulation and forward module 896 to, at module 896 places, judge whether this is this last column.If not, this cycle of treatment is returned to module 888, otherwise this cycle of treatment is returned to module 886.
Advantageously, this processing of Figure 17 A can realize with a state machine that forms the firmware in a programmable logical unit.The design of this firmware is easily to those persons skilled in art, further describes with regard to not needing here.
Refer again to Figure 17, this control logic module 882 is exported the point of the video data of this first and second channel according to this processing of Figure 17 A, and provides a control signal to following field buffer module, shown in 898.
Figure 18 has shown the details of this field buffer module 860.Except aforementioned field impact damper 853 and 855 (each uses a VRAM to realize), this field buffer module 860 also comprises a VRAM controller 902.This VRAM controller 902 control this impact damper VRAM853 and 855 and controlled by signal 898 (from this video digitizer controller 858) and signal 904 (from video output controller 862) again conversely.This video data of being exported from this video digitizer controller 858 is stored in this field buffer 853 and 855, and is read out from this field buffer 853 and 855 through a bus 906. Impact damper 853 and 855 address and enable signal by delivery on an address bus 908 of being controlled by this VRAM controller 902.
As shown in figure 19, the core of this video output control module 862 is output control logics 910 of realizing as an output state machine (Figure 35 will describe).This video data that this output control logic 910 receives from this VRAM853 and 855.This VRAM controller 902 produces this control signal 908 that is used to control this VRAM and produces required address.These of output control logic 910 controls postpone FIFO912, and a head module 912B and a FIFO add up 914.Data from these modules are buffered on this video bus by bus driver 912A.This FIFO914 adds up to will be controlled/compress front-end module 842 employed video datas by this, to carry out video data compression.Through this control register visit 864 (Figure 15), these data are sent to this module 842.
Video data format
Describe the each side of this form now, wherein this output control logic 910 impels this video data 852 to be sent on this video bus.
Each shows an image plane 920 Figure 20 A and 20B, the purpose of representing for internal data wherein, and this image plane 920 is divided into 240 row of vertical direction, and each row is divided into 640 points in the horizontal direction.The actual video data that is used to drive this display monitor is formed 480 row and multiply by 640 points, wherein this 240 data line that is provided for this internal data each in representing is carried out vertical interpolation, produces this additional row.The form that this image plane also is used the data tile shows that each data tile is measured each 8 point (Figure 21) on horizontal direction and the vertical direction.Like this, this image plane is divided into 80 data tiles in the horizontal direction, is divided into 30 data tiles (Figure 20 B) in vertical direction.
Use 4: 1: 1 data layout, represent one group of four point by the brightness data of four bytes and the chroma data of two bytes thus.In fact, each row is divided into the discrete groups of four points, and for each such group provides four luminance point bytes, and the V color data (Figure 23) of the U color data of a some byte and a some byte.This form and traditional 4: 1: 1 forms form contrast, and wherein each chroma data byte and this image plane two point multiply by the regional corresponding of two points.Here employed and in Figure 23 illustrated in this form of explanation can help to make the change dirt of this colouring information on this vertical direction to minimize, and make result from only 240 horizontal line are distributed to each video data field, any negatively influencing of image quality is obtained reducing.
In order to carry out compressed encoding, this image plane is divided into four horizontal 8 discrete groups (Figure 22) of taking advantage of 8 data tiles (tile).The continuous data tile of four levels of each group is formed one " minimum code unit " (MCU).Need these data of each MCU of expression to multiply by 8 brightness data point pieces by four 8, one 8 multiply by 8U data point piece and one 8 and multiply by 8V data point piece composition.As shown in figure 24, a preferred sequence that is used for being sent in the data of each MCU is the U data block, V data block and four brightness data pieces then.
According to employed this new video data layout in this VR/PC unit, each video data field of being exported from this module 840 by this video output controller 862 is sent out twice, once use the form of data tile, once use the capable demonstration of raster scanning.This data tile is intersected with this scan line, as shown in figure 25.In a preferred form, 15 8 multiply by 8 data tiles and were sent out this video data field that will be sent out of this piece field Data Identification a field data in the past.Then, this head is sent once more, is thereafter that second raster scanning is capable, and be another 15 data tiles then and be that the 3rd raster scanning is capable then, or the like.Intersect to send raster scanning this process capable and rectangle data tile group and continue, be sent out and all raster scannings are capable is sent out up to all data tiles.As a result, as mentioned above, this each point data code word is sent out twice, once is the part as a rectangle data tile, once is as a part that raster scanning is capable.According to sequential as shown in figure 25, it is capable and need rough 40 milliseconds one-period send 15 data tiles of each group to need a rough cycle of 27 milliseconds to send each raster scanning.
Nominally Figure 25 represents 3/4ths corresponding sending modes with this NTSC standard of the transmission rate of 60 of per seconds.In pattern as shown in figure 25, per second sends 45 row, sends identical 45 data tile with per second.Should be noted that every has sent 240 row, every has sent 3600 data tiles (2400 brightness data data tiles add 600 U data data tiles and 600 V data data tiles).Like this, when as shown in figure 25 operate this video bus 852 time, the input rate of this system is 45 of per seconds, optionally in the input with this 45 field distribution to 16 video camera.With reference to Figure 14, according to the preferred format of operational module 842 and 846, this data tile is provided on this video bus 852 tout court; Concerning on-the-spot display process module 844, the capable transmission of this raster scanning is preferred input format thus.
Forward to now with reference to Figure 26, will describe another pattern of this video bus of operation.In this pattern, use the speed of 60 of per seconds to send data tile (tile), but in these 60, only use capable form to send every one.In other words, half be sent out twice, once be sent out as row, once be sent out, and remaining field only is used as the data tile and sends as the data tile.Can use this pattern; For example, when one or more field locking channel 804 and one or two selective channels 826 are used for single camera input uniquely.Under a kind of like this situation, this AFE (analog front end) electronic installation can lock onto this input channel, and between the field without any delay, to allow the throughput of 60 of per seconds.
In the operator scheme as Figure 26, in one 27 milliseconds cycle, it is capable to send one first grating, sends 30 data tiles then in one 80 milliseconds cycle, and it is capable to send next grating then, sends 30 data tiles of next group then, or the like.(omitted this head among Figure 26, come reduced graph).Sent with one corresponding 240 the row during with interior (that is, rough 1/30th seconds), also sent and two corresponding 7200 data tiles in field.
Figure 27 illustrates another pattern of operating this video bus 852.When showing this field that is being sent out in the specific part that will use scalable manner, use this pattern as shown in figure 27 at this image plane.In this pattern, only sending needs the grating of this image that is exaggerated of generation capable, and only transmission is arranged in the capable point of those visual required gratings that this is exaggerated.This has reduced the bandwidth that need be kept in this scene display frame buffer.
In example as shown in figure 27, suppose that the convergent-divergent of realizing a twice shows.If Figure 27 and Figure 25 are compared, in the pattern of Figure 27, can observe, in the alternative time slot of the capable sending time slots of grating, in transmission, omitted all data.In other time slots, only send the least bit in each row.But, in order to keep the correct timing of this bus,, perhaps only send the raster data of lesser amt although do not send any raster data, 27 milliseconds the capable time slot that provides between every pair of data tile time slot still is provided.(among Figure 27, having omitted this head again) with reduced graph
Figure 28 illustrates employed this data layout when sending this grating line data on this video bus 852.In a preferred implementation of this system, this video bus 852 comprises 19 parallel signal lines, and in these 19 parallel signal lines 16 are that to be used for (two bytes arranged side by side) of data and three remaining bits be that be used to identify will be by the byte number of concurrent transmission.In example shown in Figure 28, suppose that this row that will be sent out is this first row in, so this first nybble (one or two byte is used for sending time slots) is used for a field 926.In 926, at first two bytes of this that is sent out are formed one 16 bit timing code at this.The information of the type of the field that following this video camera number of two byte representations and other indicate to be sent out.In this camera number serve code word joint, the one or four bit is this video camera number, and last four bits are that this that represent to be sent out will be displayed on the part (" window " number) of this image plane wherein.This window number may be represented, for example, uses this third line of one 4 * 4 windows display form, and this that is being sent out is presented in this second window.In this scene display process module 844 (Figure 14), this window number helps to form effectively windows display.
Byte 930 comprises a type and other information.In this form, do not use this first to the 4th bit and the 8th bit in the byte 930.The 5th bit is represented what whether this input video was caught by a single camera pattern (that is, only catch from a video camera video information).The 6th bit represents that this is even field or odd field, and the 7th bit represents whether to have caught the frame of the field of this vision signal.
Following two bytes, represented by label 940, formed the wardrobe of one 16 bit-rows identification number.Comprising the first nybble brightness data after the capable point data of grating, is the U color data of two bytes then, is the brightness data of another nybble then, is the V color data of two bytes then, or the like.When having sent 640 byte brightness datas and 160 byte U data and 160 byte V data, just finished the transmission of this row.Follow in the end a pair of byte later be the identification number at end of this row of expression, shown in label 942.If this row that is being sent out is the last column in this, a pair of byte packet is drawn together " next video camera " identification byte 944 that the identical data form is arranged with " video camera number " described above byte 928 down then.Should remind this scene display process module 844 in advance by " next video camera " identification byte 944, allow to produce in advance coverage information, suitable setting of buffer pointer or the like.
An alternate formats as shown in figure 28 form, in form shown in Figure 28, the color data byte of two same types is sent together, alternatively, can send each like this to the color byte, wherein each comprises and four corresponding U information bytes of last luminance point and a V information byte byte.
The form that will use its these data tile data of transmission on this video bus is described referring now to Figure 29 and 30.Shown in 952, one or two byte formed this data tile header.A byte in this shown in 954, comprises one the seven bit column identification number that uses row to represent this position of data tile in this image plane.Last bit of this byte 954 represents whether to find that this data tile is " reformed " data tile of this data compression process that will describe below being used for.Another byte in this data tile header shown in 956, comprises that 6 bits represent the line position of this data tile in this image plane.Latter two bit is retained respectively as expression two different image analysis algorithms, whether this data tile is changed (promptly, two different " change threshold values " can be used for image analysis, and these two can be different with this threshold value that is changed data tile bit that is used for byte 954, and the latter is used for data compression process).
After a byte 952 be and corresponding 64 bytes of this 8 * 8 data tile.In example shown in Figure 29, suppose that this data tile is a U color information data tile.Another data tile is made up of 64 brightness bytes or 64 V colouring information bytes.Each minimum code unit (seeing Figure 22) to continuous 8 * 8 zones of four levels of this image plane has sent four brightness data tiles, a U data tile and a V data tile (Figure 24).(80 data tiles are arranged in the horizontal direction, 30 data tiles are arranged in vertical direction because this image plane is divided into a zone, 2400 8 point * 8; See Figure 20 B), 3600 data tiles of sum comprise colouring information, are used to indicate each.
First front in this data tile field is for each data tile field sends a field head.Except employed this video camera number in data tile form and a field type byte and employed those bytes in row format some different, the row that this data tile field head is similar to Figure 28 sends this head shown in the form, the timing code that comprises two bytes, " video camera number " byte and " type " byte.With reference now to Figure 30,, with this a video camera number and a type format of data of description tile place use.This data tile camera number serve code word joint 928 ' at first comprises in the nybble and employed identical, the 16 bit video camera identification codes of this row format at it.But, because this data tile is not used in on-the-spot the demonstration, just do not need window ID information, thus do not use this camera number serve code word joint 928, other four bits.
For employed this type byte 930 ' in the head of this data tile field, identical in last 6 bits and this row format field head.Its first bit represents whether this data tile field is used as a reference diagram image field that is used to carry out image analysis, and its second bit represents whether this data tile field is used as a reference picture that is used to carry out data compression process.
Figure 31 provides a summary by this performed processing capacity of this output control logic 910 of Figure 19.As can be seen from Figure 31, this output control logic 910 comprises that 960, one of control state machine are used to set up the function 962 of this head, a translation logic 964, grating is set up logic 966, and the data tile is set up logic 968 and grating/data tile sequencer 970.The function of this output control logic is contact described this data layout of Figure 25-30 and a bus operation mode above producing.The form of use flow graph illustrates this head and sets up function 964 in Figure 31 A.Shown in the module among Figure 31 A 972, this head is set up function and is in idle condition, up to the beginning that has detected a field (first point in first row).As long as this situation taken place, this head is set up function and is just obtained video camera from this front end-plate controller 842 and identify and time stamp information (Figure 14), shown in module 974, and this form assembled (module 976) then.
Shown in Figure 31 B, this translation logic decision skew and flexible parameter (step 978) and use these parameters to drive that this grating is set up logic 966 and this data tile is set up logic 968 (step 980).
Figure 32 illustrates by this grating and sets up this processing that logic 966 is carried out.At first (step 982), this grating are set up this current grating of logical foundation and number are obtained row number.Judge then whether this is first row (step 984) of a field.If not, these wardrobe (as Figure 28 940 shown in) assembled (step 986), form these capable data of this grating then and be sent to this sequencer 970 (step 988), this cycle of treatment is got back to step 982 then.But, if find to handle first row, the inserting step 990 before step 986 so in this in step 984.In step 990, obtain the information of this a video camera number and a byte and assemble this head.
In Figure 33, shown by this data tile and set up this performed processing of logic 968.This shown processing is represented as a circulation continuously in Figure 33, comprises a first step 992, and wherein this data tile is set up logic 966 and obtained data tile identification data, promptly is used for the row and column of the data tile of this image plane.Then, in step 994, these data be assembled in this data tile header byte (as Figure 29 952 shown in).
Continuation is step 996 with reference to Figure 33 after the step 994.In step 996, this data tile is set up logic and calculate suitable VRAM address from this row and column information, and continues this one suitable (Figure 18) from VRAM853 and 855, and retrieval needs the data of this data tile (step 998) of structure.Be step 1002 after step 998, wherein this data tile is set up logic 968 and is forwarded this sequencer 970 to before with this data tile that is done.
Figure 34 illustrates this processing by this grating/data tile sequencer 970 is carried out.
Beginning in step 1004, judges whether it is to send the time that grating is capable.If like this, this sequencer 970 sends and capable corresponding these data (step 1006) of this grating set up these wardrobe that logic 996 produced by this grating after, if this is this first row, the front also has this head.
If in step 1004, do not find to send the time of this raster data, just forward step 1006 then to.In step 1006, judge this sequencer 970 whether from this data tile set up logic 968 received between two gratings are capable during with the complete data tile of interior a group of will be sent out.The number of this data tile or 15 or 30, whether this is effectively relevant respectively with pattern or the pattern of Figure 26 among Figure 25.If in step 1008, a complete data tile group has appearred in discovery, this data tile data so comprise that setting up the data tile header that logic 968 produced by this data tile (with a field head, is sent on this video bus 852 (step 1010) if desired).Otherwise step 1008 is a step 1012 later.In step 1012, to set up from this data tile each time and received a data tile the logic 968, this data tile set of counters just increases by 1, and this cycle of treatment is got back to step 1008.
After step 1006 or step 1010, this cycle of treatment is got back to step 1004.
Figure 35 illustrates this control state machine 960 (Figure 31).As can be seen from Figure 35, when field of beginning, exit from idle status 1014, enter data tile treatment state.When having finished the transmission of 15 data tiles, (or finish the transmission of 30 data tiles, this pattern with this video bus of operation is relevant), exit status 1016 then, get the hang of 1018, in state 1018, send and the capable corresponding data of grating.When having finished this row, rotate back into this data tile treatment state 1016 from this row treatment state 1018.But, when having finished last column, just finished this field, in this case, this state machine jumps to this idle condition 1014 from this row treatment state 1018.
The front end-plate control hardware
Forward Figure 36 now to, describe this control module 842 that is used for this front-end electronics now in detail.The critical piece of this control module 842 comprises 1052, one static RAM (SRAM)s 1054 of 1050, one dynamic rams of a digital signal processing integrated circuit and a DMA addressed module 1056.This DSP1050 (similar with this digital signal processing device that monitors this scene display process module and this scene photo picture analysis module 846 respectively) can be a series of TMS-C32 devices that Texas instrument company is produced.This DSP 1050 is this front-end electronics " brain " and " traffic-police ".Except other function, this DSP1050 manages the management that is used to video data and an address bus 1058 and data bus 1060 of data compression process, and this address bus 1058 and this data bus 1060 also are used for through pci interface 850 compressed video data being transferred to mainboard.
This DSP1050 also manages this control/status bus 854 (Figure 14; Be not presented among Figure 36).This state of a control bus 854 may be implemented as a high speed serialization link, this high speed serialization link sends to this digitizing with the order of this DSP1050, buffering and bus control module 840, this scene display process module 844 and this scene photo picture analysis module 846.Also delivered from status message module 840,844 and 846, that will send to this DSP1050 by this control/status bus 854.By the message (Figure 14) through 840 relayings of this module, this DSP1050 can also handle the control of this front end analog board 588 and supervision (Fig. 4 and 13).
Again with reference to Figure 36, this DRAM1052 preserved by that this DSP1050 produces and handled with initial compression statistics of using and video data data tile, and this DSP1050 visits this video data through this FIFO914 and this control register visit 864 (Figure 15) of Figure 19.This SRAM1054 is used as the common working storage of this DSP1050, and as an output buffer that is ready to be sent to through this pci interface 850 compressed video data of this mainboard.This DMA addressed module 1056 is provided at this compressed video data transferred to the DMA of this mainboard from this digital front-end plate 590 and shifted during employed address.
The tri-state bus driver represented by label 1062 is relevant with bus 1058 and 1060, and is used for the signal service stream of route on these buses as required.
Scene photo picture analysis hardware
With reference now to Figure 37, the details of this scene photo picture analysis module 846 is described.An on-the site analysis module 846 comprises a digital signal processing integrated circuit 1064, and this digital signal processing integrated circuit can be this class device that above-mentioned Texas instrument company is produced.What be included in this DSP1064 inside is to comprise that 1066, one of arithmetic and logic units start 1068, one program buffer storages of sign indicating number (bootcode) storer 1070 and latch functional module with control circuit structure 1072.Relevant with this DSP1064 is a serial ports 1074, program storage 1076 and video memory 1078.This serial ports 1074 receives control signal and this control signal is relayed to this from this front controller DSP1050 and latchs and control circuit structure 1072.Similarly, status message is latched from this with control circuit structure 1072 and is relayed to this DSP1050 through this serial ports 1074.Address and data bus 1080 interact with this DSP1064 and storer 1076 and 1078.
This program storage 1076 is preserved the software that this DSP1064 of control carries out one or more image analysis algorithm.This image analysis algorithm software is loaded in this program storage 1076 by this control DSP1050.Conversely, this algorithm software may be forwarded to this control DSP1050 before this mainboard.The starting resource of some or all image analysis algorithm software may separate with this VR/PC unit and away from a device of this VR/PC unit.
This video memory 1078 is preserved this video information that will be analyzed by this scene photo picture analysis module 846.The form that this video information is used the data tile receives from this video bus 852, then in being saved to this video memory 1078 before, formatted in form logical one 082.
Preferably, this video memory 1078 has enough preservation capacity to preserve the reference picture of a plurality of independently video streams, the perhaps statistics of deriving from reference picture.For example, can imagine that this scene photo picture analysis module 846 can be simultaneously to carrying out image analysis by 16 16 on-the-spot input video streams that video camera produced that link to each other with this VR/PC unit respectively.In addition, the parameter that is applied to execution in this corresponding image analysis algorithm of each input video stream, that be used to retrain this algorithm can change, and one or more be different from fully other simultaneously algorithm of applied vision algorithms can change at the difference of image feature.For example, a circumference is violated detection algorithm and can be applied to some input imagery streams, and " museum " algorithm can be applied to one or more other input video streams.Will be further described below this circumference violates detection algorithm and is somebody's turn to do " museum " algorithm.
Should be appreciated that the result who is handled by these module 846 these performed image analysiss is reported to this controller DSP1050 through this serial ports 1074.
On-the-spot display process hardware
With reference now to Figure 38, the details of this scene display process module 844 is described.
The critical piece of this scene display process module 844 is DSP1084, one first flexible and color space conversion circuit 1086, one shows VRAM1090, an overlay planes produces circuit 1092, covering mixer and serializer 1094, one advanced persons/go out earlier storer 1096, one second flexible and color space conversion circuit 1098,1102, one NTSC/PAL coding circuits 1104 of a SVGA feature connector and a synchronous generator 1106.
Preferably, use the commercial components of standard to realize that all are presented at the circuit module among Figure 38.For example, this scene display controller DSP1084 device that above-mentioned Texas instrument company is produced preferably.This DSP1084 receives command messages from this controller DSP1050, and status message is sent to this controller 1050.Under the supervision of this controller DSP1050, this scene shows that control DSP1084 controls the operation of this scene display process circuit structure 844, and, especially, control two flexible/color space conversion circuit 1086,1098 and these overlay planes generators 1092.
This first flexible/color space conversion circuit 1086 receives and handles this grating line data that is provided on this video bus 852.If occupied the entire field video display part of this display screen by this represented image of the video data lines of this reception, circuit 1086 is not carried out any flexible processing then.But, if show the on-the-spot image of a separator screen, comprise and the corresponding image of two or more live video image streams, in circuit 1086, carry out flexible the processing then.For example, if use 2 * 2 form that four images are presented in the respective window, in circuit 1086, each image all dwindles one times in the horizontal direction with on the vertical direction then.In addition, in circuit 1086, also finish the conversion of color space, so that be switched to the RGB data from this YUV color data that this video bus received.
From these flexible/color space conversion circuit 1086 outputs, this is converted (and if desired, by flexible) of being provided to as an impact damper of RGB data shows VRAM1090, this video data that is cushioned is output to this covering mixer/serializer 1094 then.Simultaneously, the control signal from this DSP1084 is responded, this overlay planes generator 1092 provides one to cover image, for example may represent to comprise the alphabetic character of the topic head at " scene " or " video camera one ".This pictorial data of representing this covering image is provided to this covering mixer 1094 from this overlay planes generator 1092, to mix with this live video pictorial data.After in the serializer part of this circuit 1094, being carried out suitable buffering, comprise, if desired, be assigned to a selecteed display window, this video information (of may be several video image streams is compound) is passed through FIFO storer 1096 and is transferred to this second flexible/color space conversion circuit 1098.
If the part occupied from this image of this front-end electronics output is little than the whole image plane of this display monitor, at circuit 1098 just to passing through the processing of stretching of this image that this FIFO storer 1096 exports.If the entire display screen curtain on this monitor is used to show this scene image of exporting from this front-end electronics, in circuit 1098, just do not carry out flexible the processing then.But, if this live video only occupies a part (as shown in the figure, for example, as Figure 11) of this video screen, just carry out flexible the processing then, so that the size of this image that this front-end electronics produced is adjusted to this window that is suitable for being assigned on it.In the demonstration of Figure 11, can observe, one of this screen display by below part in, the graphical user interface feature is provided.As will be explained below, produce this GUI element at mainboard.
Rgb video data (if necessary, being carried out flexible processing in circuit 1098) through the SVGA feature connector 1102 (Figure 38) of standard, are used as from this SVGA of this digital front-end plate 590 (Fig. 4) output and send to this mainboard.In addition, 1098 pairs of these RGB data of this circuit are carried out a color space conversion (if desired, after the processing of stretching), use the form of yuv video data that one second output is provided.These yuv video data are provided to a NTSC or PAL scrambler 1104, this scrambler 1104 uses a synchronizing signal that is produced in synchronizing generator 1106 to form a NTSC that may be used to drive a NTSC monitor (perhaps PAL, if situation needs) analog output signal.
Front-end software
Figure 39 represents to control the summary of this software of the operation of this video board controller DSP1050.This software of controlling this DSP1050 comprises a video board main control software module 1108 that other software modules of this DSP1050 are arbitrated.Other software modules comprise analog board management (video camera ordering) module 1110, a task manager 1112, a time tracking module 1114, an on-the site analysis module management person 1116, an on-the-spot display module supvr 1118, Compression manager person's module 1120, one data tiles comparison module 1122, one mapping generators module 1124 and a Drive Module 1126 that is used for the mainboard pci interface.
This software module 1110 through 1118 and 1126 can be counted as handling " management " responsibility of this DSP1050, and module 1120-1124 relates to the function relevant with this compression of video data.In this " management " software module, except pci interface driver 1126, all software modules are to carry out customary function in essence, for example will be from the message relay of this mainboard miscellaneous part to this Video processing front-end electronics, and will be from the message relay of the miscellaneous part of this Video processing front-end electronics to this mainboard.The person skilled in art can use the standard programming technology that these software parts are provided easily, so they further are not discussed.
But, the function of this pci interface driver 1126 will be described further with reference to Figure 40.Basically, this pci interface is carried out two functions: (1) transfers to this mainboard with compressed video data from this front end-plate by dma operation; (2) transfer command and status message between this mainboard and this Video processing front end-plate.The capacity of the one-way video data service from this front end-plate to this mainboard is much larger than the capacity of this two-way news business.
As can be seen from Figure 40, by performed this processing of this pci interface driver from judge a message whether be from this mainboard input (step 1128) and.From this mainboard input, just carry out the step 1130 of this message of decoding if in step 1128, find a message then.Then, this decoded messages is used in supvr's module that is suitable for being shown in Figure 39 a relevant handled form of module and places (step 1132), and should be sent to this related management person (step 1134) by the message of reformatting.This processing is circulated back to step 1128 then.Typically, receive at this front-end processing plate, comprise program and/or order data from the message of this mainboard, for example catch in the operation in the field of this analog video plate, change the required data of sequence of video camera, be used for the change of the parameter of video data compression operation, will be by the selection or the adjusting of the performed graphic scene analytical algorithm of this front end-plate, or the like.
If do not have the message of discovery from this mainboard input in step 1128, this drive processes just before step 1136, in step 1136, has judged whether that a message will be sent to this mainboard from this front end-plate.If have, with regard to execution in step 1138, with the output message reformatting, being handled by " system commander ".As will be seen, should " system commander " be a software module, this software module is controlled this mainboard CPU and is used as a central message clearing house sending out message between this software object of being supported by this mainboard CPU mutually.After step 1138 is step 1140, in step 1140, connects through this PCI between this front end-plate and this mainboard, and output message is sent to the commander of this system.This cycle of treatment is got back to step 1128 then.
If in step 1136, do not find output message, next step is a step 1142.In step 1142, judge whether compressed video data has been ready to be transferred to this mainboard.If no, this cycle of treatment is got back to step 1128.But, if find that in step 1142 video data has been ready to be transferred to this mainboard, this processing proceeds to step 1144 then, in step 1144, a counter in this DMA selected cell 1056 (Figure 36) by in this main board memory space, this video data target location that will be transferred to comes initialization.After the initialization of this destination address, just begin of the DMA transfer (step 1146) of this video data to this mainboard.When carrying out the transfer of this video data, this processing of Figure 40 may detect an overtime condition, as shown in the step 1148.Should be noted that if for a certain reason this DMA shifts and failed, and an overtime condition just may take place.If detected one overtime, just produce a message, should overtimely be notified to this mainboard (step 1152).After this reformatting and message forwarding step 1138 and 1140, this processing turns back to step 1128 then.If an overtime condition does not take place, this interface driver is handled and just is in idle condition, shown in module 1154.
Video data compression
With reference now to Figure 41 to 44 and 39A, discuss in this upward performed video data compression operation of this digital front-end plate 590 (Fig. 4).
This VR/PC unit disclosed herein has adopted a new video data compression technology, under the typical work condition of this unit, the effective compression ratio that this new video data compression technology can provide is rough at least 250: 1, and the quality of an image that is provided is enough for the video monitor Secure Application at least.The compression of this height allows the effective use to preservation capacity (being hard-disk capacity basically), and realizes data transmission and operation in this VR/PC unit.In addition, as what will be understood from discussion part subsequently, image analysis algorithm that subsequently will be performed to this compressed video data has been held in this new video data compression technological synthesis disclosed herein.In addition, even compress technique of the present invention provides significant high compression ratio, still can carry out difficult playback function quite effectively, for example playback in reverse.
With reference now to Figure 44, summary of video data compression technology of the present invention is described and this is compressed the form of video data.
After compression, be represented as a series of data fields 1130 by this video image stream that given video camera produced.In this sequence, two class fields are arranged: reference diagram image field 1132 and " difference " image field 1134.The appearance of this reference field 1132 in this data fields sequence fixed at interval.For example, in a preferred implementation of compress technique of the present invention, per the 33rd field is a reference field in this sequence; That is, be 32 " difference " field 1134 between every pair of continuous reference field 1132.In this reference field 1132 each is not carried out compressed encoding with reference to any other image.On the other hand, compressed encoding is carried out with respect to one or more last image in each difference field 1134.
The beginning of each reference field field head of 1132 usefulness (not being presented among Figure 44), this data tile field identical (in Figure 30, not having the time code word of demonstration also to be included in this head) that this head may be discussed with the top Figure 30 of contact.Other parts of this reference field 1132 by with 2400 data tiles of this image plane in each corresponding compressed video data form.Especially, forming each (2400 brightness data tiles and 1200 color information data tiles) that this compressed video datas of other parts of this reference field handled in these 3600 data data tiles by the compressed encoding circuit structure of this JPEG chip 848 (Figure 14) forms.The person skilled in art all knows, must comprise traditional coding step, for example quadrature (DCT) conversion, the digitizing of coefficient value, and variable-length encoding by these chip 848 these performed JPEG encoding process.It to the compression factor of this reference field 30: 1 to 70: 1 magnitude.
On the other hand, in this difference field, each should " difference " data fields 1134 only comprise the compressed video data " variation ", selecteed data tile of expression with respect to a last image.When some motion being arranged in this video image stream or do not move, in this variance data field 1134, only represent data data tile seldom or do not represent any data tile, so that can realize the compression of a height.
From the lower part of Figure 44 as can be seen, typical variance data field 1134 is made up of a field 1136, this head is later on a mapping (enum) data 1138, be considered to the part of this image plane of " change has taken place " in the current image of these mapping (enum) data 1138 expressions, and correspondingly represent by the video data in current variance data field 1134.For example, this mapping (enum) data may comprise each a bit that is used in 600 minimum code unit of this image plane (MCU), can remember that a discrete levels sequence of a MCU and four data tiles is corresponding.For example, one has been found that with expression this MCU has taken place to change and used with six corresponding JPEG compressed video datas of data data tile of this MCU represented with corresponding " 1 " bit of a specific MCU.Be illustrated in the data that do not comprise this MCU in the current variance data field with corresponding this bit value " 0 " of a specific MCU.
After reformed MCU mapping 1138 is data segment 1140, and this data segment 1140 is by forming with this mapping (enum) data 1138 represented corresponding JPEG compressed video data of this reformed MCU.The number of " 1 " bit in this mapping (enum) data is identical with the number of the MCU part 1140 that is encoded.
Just now described this compress technique has been considered an image reconstruction technique that illustrates in Figure 61.According to this technology, come this JPEG compressed encoding is decoded by using a traditional mode, just can reformulate a reference field 1132.Then, for difference field subsequently, the MCU data that each section taken place to change are carried out the JPEG decoding, and suitable part that each MCU in the video data that is produced substitutes as this reformed MCU mapping (enum) data is represented, in this image plane.The process of this first difference field of reconstruct can be regarded as one is used this MCU mapping (enum) data as a process that instructs " stickup " stamp (reconstruct MCU).Then, to each difference field subsequently, repeat this process.
The key component that the compressed encoding of this difference field is handled must judge whether and will carry out JPEG coding (because it is different with reference to MCU accordingly with) or abandon this MCU (because it is identical with reference to MCU with this) a specific MCU.In a preferred embodiment of the present invention, each data tile in this MCU by with compare with reference to a corresponding data tile among the MCU at this, if and found that the corresponding reference data tile of any data tile and this is different, would just think that a MCU in this difference field is different.By calculating specific statistics from the point data of forming this reference data tile, and according to a threshold value, the identical statistics of this statistics with this reference data tile that produces previously compared, just can judge whether a data tile is different with a reference data tile.In a preferred embodiment of the present invention, the user can regulate this threshold level of expression " same degree ", so that this user can trade off to compression factor and image quality.
Illustrate to Figure 43 A-43C principle the aspect of this data tile comparison procedure.This data tile that Figure 43 A shows is one 8 * 8 dot matrix.This data tile of Figure 43 B presentation graphs 43A how to be divided into be each other separately and each comprise four a four/part of one 4 * 4 dot matrix.By carrying out addition to 8 in 16 points of each four/part, if with addition and divided by 4 (two binary digits move right), just can calculate this four/a part of data tile comparative statistics data.The statistics of one 9 bit 1/4th consequently.(these 8 with can be counted as one " contraction-expansion factor " by the divisor that removed and can be a number different with 4.)
Shown in Figure 43 C, use this 8 point that will be added selected be known as " inspection panel " sub sampling.May use two inspection panel sub sampling technology: the point that the usefulness among Figure 43 C " X " is represented can be selected to be used for addition, perhaps selects the point that does not have usage flag to represent.Under any situation, can observe, this inspection panel sub sampling must be that a factor is 2 sub sampling, and each row is offset.In a preferred embodiment of the present invention, only one of these two inspection panel sub sampling technology are used to all four/parts of all data tiles, so that do not have a pair of level or vertical adjacent point to be used to calculate the statistics of this data tile feature.
This identical technology is used to calculate the statistics of this reference data tile and the statistics of those data tiles in this difference field.If the difference of the statistics of one of the statistics of four a four/part of a reference data tile corresponding reference four with this/part is greater than this threshold value, this whole data tile then, whole M CU is considered to this with reference to MCU variation has been taken place thus.Should be appreciated that this process is used to six data tiles (four brightness, two colors) of each MCU.
Figure 39 A, 41A, 41B and 42 illustrate the processing that the compressed encoding with this difference video data field continues.With reference to figure 41A, of describing this Compression manager person's software module 1120 is set up operation (Figure 39) earlier.This Compression manager person's establishment stage in step 1142, has received the threshold value table that is respectively applied for from this input video stream of corresponding video camera since a step 1142 from this mainboard.On the image quality of these and this corresponding video signal flow and the plate that the corresponding threshold value table of compression factor parameter is stored in this front end-plate control module 842 among the RAM.Then, be written into the threshold value in this table, (step 1146) handled in the performed compression of DSP1050 and (step 1148) handled in this compression that resets to be used for.Specifically, in step 1148, suitable instruction is sent to this JPEG chip 848 required parameter is set, for example selection of quantization table and Huffman code tables, monochrome and how color-coded selection or the like.
Figure 41 B illustrates the operation of this Compression manager person's software module.Shown in module 1152, in 16 input video streams each is continued this identical operations process.At first, shown in module 1154, retrieval and the corresponding threshold data of video camera that produces this current difference field that will be encoded.Judge that then whether this threshold value setting recently appliedly is provided with different (modules 1156) with those.If do not have, using the mode identical with the mode that nearest variance data place is used to carry out compression handles, but, if this threshold data be employed recently different, just upgrade operation table (step 1158) and this compression processing (step 1160) that resets that this compression is handled.
Should be appreciated that identical threshold parameter can be used to handle brightness and color data data tile, perhaps in one aspect, this brightness data is used a threshold value, and in yet another aspect, (U, V) data is used another threshold value to color.
Figure 42 illustrates by this performed processing of this data tile comparison software module 1122.Beginning, by the data tile data of visit in this FIF0914 (Figure 19), read in this video bus and operate these 15 data tiles that the data tile stage of (seeing Figure 25) will be sent out, shown in step 1164, following processing is applied to each data tile then: retrieve this fixed reference feature statistics (step 1166) of this corresponding reference data tile and calculate the characteristic statistics data (step 1168) of this current data tile according to this technology that contact Figure 43 describes.If according to this applied threshold value, calculated characteristic statistics data of the fixed reference feature that is retrieved and this current data tile, find that a data tile is different with this reference data tile, then this data tile is labeled as " difference " (step 1170), and will this current data tile be preserved by the calculated characteristics statistics substitutes last fixed reference feature and as the fixed reference feature that is updated.But, if in step 1168, find this current data tile and this reference data tile " identical " (this threshold value of diversity ratio that is feature is little), then this data tile of mark not.Shown in step 1172, continue treatment step 1164 to 1178, compared up to all 15 data tiles with this fixed reference feature.Then, these data tiles comprise the data tile that is marked as " difference ", are returned to this video output module (step 1174).Can remember from the discussion of front, all data tiles, or be marked as " difference ", or be arranged in identical MCU with the data tile of " different ", all be provided to this JPEG process chip 848 and carry out the JPEG compression and be included in this compressed variance data field.Other all data tiles are not provided to this JPEG chip, but are abandoned simply.
Describe by the performed processing of this mapping generator software module 1124 (Figure 39) below with reference to Figure 39 A.
Shown in module 1176, each data tile is carried out following processing.At first, judge whether this data tile is marked as " difference " (step 1178).Then, as step 1180, shown in 1182 and 1184, if this data tile is marked as " difference ", just one is set to 1 with the corresponding sign of this current data tile, otherwise just with this current data tile accordingly this sign be set to 0.Shown in step 1186, this sign is fused in the current mapping byte that is being configured then.Be fused to this mapping byte (step 1188) if construct all 8 signs of this mapping byte from them, just begun a new mapping byte (step 1190) then.After all data tiles of this image had been constructed the mapping byte, the mapping (enum) data that is produced was transferred to this mainboard (step 1192) through this pci interface 850.
Although the discussion of prior figures 39A has illustrated each data tile and has used a mapped bits, should be appreciated that, as a preferred alternate embodiments, if mapped bits be assigned in six data data tiles among each MCU and this MCU any one be marked as " difference ", the sign of this given MCU just is set to " 1 ".
Mainboard software
The operation of control this mainboard CPU592 (Fig. 3) will be discussed now, make this software of this VR/PC unit of as a whole control thus.In a preferred implementation of this VR/PC unit, used the microprocessor (for example, a Pentium processor) of a standard, and under the operating system of a standard, worked, its objective is in order to make the expense minimum of mainboard.Because the multitask environment of Windows95 and its software development ability of supporting are so in a preferred implementation of this VR/PC unit, this mainboard CPU uses well-known Windows95 operating system.Application software module discussed below is according to being realized by component object model (COM) structure that Microsoft proposed.Use the language of this object based programming of C++ to produce application module.
Figure 45 provides a summary of the software part of forming this mainboard CPU application.Following the parts that illustrate are numbered: system commander 1202, set up supvr's parts 1204, planning parts 1206, safety management person's parts 1208, user interface component 1210, alarming processing parts 1212, front console parts 1214, Compression manager person's parts 1216, videograph parts 1218, image processing facility object 1220-1 is to 1220-p, image processing facilities management person parts 1222, video is searched parts 1224, archive management person parts 1226, video playback parts 1228, image analysis (playback) instrument object 1230-1 is to 1230-M, database search tool management person parts 1232, video is preserved parts 1234, and remote-control device object 1236-1 is to 1236-N and remote object supvr parts 1238.Except this remote object 1236, beyond playback analysis tool object 1230 and the image processing facility object 1220, all are not that the commander of this system 1202 parts all are shown as and communicating by letter that the commander of this system 1202 carries out that two-way news transmits.
The commander's of this system effect is as a central message switching centre, allows message to transmit between other application software parts.All pass through this distance supervisor parts 1238 to this remote object 1236 with from the message of this remote object 1236, all pass through this database search tool management person 1232 and all pass through this image processing facilities management person parts 1222 to this image analysis tool object 1230 with from the message of this image analysis tool object 1230 to this image processing facility object 1220 with from the message of this image processing facility object 1220.By the multitask environment that is provided by Windows95 operating system, can expect that each software module and object will use its processing threads or alternatively, use this main GUI thread.This application software architecture is message-oriented and event driven.The commander of this system 1202 is shown in Figure 46 A.According to the standard method of this CoM structure, the commander of this system 1202 supports two interfaces, IUnknown (label 1240) and INotifySrc (label 1242).Understand as those persons skilled in art, IUnknown interface 1240 is one and is used for reference count, the standard C OM interface that discharges storer and the interface of being supported by other com objects is conducted interviews.INotifySrc interface 1242 is to allow software part to come to after receiving specific message and representing interest the model that is established in the COM of standard design guidelines.The commander of this system keeps a mapping that message is relevant with interested group, when a message was come the commander of this system, the commander of this system searched all interested groups and excites calling the parts of interested group (party) by this INotifySrc interface.The commander of this system is employed software at first to be written into and initialized parts.Then, the commander of this system judges that from the login of a system all will be by the initialized parts of the commander of this system, these parts are written into each parts then and call an initialization function that transmits this INotifySrc interface pointer, so that can oneself be logined.
Figure 46 B illustrates other software parts that use general type.Should be noted that miscellaneous part typically supporting interface Iunknown and Inotify.
About using the background material of the message of interface to find in one piece of paper between this COM structure and the object, this paper is that K.Brockschmidt is published in May, 1996 Microsoft Systems Journal" how OLE and COM solve the problem of component software design " on the 63-80 page or leaf, one piece of relevant paper is in June, 1996 Microsoft Systems JournalOn the 19-28 page or leaf.
Figure 47 illustrates by the performed processing of this safety management person's parts 1208.From module 1244 as can be seen, this safety management person is in idle condition, after a user attempts to login.When detecting the attempting an of login, just judge whether (step 1246) this logging request is effective.If this logging request is invalid, these parts are circulated back to module 1244.If but this logging request is effective, this user just logins the system's (step 1248) that enters then, and it just judges which kind of system features the user that (step 1252) this login enters is allowed to operate then.This is to retrieve the user-dependent feature setting that enters with this login and realize by visiting a safety database 1254.According to this feature setting that is retrieved, these safety management person's parts send the order that allows operation to (step 1256) parts that each this user is allowed to visit then.The module 1258 of Figure 47 represents that also these safety management person's parts further provide a processing, and the power user who is authorized to thus can enter and edit this feature setting in this safety database 1254.
Figure 48 illustrates the operation of a typical subject of this remote object 1236.The function of this remote object is driver or the interface as the external device (ED) of this VR/PC unit.A kind of like this device may comprise external medium drive assembly (for example, an outside DAT driver), and other VR/PC unit perhaps are similar to those and are presented at this locality or remote node among Fig. 1.Preferably, the external device (ED) to each and this VR/PC unit communicate all carries out specific description with a remote object.
Shown in module 1260, be received through the received data of communication link that are connected with this remote-control device, be cushioned then (step 1262).Received data are translated into this employed messaging protocols of application software parts (step 1264) then, and the message that is produced is sent to this distance supervisor 1238 (Figure 45).Continuation is with reference to Figure 48, when a message will be by when this VR/PC unit sends to a remote-control device, this message by in this remote object 1236 suitable one from this distance supervisor 1238, receive.This message is translated into a form (step 1266) suitable, that send to this external device (ED) by this remote object 1236, is placed on then in the output buffer (step 1268), and is sent (step 1270) through this communication link.
Figure 49 illustrates by the performed processing (Figure 45) of this front console software part 1214.When user of this VR/PC unit handles switch on this front console 586 (Fig. 6), this front console software object 1214 received a corresponding signal (step 1272, Figure 49).This switch control signal is translated into the employed messaging protocol of this application software (step 1274) then, and the message that is produced is before forwarded to the commander of this system.
This front console software object is also managed the state that is provided at the LED on the physics front console 586.When this LED show state that will change on this front console 586, this front console software object 1214 receives a suitable message from the commander of this system.This front console software module is carried out work then, and this message translation is become LED register command (step 1276), and exports the data that produced, to be kept at (step 1278 in this LED register; Do not show the LED register).
Figure 50 illustrates by this and sets up the performed processing of supvr's parts 1204.Beginning (step 1280), it has judged whether to receive a request about the information of setting up that is saved previously.If received this request, just satisfy this request (step 1282).(not receiving about setting up the request of information, after step 1280 if perhaps have) after the step 1282 is step 1284, and in step 1284, it has judged whether to receive a request that will change the information of setting up that is saved previously.If received this request, just ask to change this and set up information (step 1286) and cycle of treatment turns back to 1280 according to this.If do not receive this request in step 1284, this processing directly turns back to step 1280 from step 1284 then.
With reference now to Figure 51 and 52, describes by this video and to preserve the performed processing of software part 1234.
As can be seen from Figure 51, the data that are stored on this hard disk drive comprise compressed video data 1288 and index data 1290.This video data is with corresponding from this inlet flow of all 16 video cameras (if having 16 video cameras to link to each other with this VR/PC unit and at work), and its form is and .AVI (audio/video intersect) conformance to standard of Microsoft about audio/video file.Although the embodiment of this VR/PC unit as described herein is not preserved audio-frequency information, can imagine and revise this system, so that provide audio pickup device (microphone) and digitized voice data to be stored in together with relevant video information.Flow file layout that corresponding data are used regular length with this incoming video signal and intersect preservation, wherein the file of the N shown in Figure 52 1292 is recorded on the hard disk.A preferred size of each file 1292 is rough 20 megabyte.By continuous video data stream is divided into file, can obtain restriction owing to extract the loss of data or the data corruption on the hard disk that cause.
Except the permanent video data file 1292 of standard, on this hard disk, also preserved the video data in the pre-alarm bumper portion (label 1294) that maintains this hard disk.Preferably, this to alarm in advance that impact damper 1294 uses the mode of an intersection and use be that the full frame speed (45 of the per seconds that distributes between these 16 video cameras) of this system is preserved and from this incoming video signals of all 16 video cameras basically.By contrast, should be appreciated that, not some or all 16 current may at all not being recorded in this accurate permanent file 1292 of video camera, the speed that perhaps may be used one " descending in time " is preserved, and the frequency than per second 45/16 is low basically for the speed that should " descend in time ".Preferably, this is alarmed impact damper 1294 in advance and can be used as a circular buffer realize on hard disk, for example, may preserve all in the past 60 seconds with inherent front-end electronics institute video captured field.
Refer now to the index data on this hard disk, the whole index that has covered All Files 1292 is represented with label 1296.For each file in this N file 1292, the date and time of a beginning all is provided, and the date and time of an end.To each independent file 1292, also provide an index additional, supporting paper.The index of this supporting paper is represented with label 1298, and for each video data field provides this field captive date and time, is caught this video camera, information and this field skew in this document that incident is relevant.Shown in label 1302, the given event information that is used for a particular field may comprise and being illustrated in when catching this, and the data of polytype incident have taken place.The detection of this incident can be finished by the contact alarm sensor 526 that Fig. 1 discussed and/or by the feature of analyzing this image stream.Can analyze this image in the time that receives this image stream, or the time after, analyze image when playback should image stream.Except detection incident itself had taken place, this image analysis algorithm that is used to the incident that detects may return a confidence factor value.In this case, this confidence factor that is provided by incident Detection Algorithm may be provided these data that expression has detected an incident, shown in label 1304.
In a preferred embodiment of the present invention, this index information 1290 is stored on the hard disk identical with the hard disk of preserving this associated video data file 1292, and this index information also is stored on one second hard disk.In order to search the position of this user's interest video data on this first hard disk, may visit this second hard disk then, preserve the operation of new video data thereon and can continue and visit this first hard disk, and can indexed search not interrupt.In an embodiment of the invention, two hard disks are provided, one in two hard disks is used to video data and preserves (with relevant index), and another hard disk is not used to the video data preservation, but is dedicated to backup or " shade " index information and program or similarly preservation.In another embodiment of the present invention, provide the hard disk drive of three or more.In one embodiment of back, one of hard disk drive is exclusively used in shade index (shadow index) and program information is preserved, and two other or more hard disk are used to preserve video data.
This video is preserved software part and carried out function: the buffering of management pre-alarm video data on this hard disk or these hard disks is kept at this input video stream on this hard disk and retrieval is stored in video data on this hard disk.Figure 52 uses the form of flow graph to illustrate by this video and preserves the performed processing of software module.Beginning, it judges in step 1306 whether this video preservation software part has participated in the fixed video data preservation part of pre-alarm buffer management part or its function now.If do not participate in pre-alarm buffer management, this processings is preserved next in current file that is opened on this hard disk in a large number will be by the video data (step 1308) of the permanent preservation of standard.As previously mentioned and as subsequently discussion, should be appreciated that the quantity of multitude of video data and data is corresponding, these data are wanted can be handled easily and be cushioned with preparation to be written to hard disk.Upgrade corresponding index data (step 1310) then.The end of video data file 1292 before next processing has judged whether to arrive and deserved.If, just judge then this hard disk of being used to preserve video data or these hard disks whether full (step 1314).If not, on this hard disk or these hard disks, open another video data file (step 1316).If this hard disk or these hard disks are full, step 1314 is exactly a step 1318 later.In step 1318, judge that this video data is preserved hard disk or whether these hard disks are used with the mode of a ring.If not, this video is preserved software part and is sent the message (step 1320) that an expression has arrived the end of preservation capacity to the commander of this system then.But, if in step 1318, find that hard disk is to come work with the mode of an annular, this document index list that resets then and locate to proceed preservation (step 1322) in " beginning " of this hard disk.
If to carrying out this pre-time of alarming the buffer management function, this processing forwards step 1324 to from 1306 then in step 1306 discovery.In step 1324, judge whether to detect alarm conditions.If no, next piece next one that will be stored in the ring bumper portion that video data in this pre-alarm impact damper is placed on this hard disk is preserved in the position (step 1326).It judges whether to have arrived the end (step 1328) of the ring bumper portion of this hard disk then.If arrived the end, this pointer that is illustrated in the next savepoint in this ring impact damper just is moved to the front (step 1330) of this ring impact damper.Otherwise the next one that this pointer is moved in this ring bumper portion of this hard disk is simply preserved on the position (step 1332).
If in step 1324, find to have detected alarm conditions, step 1334 is followed after step 1324 then.In step 1334, be stored in the permanent preservation part that this video data in this ring impact damper is copied into this hard disk.Shown in step 1336, copy these data to this hard disk should forever preserve partly from this ring impact damper and continue, up to having finished copy, this pointer is reset to the beginning of this ring bumper portion then.Alternatively, a new portion of this hard disk is assigned with as this ring impact damper, the front is assigned with as the part of this hard disk of a ring impact damper has formed this permanent part of preserving part, to keep this video data that is recorded in this pre-alarm impact damper before detecting these alarm conditions.
Figure 53 illustrates by this video and searches this processing that software part 1224 is continued.It is (as will be described below when a user has excited a seek operations, by carrying out reciprocation) with this graphical user interface, this video is searched software part and is handled, come to obtain the expression date from this user, time and this user the search parameter (step 1340) of source video camera of interested video information.In addition, this video is searched parts and is obtained the expression of whether using an image analysis algorithm (" instrument ") about this search from this user.Step 1342 is judge modules, and wherein it judges whether to ask a search based on an image analysis.If, this video search software part obtain from this user the image analysis algorithm that expression will be performed type input and, if suitable, obtain the parameter (step 1344) of one or more execution that retrains this image analysis algorithm.According to this information, this image analysis algorithm is prepared in this processing then, and this video data that is retrieved in this search is analyzed (step 1346).(perhaps, if this user does not ask image analysis, after step 1342) is step 1348 after step 1346, in step 1348, searches this database and retrieves the video data of being asked by this user.If asked image analysis, this analytical algorithm is added to this video data that is retrieved then.Under any situation, all report result's (step 1352) of this search.
With reference now to Figure 54, is described in detail in this treatment step shown in the step 1348 (" carry out search ") of Figure 53.This step 1348 initially comprises sets up (Figure 54, step 1354), the tabulation of one or more date and time file of mating mutually with this search criteria by this user's defined.For each file in this tabulation, shown in step 1356, carry out this following step: each file is produced this date, and a tabulation of time and video camera clauses and subclauses judges whether to use an image analysis algorithm (step 1360) then.If no,, just submit to simply then to be used for producing report (step 1362) if promptly this user does not ask image analysis.But, if asked search based on an image analysis, then to each clauses and subclauses in this tabulation of assembling in step 1358, shown in step 1364, just carry out following processing: at first, this image analysis algorithm (step 1366) that resets, and use this image analysis algorithm to analyze and the corresponding video image sequences of these clauses and subclauses (step 1368) then.Judging in step 1370 then whether this image sequence has will be by the feature that this image analysis algorithm detected.If this sequence is added to positive the results list (step 1372), and the index information of this document is updated and represents to have detected this incident (step 1374).That is, in Figure 51 1302 shown in, the data relevant with this incident are updated and represent to have detected this incident, and are used for the confidence factor that this event detection is judged.Should be appreciated that if do not find to have this feature of this image stream, this sequence is not added to this results list and does not upgrade this index information.Under any situation, after step 1374,, just after step 1370, judge more clauses and subclauses (step 1376) in this tabulation, whether to have occurred if perhaps do not detect interested feature.If no, with regard to the result (step 1362) of report as a result of this image analysis.But, if more clauses and subclauses, just retrieve next clauses and subclauses (step 1378), next clauses and subclauses are carried out this circulation of beginning in step 1364.
Figure 55 provides a summary by this performed processing of this video jukebox software parts 1228.By the slow rotary switch 660 of user's manipulation on this front console 586, perhaps interact by user and this graphical user interface, excite the video playback operation, as described below.In some cases, after having finished a search, just import this video playback capability automatically, show this video data by this user's request.
Shown in Figure 55, the beginning step of this video playback capability is to judge sent out which kind of play command (step 1380).If sent out a pause command (step 1382), just interrupt this video data decompression operation (step 1384) then, and this video playback capability has been suspended video playback (step 1386) to the commander of this system report.If used the speed of a given per second X field, sent out a quick play command (step 1388), just interrupt this decompression operation (step 1390) again then and this quick mode that resets is play visual speed (1392).Just restart this video data decompression operation (step 1304) then, and report the playback rate (step 1396) of this new request to the commander of this system.
Come reverse play (step 1398) if selected to use the speed of Y image of per second, just interrupt this decompression operation (step 1402) then again, and the visual speed (step 1404) of this reverse regeneration mode that resets, and begin a reverse decompression operation (step 1406).After step 1406 is abovementioned steps 1396, and in step 1396, this requested playback rate is reported to the commander of this system.If in step 1382, situation that will be detected in 1388 and 1398 all occurs, and then a broadcast state does not have reformed message to be sent to the commander of this system (step 1408).
Video data decompression (forward direction)
Below with reference to Figure 56 this treatment step 1394 shown in Figure 55 is described in more detail.
This processing shown in Figure 56 is from receiving the order (step 1410) that restarts of this decompression engine.Be later on step 1412, in step 1412, this playback rate timer be set, and remove this and withdraw from the sign that decompresses according to the speed of employed per second X Images.
Be later on step 1414, in step 1414, finding out this will be by first reference picture in the video flowing of playback in the later position of point that will begin playback.(reformulation) this reference picture then decompresses in step 1416.After step 1416, be step 1418, in step 1418, judge whether to have asked to interrupt this decompression engine.If a message is sent to the commander of this system, represents to have interrupted this decompression engine (module 1420), and stopped this decompression operation.But if do not find to interrupt this compression in step 1418, this processing just moves to step 1422 then, in step 1422, has judged whether the time that next image is decompressed.If not, this cycle of treatment turns back to step 1418.But if the time that next image is decompressed, this processing forwards step 1424 to, and in step 1424, this next image that judgement will be decompressed is a reference picture or a difference image.If this next image is a reference picture, just use one to be used for processing (module 1426) that a reference picture is decompressed, this cycle of treatment turns back to step 1418 then.If this next image is a difference image, just use one to be used for processing (module 1428) that this difference image is decompressed then, this processing turns back to module 1418 again then.
With reference now to Figure 57, describes and to be used for processing ( module 1416 and 1426 of Figure 56) that reference picture is decompressed.Shown in Figure 57 this handled by the circulation cover and formed, being each (each image 30 row) of going that are used for this image minimum code unit by module 1430 represented outer circulations wherein, be (20 MCU of each row) that are used for each MCU of this current line by module 1432 represented should interiorly circulating.
In step 1434, each in six pieces of processing jpeg encoded data is come original six data tiles (four brightness and two colors) that this JPEG coding is carried out inverse operation and recovers video data basically.Controlling a general purpose microprocessor, to come the program that the JPEG coding video frequency data is decoded be well-known, so do not need here to be described.Decoded, be copied into an output buffer (step 1436) then with the corresponding video data of this MCU.As long as decoded this reference picture all the row in all MCU and they have been placed in this output buffer, the data that are cushioned of representing this whole decoded picture are transferred on the level of bit, to be presented on this monitor (step 1438).
With reference now to Figure 56 and 58, the module 1384,1390 and 1402 of Figure 55 further is discussed.Shown in Figure 58, when having received an order that stops to decompress (step 1440), the sign (step 1442) of " withdrawing from process " just is set then.If a quick playback operation has taken place then, this sign that withdraws from process triggers a searching (Figure 56) certainly in module 1418 then, causes closing this decompression engine, and sends a suitable message to the commander of this system.As what will can see from the discussion about the processing of reverse regeneration, this withdraws from process identification (PID) has similar effects to this reverse regenerative operation.
With reference now to the expression of Figure 59 and disparity map image data as shown in figure 44,, describes by the performed processing (" decompression difference image ") of module 1428.Shown in module 1444, the beginning step of this difference image that decompresses is the data of reading in expression and the position of corresponding, the reformed MCU data of this difference image in this image plane.Carry out this circulation cover then, wherein this outer circulation is represented and each row of the MCU in this image plane all carried out (each image 30 row) by module 1446, should interiorly circulate, and each MCU in this row is all carried out (20 MCU of each row) by shown in the step 1448.
To each MCU, from this mapping (enum) data, pick up corresponding this bit (step 1450) with this MCU, judge then whether (step 1452) this MCU in this image plane in this current image change has taken place.For example, if the value of this bit is " 0 ", this MCU does not just change then, but when the value of this bit is " 1 ", just represent that this MCU has taken place to change and with this MCU corresponding this more new data be included in deserve before in the video data field.If run into " a 0 " bit, this processing just circulates simply then, to pick up the bit of next MCU.When the motion in this image seldom or was not moved, this MCU mapping generally was very sparse, so that the whole row of MCU may all not change.But when having run into a reformed MCU, this processing of Figure 59 just forwards module 1454 to, in module 1454, and next piece of the reformed MCU data of decoding.The decoding of these MCU data may be by being carried out by the identical program of the standard program of reference with the step 1434 of getting in touch Figure 57.
Continuation is with reference to Figure 59, after these reformed MCU data of having decoded, this processing decides a suitable skew according to current line and the column count of this MCU, so that decoded just now MCU data block is introduced on the correct position of this image plane (step 1456).Then, according to the skew that is produced, this decoded of output MCU data is refreshed this display buffer (step 1458).Figure 61 with diagrammatic representation step 1456 and 1458 result.Figure 61 explanation connects a MCU ground based on a MCU and produces current this decoded difference image, upgrades an image that is shown previously.As previously mentioned, locational " stamp " that this reformed MCU can be counted as being adhered in this image plane, this reformed MCU mapping (enum) data of basis is determined.
After the step 1456 and 1458, this processing circulates, to obtain this mapping (enum) data bit of next MCU in this image plane.
With reference now to Figure 60, the additional detail of the step 1454 of Figure 59 is described.Begin to pick up the compression encoded video data (step 1460) of a buffer sizes after to reformed MCU data one decodes.Judge whether that then enough compressed video datas carry out decoding processing (step 1462).If have, just use this canonical solution coded program noted earlier, come carrying out inverse operation (step 1464) by this performed on this front end-plate JPEG coding.If when in step 1462, finding not have enough compression encoded video datas to decode, just refill this impact damper then, shown in step 1466.In addition,, run into an end that data are preserved file, just opened next data file (step 1468 and 1470) then if when refilling this impact damper.At an alternative embodiment with preferred embodiment, being used for replenishing visual data can disposablely finish, and can not want step 1462 like this, 1466,1468 and 1470.
With reference now to Figure 62 A and 62B, be described in the substitute technology that refreshes this display buffer during the playback operation.
Figure 62 A has shown a technology that makes the refresh rate that is used to provide 18 of per seconds in this VR/PC unit.Beginning to 600 MCU use JPEG decoding of the point data in the reference picture, is perhaps used JPEG decoding (step 1472) to reformed MCU data in the difference image.Mainboard CPU uses a vertical interpolation operation to come to obtain 480 row point data (step 1474) from the internal data of this 640 point * 240 row is represented then.The back is a further software processes step, wherein this yuv data is translated into RGB data (step 1476), these data that are translated comprise 640 point * 480 row, every is three bytes, and in step 1478, be cushioned, the speed that is used 18 of per seconds is then carried out transfer of bits, drives a SVGA monitor.
Figure 62 B illustrates a substitute technology of the refresh rate that 30 of per seconds are provided during playback.According to this technology, with identical as the technology of Figure 62 A, used identical beginning JPEG decoding step 1472, but these decoded data are sent to a hardware module 1480, this hardware module 1480 is used one 2 times vertically scale function and then 480 line data that produced is sent to another hardware module 1482, so that color space is translated RGB from YUV.Directly from these hardware 1482 these RGB data of output, drive this SVGA then with the refresh rate of using 30 of per seconds.Though its cost has provided additional hardware component 1480 and 1482, the what is called that illustrates among Figure 62 B " is directly drawn " technology, except providing one faster the refresh rate, has also reduced the burden of this mainboard CPU.
With reference now to Figure 63 A and 63B, be described in employed alternative record and playback strategy in this VR/PC unit.
This first alternative strategy that Figure 63 A Central Plains illustrates reasoningly farthest obtained this unit write down simultaneously respectively by several (for example, 16) dirigibility of the video signal flow that video camera produced that links to each other with this unit, but the vertical resolution of the image that is provided is 240 row, is half of commercial broadcasting standard generally.But, have been found that and use interpositioning to produce 480 row that this vertical resolution is enough at least concerning video surveillance applications.Under any situation, in the technology shown in Figure 63 A, " three-band (tri-corder) " time slot 1484 (corresponding with one of three field locking channels 804 of this front end analog board (Figure 13)) is assigned to a field that is produced by video camera X a preset time.Operate this front end analog board like this, make the shake of expression motion or reformed MCU with wrong minimum so that only catch odd field.From being hunted down then by parallel processing of this video camera X, carry out digitizing and compression by this front-end electronics, shown in step 1486, and be used as a single .AVI data stream and be stored on this hard disk 1488.
When the request playback should be flowed by the image that video camera X is produced, just from hard disk 1488, this corresponding .AVI stream of software regeneration that the mode that (module 1490) discussed above using is decoded (decompression) uses this recovery stream to drive a SVGA monitor (module 1492) then.
Figure 63 B illustrates a substitute technology, wherein operate this VR/PC unit like this, when coming that a video camera carried out playback, basically the commercial broadcasting vertical resolution of standard is provided, but its cost be reduced widely other video cameras of linking to each other with this VR/PC unit the record resource that can use.In the technology of Figure 63 B, in the field locking channel 804 of this front end analog board two are represented, are exclusively used in separately and catch odd field and the even field that is produced by video camera X by " three-band " time slot 1494 and 1496 among Figure 63 B.Should only catch even field by " three-band " time slot 1494, " three-band " time slot 1496 is only caught odd field.
Subsequently in the processing that is saved in disk, the odd field of the even field of this video camera X and this video camera X is processed into these two fields two incoherent video signal flows seemingly.Like this, the digitizing and the compression of the digitizing of even field and compression and odd field are separated, shown in module 1498 and 1502.Because can obtain one the 3rd bid, so the 3rd channel can be used by another video camera or be shared by other video cameras, so that digitizing of one or more additional stream (not having among Figure 63 B to show) and compression are carried out with the digitizing and the compression of this odd field that is produced by this video camera X and even number field flow.In any one incident, these two independently stream be used as two independent .AVI files and be kept on this hard disk 1488, and by with two independently file manage and retrieve.As a result, the arrangement shown in Figure 63 B allow this VR/PC unit to preserve to be produced by video camera X, vertical resolution is the images of 480 row.
Because this even field and odd number field flow are independent the preservations, when carrying out playback, several software decode options are arranged, shown in module 1504.For example, because all required data of vertical resolution appear on this hard disk 1488 fully, these two streams are played and intersect, and provide 480 row of an intersection to show, shown in 1506.When using forward direction or reverse movement to play, the method that calculated amount is less can be known as " interrupt and fill " (label 1508), the only stream in these two streams of a playback, and carry out vertical interpolation 480 row are provided.But, when this played-back image stream is suspended, may be reproduced from the field that other streams come and produces an image that complete vertical resolution is arranged.
Reverse video data decompresses
Describe now when always regenerating this image stream with a negative side, the required processing of this compression encoded video data decompresses.Beginning will provide a concept nature of this processing to summarize with reference to figure 64A and 64B.
In Figure 64 A, label 1510 has shown form and has got in touch the identical a series of compressed video datas field of form that Figure 44 discusses.But for the example that illustrates among Figure 64 A and the 64B, supposing only has three difference images 1134 between two continuous reference pictures 1132, rather than actual employed 32 difference images in the preferred implementation of VR/PC unit.From form shown in Figure 44 as can be seen, label 1138-1 represents to be included in the reformed MCU mapping (enum) data of this disparity map image data field 1134-1 in the 1134-3 to 1138-3.Use identical expression, label 1140-1 represents to be included in the reformed MCU data of this disparity map image data (original text for reference to) 1134-1 in the 1134-3 respectively to 1140-3.Should be appreciated that the direction that direction from left to right in this image field sequence 1510 and the time when writing down this advance is corresponding.In other words, the time series that produces and write down this is 1132-1,1134-1,1134-2,1134-3,1132-2.Label 1512 among Figure 64 A illustrates in reality performed series of preprocessing step before carrying out playback in reverse at this image sequence shown in the reference number 1510.
Produced with one with begin to be compressed the corresponding image of the image that forms this reference data field 1132-1, then this pre-treatment step read with following difference image 1134-1 accordingly by " change " MCU mapping (enum) data 1138-1.Because this mapping (enum) data 1138-2 represents next difference image 1134-1 and this current visual 1132-1 MCU part different, in image plane, this mapping (enum) data 1138-1 also represents when forming next image, will be by the MCU of " stickup " in this current image.Because this reason is selected the corresponding visual MCU with this data fields 1132-1 according to mapping (enum) data 1138-1, be formed on " back is to stamp " the data 1514-0 that will be used to this current image of reconstruct in the playback in reverse.After having preserved " will be reformed " MCU1514-0, use this mapping (enum) data 1138-1 again, this time be to come reconstruct and the data fields 1134-1 accordingly should image, upgrade image accordingly with this data fields 1132-1 by " stickup " reformed MCU data 1140-1.
As getting in touch described this process of Figure 59, this mapping (enum) data 1138-1 is used to use the mode that illustrates in Figure 61, come " guiding " decoded MCU " stamp ".(still, different with the process of Figure 59, the reference picture that is produced is not output to display.)
On this aspect, the mapping (enum) data 1138-2 of advisory data field 1134-2 decide will with which and 1134-1 accordingly should image MCU as preserving with corresponding " will the be reformed " MCU1514-1 of this data fields 1134-1.Then, identical with the front uses this mapping (enum) data 1138-2 to come by pasting MCU data 1134-2 for the second time, with produce one with the corresponding reconstructed image of this data fields 1134-2, upgrade with a 1134-1 accordingly should image.
Below, use this mapping (enum) data field 1138-3 decide will with which and 1134-2 accordingly should image MCU as preserving with corresponding " will the be reformed " MCU1514-2 of this data fields 1134-2.Then, use this mapping (enum) data 1138-3 to guide MCU data 1134-3 again, to produce one and a corresponding reconstructed image of 1134-3 by the image that upgrades this 1134-2.Be represented as 1134-3R reconstructed image (reconstruct), that produced among Figure 64 B, then in the playback in reverse sequence below, preserve to " reference " image after being used as one.Finished this preprocessing sequence 1512 now, then as the direction from right to left of front, this playback in reverse is handled may forward a sequence 1514 shown in Figure 64 B to.
Beginning is in sequence 1514, and output is shown to " reference " visual 1134-3R (this image of being caught before the reference picture 1132-2 is corresponding with catching) by the back of reconstruct.
Then, use this back of guiding to upgrade visual 1134-3R, upgrade this back and produce and the corresponding image of " the next time is upward than the image of morning " (promptly corresponding to difference video data field 1134-2 image) to reference picture 1134-3R so that MCU connects a MCU ground to this mapping (enum) data 1138-3 of stamp (" will reformed MCU ").Then, in turn, use reformed MCU mapping (enum) data 1138-2 will be reformed " back is to stamp " 1514-1 be directed on the position suitable in this image plane, upgrade image accordingly, to form and the corresponding image of this video data field 1134-1 with data fields 1134-2.
Below, use the image that will reformed MCU1514-0 upgrades this 1134-1, to form and a corresponding image of 1132-1 by mapping (enum) data 1138-1 guiding.Alternatively, whole reference field 1132-1 is decoded again, and the time that 1514-0 spent is long but the time ratio of this cost uses " back is to stamp ".In this, this process of being discussed with respect to sequence 1512 above can using again, but, what use specifically is in image stream, at front, last this reference picture that occurs of this reference picture 1132-1 be right after one group of disparity map image data field before this reference picture 1132-1.
In the remaining part of Figure 64 A and 64B, shown illustrating the decompression technique that is used for the playback in reverse operation simply.As the example of a simplification, shown an image sequence 1516.This sequence 1516 comprises visual 1132-1E, 1134-1E, 1134-2E, 1134-3E and 1132-2E.Done following supposition, so that can not use unsuitable complex image to illustrate the principle of this decompression technique:
(1) supposes that each image is made up of one 4 * 4 minimum code cellular array.
(2) shape of each minimum code unit is square, rather than 4 * 1 rectangle data tiles of the MCU structure in the actual preferred implementation of this VR/PC unit.
(3) this initial visual 1132-1E is complete white.
(4) view image fields at the most left MCU place that accurately enter the lastrow of MCU dimensionally with a corresponding object of MCU, and accurately enter in the time of this image 1134-1E, this process is used the accurately speed of a MCU of every frame then, passes this image plane with a direction towards the right side.
1518 shown be respectively with this difference image 1134-1E, 1134-2E and 1134-3E corresponding mapping (enum) data 1138-1E, 1138-2E and 1138-3E.(still, should be noted that 1518 only demonstrate 4 bits in front of mapping (enum) data 1138-(N) E.According to this exemplary image that is presented among the 15-16, last 12 bits of each mapping (enum) data will be " 0 " entirely, and be omitted to simplify this image.)
Inspection is displayed on this mapping (enum) data in 1518, can see, this image 1134-1E only has a reformed MCU (first MCU in top line), so that the value of only corresponding first bit is " 1 " among this mapping (enum) data 1138-1E.In the visual 1134-2E of the next one, change has taken place with respect to last image in preceding two MCU in top line, so the value of preceding two bits in this mapping (enum) data 1138-2E is " 1 ", remaining bit is value " 0 ".In the visual 1134-3E of the next one, change has taken place with respect to last image in the second and the 3rd MCU in top line, so produced the mapping (enum) data " 0110 " shown in 1138-3E.
These corresponding reformed MCU data are illustrated in 1520.As can be seen from the figure, for the data of this first difference image of expression 1134-1E, only a single MCU data block (black " stamp ") has been formed this reformed MCU data 1140-1E.The reformed MCU data 1140-2E of next image comprises one black " stamp " after one white " stamp ".The reformed MCU data 1140-3E of next image is identical, promptly comprises after one white " stamp " by one black " stamp ".
After pre-service above-mentioned discussion,, just read this mapping (enum) data 1138-1E about this sequence 1512.Value " 1000 " means a MCU who only needs to preserve this reference picture 1132-1E, and forming thus will reformed MCU data 1514-0E (" back is to stamp "-entirely is white).Below, use this mapping (enum) data 1138-1E to come the MCU position in the top line of this MCU to use the reformed MCU data of all black stamps, produce by the difference of reconstruct image 1134-1E, upgrade this reference picture 1132-1E thus.Read this mapping (enum) data 1138-2E of next difference image then.This value " 1100 " expression needs to preserve this by preceding two MCU of reconstructed image 1134-1E (a black stamp is later on a white stamp), and forming thus will reformed MCU data 1514-1E (" back is to stamp ").Then, a reformed MCU meets a reformed MCU, upgrades this image 1134-1E, to form this image 1134-2E.Especially, a bit connects a ratio and specially reads this mapping (enum) data 1138-2E, represented as value " 1100 ", first stamp of these MCU data 1140-2E is directed into the MCU position in the top line, and the next stamp among the 1140-2E (complete black) is directed into the 2nd MCU position in the top line then.Then, preserve " will be reformed " MCU again.So just read this mapping (enum) data 1138-3E, and find the second and the 3rd MCU (corresponding with a black stamp of a white stamp) in this top line with the back be selected to form should (" back be to stamp ") will reformed MCU data 1514-2E.
Be that a MCU connects the step that this image of MCU ground renewal 1134-2E forms this image 1134-3E below.As previously described, use this mapping (enum) data 1138-3E that the reformed MCU1140-3E of this forward direction is directed on the correct position in this image plane, finish this step.Because 1134-3E is last the difference image before a reference picture, show so preserve this visual 1134-3E by reconstruct, then should be used as one by the visual 1134-3E of reconstruct after to " reference " image.
In this, finished this pre-treatment step, and actual playback in reverse operation may take place.Beginning shows that this is by the visual 1134-3E of reconstruct.Then, read this mapping (enum) data 1138-3E and use this mapping (enum) data 1138-3E to guide this back, form this image 1134-2E to upgrade this image 1134-3E to MCU data 1514-2E.Below, read this mapping (enum) data 1138-2E and use this mapping (enum) data 1138-2E to guide this back to MCU data 1514-1E, form this image 1134-1E to upgrade this image 1134-2E.At last, use this mapping (enum) data 1138-1E to guide this back, form this reference picture 1132-1E to upgrade this difference image 1134-1E to MCU data 1514-0E.From foregoing data as can be seen, then below to this reference picture be created in the former difference group of picture of this image 1132-1E at first and carry out this preprocessing sequence.
From the description of front, be appreciated that this mapping (enum) data 1138 is used to playback in reverse decompression operation and forward direction playback operation about employed this decompression technique the playback in reverse operation.Especially, in this playback in reverse decompression technique, use the purpose of this mapping (enum) data 1138 obviously to have three:
(1) for use one with in forward direction playback decompression operation the similar mode of employed mode, during this pre-treatment step, at a forward direction reconstruct difference image.
(2) in order to select this " back to stamp " (will reformed MCU) 1514.
(3) for during this actual playback in reverse, guide this reformed MCU.
Can be sure of, during reverse decompression, repeatedly use and use efficiently this mapping (enum) data represented this compressed video data form (Figure 44) disclosed herein and accordingly of compress technique particularly advantageous aspect.
Compare with the background example that illustrates among Figure 64 A and the 64B, described in more detail by the represented processing of the module 1406 of Figure 55 with reference now to Figure 65-68.
Beginning is with reference to figure 65A and 65B, and this back is afterwards this image speed timer is set and removes this and withdraw from mark (step 1522) to playback rate according to required to the beginning step of decompression.Then, in step 1524, find out in time and corresponding this reference picture of the time point that this playback in reverse operation will occur, this reference picture (step 1526) then decompresses.Is step 1528 after the step 1526,1512 represented these preprocessing sequences among step 1528 and Figure 64 A are corresponding.
Next step is a step 1530, in step 1530, judges whether to be provided with the mark that this withdraws from process.If be provided with this mark, a suitable message sent to the commander of this system (step 1532), and stop this back to decompression operation.If do not have to find to be provided with the mark that this " withdraw from process " in step 1530, then this processing just proceeds to step 1534, in step 1534, required played-back image speed is judged whether it is to decode and show the time that next is visual.If not, this cycle of treatment turns back to step 1530.But, if the decoding and show the time that next is visual, just carry out step 1536, in step 1536, judge that this next image is a reference picture or a difference image.If this next image is a reference picture, just according to a suitable process this reference picture (step 1538) of decoding, set up next " back is to historic buffer " (step 1540) then, the preprocessing process that is about to be presented at 1512 among Figure 64 A is applied to down (being right after before writing time) one group of difference image.
If in step 1536, find will be decoded this next image be a difference image, that uses the front then and produced is back to historic buffer and according to a suitable process this difference image of decoding.
Forward Figure 66 now to, will further explain " setting up the back ", the step 1528 among Figure 65 A and the 65B and 1540 to historic buffer.Beginning (step 1544), just this this reference picture (shown in the 1132-1 of Figure 64 A) of record is decompressed before the difference group of picture that will decode now, and the form that is used record then is kept in the impact damper of " Shang Yichang ".Then, in step 1546, this process is picked up and these difference image 1134 corresponding mapping (enum) datas, and this variance data 1134 is being recorded later in the time of this decoded just now reference picture.This picked mapping (enum) data uses and the corresponding reformed MCU data 1140 in this first difference field, comes a MCU to connect a MCU ground renewal and is somebody's turn to do " a last field buffer " (step 1548).Be a circulation then, shown in step 1552, and use this circulation in other difference images 1134 in this now decoded difference group of picture each.As this round-robin one first step, it is a step 1554, and picks up and be used for recently by the mapping (enum) data of this difference image of the back of this difference image of reconstruct.Then according to next difference field mapping (enum) data produce (step 1556) from just now by the difference image of reconstruct will reformed MCU (back is to stamp 1514), use the reformed MCU data 1140 of forward direction to upgrade a field buffer on this then, come this next difference image (step 1558) of reconstruct.After all required iteration of circulation 1552-1558, produced 1512 represented back by among Figure 64 A and the 64B to the historic buffer data.
Figure 67 detailed icon has illustrated module 1544 corresponding these processing with Figure 66.The form of this processing of Figure 67 is three circulation covers, and the outer circulation of this circulation cover is represented by module 1560, and each row among the MCU is carried out this outer circulation.It is represented by module 1562 to circulate in the middle of this circulation cover, and each MCU in this row is carried out this centre circulation.The interior circulation of this circulation cover is represented by module 1564, and should interior circulation to each the data tile execution among this MCU.Shown in module 1566, each data tile is carried out a decompression and recording processing then.The details of module 1566 is displayed among Figure 68.Beginning shown in step 1568, is handled the coefficient value that recovers to be quantized with corresponding this compressed video data of this data tile.Then, this DC coefficient is come record as an absolute value, rather than come record as comparing a difference value that is produced with last data tile DC level.
Then, come other coefficients are carried out huffman coding and variable-length encoding, and preserve the data tile data (step 1574) that this is recorded according to the video data compression technology (step 1572) of standard.(as described in just now, during the forward direction playback operation, can carry out after producing one, shown in the module among Figure 56 1575, realize the quick switching of operation from the forward direction to the playback in reverse to the operation of historic buffer.)
Compressor reducer manager software parts
With reference now to Figure 69 A and 69B and 70A and 70B, describe these compressor reducer manager software parts 1216 that are presented among Figure 45 in detail.Should be appreciated that this compressor reducer manager software parts 1216 are handled the communication of carrying out between this mainboard CPU and this front-end electronics (particularly front controller DSP1050, as Figure 36).
Figure 69 A and 69B illustrate these compressor reducer manager software parts and relate to processing command, promptly are sent to the message of this front-end electronics from this mainboard, the aspect.Beginning judges whether to send an alert command in step 1576.If send an alert command, be set in the alarm conditions data byte (step 1578) with corresponding this bit of these current alarm conditions or a plurality of alarm conditions or a plurality of bit, use a suitable form to produce a message that delivers this alarm conditions byte then, to be received (step 1580) by this front controller DSP.
An order of the sequence of the video camera inlet flow of preserving if will be referred to be hunted down sends to this front end-plate (step 1582), produces the data set (step 1584) that expression is used to catch the required sequence of this camera signal stream then.The message that is produced is formatted, to be received (step 1586) by this front end-plate controller.
In step 1588, judge whether that rate command be preserved in a field sends to this front-end electronics.If being preserved rate command, a field sends to this front-end electronics, with regard to producing this front-end electronics is programmed to the data (step 1590) of using this required speed to come the capture video data place to need, then, the message that is produced is formatted, to be received (step 1592) by this front-end electronics.
In step 1594, judge whether and an order that relates to an image analysis algorithm will carrying out in the scene photo picture analysis module 846 (Figure 14) in this front-end electronics will be sent to this front-end electronics.Continue with reference to figure 69B, if send an image analysis algorithm order, just produce then relate to this image analysis data (for example, the data of this algorithm that detailed description will be performed, perhaps be used to retrain the parameter of the execution of this algorithm) (step 1596), be used a suitable form then and place, to be received (step 1598) by this front-end electronics.
In step 1602, judge whether and an order that relates to video data compression will be sent to this front-end electronics.If an order that relates to video data compression is sent to this front-end electronics, just combination is used to select the data (step 1604) of a compression parameters.For example, can send a compression parameters, this compression parameters is used to select a normal data tile " homogeny " threshold value, perhaps second stricter threshold value that better reproduced picture quality is provided, perhaps one not too strict and can reduce in the 3rd parameter of par of the compressed video data that every place produces one.Alternatively, this compression parameters can be used for being provided for quantizing the number at the bit of the coefficient of this JPEG chip 848.
Should be noted that preferably, the packet of being assembled comprises which video camera of sign will use the data of these data in step 1596 and 1604, so that this image analysis algorithm and/or should compression handle can be all different to each video camera.
Step 1604 is a step 1606 later, in step 1606, this compression parameters bag is formatd, to be received by this front-end electronics.In step 1608, judge whether and a message will be delivered to this front-end electronics, to control the operation of this scene display process module 844 (Figure 14).For example, from the size of this image of this scene display module 844 outputs, the number of the video flowing that will be shown simultaneously, for example 1,4,9, perhaps 16, and/or the distribution of video camera stream between a plurality of display windows can be changed.If made a sure judgement in the step 1608, in step 1610, produce this suitable on-the-spot display command sequence then, and in step 1612, this suitable on-the-spot display command sequence is formatd, so that the message that can correctly be handled to be provided in this front-end electronics.
In step 1580,1586,1592,1598,1606, in 1612 after any one step, be a step 1614 perhaps, in step 1614, this related message is coupled to this pci bus 596 (Fig. 3) to be sent to this front-end electronics.
With reference now to Figure 70 A and 70B, describe and to relate to by aspects processing, these compressor reducer manager software parts of this mainboard from the status message that this front-end electronics received.Shown in step 1616, received the message data of coming through pci bus, just detect the type of this message then.
If finding this message is an alarm information (step 1618), this alarm data (step 1620) of just decoding then, and use a message (step 1622) that the suitable form of this mainboard CPU software part is produced this alarm data of report.For example, this alarm information can describe type that detected by an alarm sensor 526 (Fig. 1) or by this alarm event of being detected by the performed scene photo picture analysis of this front-end electronics in detail.
Continue with reference to figure 70A and 70B, if finding this input message is a message (step 1624) that relates to the sequence of the video camera stream that is used to write down, just decoding is included in these data (step 1626) in this message, and it is formatd to exchange message (step 1628) with this mainboard CPU software.
If relating to a field of being realized by this front-end electronics, this input message catches speed (step 1630), this field rate data (step 1632) of just decoding then, and use a form that is suitable in this mainboard CPU software, using to come these data are formatd (step 1634).
If this input message relates to an on-the-spot image analysis algorithm (step 1636) performed in this front-end electronics, just decoding then, this relates to the data (step 1638) of this algorithm, and uses a form that is suitable for using in this mainboard CPU software to come these data are formatd (step 1640).
If this input message relates to a parameter (step 1642) that is used for being controlled at the squeeze operation that this front-end electronics is being performed, this compression parameters data (step 1644) of just decoding then, and use a form that is suitable in this mainboard CPU software, using to come these compression parameters data are formatd (step 1646).
If this input message relates to the on-the-spot display process of carrying out (step 1648) in this front-end electronics, this data (step 1650) of just decoding then, and use a form that is suitable in this mainboard CPU software, using to come these data are formatd (step 1652).
In step 1622,1628,1634,1640,1646, perhaps in 1652 after any one step, it is a step 1654, in step 1654, related message is before forwarded to the commander of this system software part 1202 (Figure 45), and is passed to other software parts that need to receive this message by the commander of this system.Typically, in module 1624,1630, this detected message is that this front end-plate of expression has realized that the front is sent to the status message of the order of this front-end electronics from this mainboard in 1636,1642 and 1648.
With reference now to Figure 71, the aspect that relates to these compressor reducer manager software parts of handling the input compressed video data is described.Supposed that the DMA that a video data has taken place shifts.This compressor reducer supvr judges whether this DMA video data transfer has finished (step 1656) then.Shift if finished this DMA video data, just shift the received just now video data that this is cushioned, it is recorded on this hard disk (step 1658), judged whether to have arrived the end (step 1660) of an image then.If do not arrive the end of an image, this cycle of treatment turns back to step 1656.But, if arrived the end of an image, just produce then and corresponding this index data (step 1662) of received this video data just now, just produce and preserve this video data of expression the position this map and this video data is kept at the later position (step 1664), position of this map.At last, in a step 1666, this received video data and this index data are before forwarded to this video and are preserved software part 1234 (Figure 45).
With reference now to Figure 72, this videograph software part 1218 (Figure 45) is described.Basically, this videograph software part is carried out three functions.The first, the parameter that these parts are before forwarded to this front-end electronics by generation (step 1668) by these compressor reducer manager software parts 1216 quilts is provided with message and sets up this squeeze operation.Further, suitable initial message is provided to this video and preserves software part 1234 (Figure 45), shown in step 1670.At last, the suitable message that this performed compression processing operation is produced in step 1672 in this front-end electronics excites and is carried out preceding commentaries on classics by these compressor reducer manager software parts.
With reference now to Figure 73, the aspect that relates to the archive management person software part 1226 (Figure 45) of handling the request of search is described.
Figure 74 A and 74B illustrate the aspect of the archive management person software part that relates to command process.In essence, processed order relates to those and needs to carry out " magnetic tape recorder " function or the similarly order of function in the processing shown in Figure 74 A and 74B.
A first step 1674 this archive management of the expression person parts that are presented among Figure 73 have received a search request.This search request analyzed (step 1676) and then this order be translated into a form (step 1678) that is suitable for further being carried out processing at this archive management person software part.
With reference to figure 74A and 74B,, just produce one then and begin message (step 1682) accordingly if detected a start-of-record order (step 1680).If detected one cease and desist order (step 1684), just produce one then and stop message (step 1686) accordingly.Be written into order (step 1688) if detected one, just produce one then and be written into message (step 1690) accordingly.If detected a play command (step 1692), just produce the corresponding message (step 1694) of playing then.If detected an eject command (step 1696), just produce a corresponding Pop-up message (step 1698) then.Restart order (step 1702) if detected one, just produce one then and restart message (step 1704) accordingly.If detected a seek command (step 1706), just produce the corresponding message (step 1708) of searching then.If detected the order (step 1710) of unrolling, just produce the message (step 1712) of unrolling accordingly then.If detected a skip command (step 1714), just produce a corresponding redirect message (step 1716) then.Under the situation of each message of being mentioned in this section, this message is before forwarded to a file software object (step 1718).This files to liking a driver software function of control this file DAT driver 626 (Fig. 3) or an outside file medium actuator unit that connects.For example, a DAT driver independently, the DVD driver, magneto optical driver perhaps similarly can be connected to this VR/PC unit by SCSI mouth 628.
Can imagine that this file is preserved and/or search operaqtion can use two or more archive devices to operate simultaneously, these archive devices may comprise the outside device that is connected of DAT driver 626 and one or more.
The alarm processor parts
The operation of this alarm processor software part 1212 is described below with reference to Figure 75 and 76.For Figure 75, suppose, received an alarm information from this front-end electronics.In step 1720, judge according to the agreement of a standard or the agreement of a client definition whether this user is selected to handle these alarms then.If selected the agreement of a standard, be step 1722 after the step 1720 then.In step 1722, this alarm processor impels one or more predetermined alarm output signal of generation according to the type of received alarm information.For example, this alarm output signal or a plurality of signal can be to close the door automatically or lock a door automatically, excite alarm device or visual alarm to show, perhaps similar signal.Step 1722 is a step 1724 later, in step 1724, produces a message and impels this front-end electronics to change video field signal by this sequence of catching from the corresponding video camera that is connected to this VR/PC unit.
Next step is a step 1726, in step 1726, judges whether this VR/PC unit is being used a pre-alarm buffer mode and is operating.If this VR/PC unit is being used a pre-alarm buffer mode and is operating, and is step 1728 after the step 1726 then.In step 1728, message of this alarm processor software part distribution, this message are to be used for indicating this video to preserve these data that software part is captured in this pre-alarm impact damper, and is described as front contact step 1334-1338 (Figure 52).This video hold function can be organized into like this, so that " forever " that all data in this pre-alarm impact damper are transferred on this hard disk preserved, perhaps this video hold function can be organized into like this, so that only shifted like this with the corresponding video data of particular camera field.
After the step 1728 step 1730 (do not operate if this VR/PC unit is used a pre-alarm buffer mode, just be located immediately at after the step 1726).In step 1730, this alarm timer is set (if perhaps alarm conditions in action, just expand this alarm timer), and this detected alarm event is added in the alarm event tabulation of being safeguarded by this alarm processor software part.
The alarm that Figure 76 illustrates according to a standard excites camera sequence figure, this camera sequence of realizing in step 1724.This sequence that provides in the MV200 simulation multiplexer that assignee of the present invention sells is provided this sequence shown in Figure 76, and (see nineteen ninety-five, RobotResearch company publishes Multivision Optima II TM Multiplexers, Installation and Operatin ManualThe 33rd page).In a sequence of the video field 1732 shown in Figure 76, this module 1734 that sign " A " arranged is with one or a plurality of are corresponding by one or more video camera produced, and this one or more video camera is pre that may to produce with respect to detected alarm be interested vision signal.In the module 1736,1738 and 1740 each represents that respectively from three be not a video field signal of catching the interested especially different cameras with respect to these alarm conditions.Like this, as in the represented in the above MV200 multiplexer, an alarm is responded, to these video camera rearrangements, so that by relevant with this alarm one field that video camera or a plurality of video camera produced with respect to other video cameras, be given more frequent record time slot.
Step 1730 is a step 1742 later.This records series shown in step 1742 expression Figure 76 is kept, up to this alarm timer expiry.About this last alarm whether overtime judgement having taken place has made in step 1744, overtime if this last alarm has taken place, just close this alarm timer (step 1746).
As long as it is over and done with to use the field rate of an increase to write down time of the video camera that this alarm is correlated with, shown in each Figure 76, the script holder of these video cameras records speed and just is reduced to these video cameras and is detecting the field rate that is prescribed before this alarm.Should be appreciated that, the record field rate of front defined may be " 0 " (promptly, the state of this video camera is only to have alarm to write down), perhaps use common sequence to carry out record, and other video cameras are in the normal recordings state, or the recording status of " time delays (time-lapse) ", in this state, use than a low field rate of speed that writes down video camera with normal sequence and write down this video camera.
If in step 1720, judge that the alarming processing pattern of a customization is effective, after step 1720, be step 1748 then.In step 1748, this alarm processor software part determines this video camera, the type of incident and about time of the alarm conditions that are detected.Be later on step 1749, in step 1749, the corresponding event response that decoded video camera, event type and time data are used to pick up from an event response descriptive data base 1746 is described.Step 1749 is a circulation later, and shown in step 1750, each order during this circulation is described the event response that is retrieved at this all is performed.This circulation is by step 1752, and 1754 and 1756 form.In step 1752, read with the current line in this description and order accordingly.In step 1754, to ordering a corresponding message to be encoded with this, in step 1756, this message is sent to the commander of this system software part.
Be an example that typical event response is described below:
Event response is described (example)
Alarm 1 output=ON (1)
Alarm 2 output=ON (2)
Video camera 1 speed=30 (3)
Video camera 1=ON (4)
Wait for=30 (5)
Restart (6)
Can observe, the exemplary event response that proposes is above described and is comprised six row.This first line display is alarmed 1 output signal and is set to out.For example, this may be one and excites for example light signal of a flicker of a visual warning indicator.This second line display, second alarm output signal is set to out.For example, this may can listen alarm indicator with handling one, for example a warning horn.
This third line is represented to catch record and is set to 30 of per seconds from the speed of the field of video camera 1.Remaining record bandwidth will be sorted between other video cameras that write down then in front distributes.
This fourth line represents that the recording status of video camera 1 is set to out " ON ".This order will cover the front, and any one uses software to close the order of video camera 1.
Four defined states of row in front during the 5th order expression is described by this response were kept 30 seconds.
The 6th row of this description is that last column represents that the mode of operation of this system front will restart after alarm response in 30 seconds.
The image processing facility
Referring now to Figure 77 this image processing facilities management person software part 1222 (Figure 45) are described.Beginning, in step 1758, the effect of this image processing facilities management person software part is that this user to this VR/PC unit provides the option that can use this user, handles shown image or image sequence by this system.Step 1758 is a step 1760 later, in step 1760, judges whether this user indicates the selection of having finished an image processing facility and its parameter.If the selection that this user does not have indication to finish this facility and parameter is handled, and is step 1762 then, in step 1762, judge whether this user indicates and cancelled a current selecteed facility and its parameter group.If this user does not also indicate and does like this, after step 1762, be step 1764 then.Step 1764 expression will be arrived step 1772 to a facility execution in step 1766 of being selected by this user.From following discussion, be appreciated that to this user can with this image processing facility option can use the form of a menu or as incompatible being provided of icon sets of representing one " tool box ".This image processing facility option conceivable, that be included in this VR/PC unit is an indentation option, a color adaptation option, a contrast adjustment option, a focus adjustment option, a histogram balance option or an object identification option.
Step 1766 must comprise the input that receives the parameter relevant with selecteed image processing facility from this user.This parameter may be included in this image plane, this selecteed facility will be employed or not be employed a zone or a plurality of zone.Other parameters may comprise that the degree of operation intensity of this selecteed facility or numeral or quantity control example are as a slider bar.For example, if selected a convergent-divergent facility, just may select the degree (2,3,4 times or the like) of convergent-divergent.
In step 1768, this parameter or a plurality of parameter selected by this user are translated into the unit relevant with this image plane, for example put the position.If suitable, they also may be translated into color or contrast or focus adjustment controlling value or similar.Then, in step 1770, this image processing facilities management person parts use this parameter value that is translated, and produce to indicate " preview " image of the possible effect of selecteed image processing facility to this user.Then, in step 1772, this preview image is shown.
Step 1772 is a step 1774 later, in step 1774, judges whether this user agrees with carrying out this selecteed facility and parameter.If this user agrees with carrying out this selecteed facility and parameter, is exactly later on step 1776, in step 1776, selecteed facility and parameter are applied to this image or image sequence.This cycle of treatment turns back to step 1760 then.But, if in step 1774, this user represents that this selecteed facility and parameter are provided with unsatisfactory, then should image or image sequence be restored to the condition (step 1778) of before this image processing facility is employed, be dominant (prevailing), and this cycle of treatment turns back to step 1760.Further, can see,, judge that this user has selected to cancel selecteed facility and parameter, just enter this step 1778 again if in step 1762.
Certainly,, judge this user and wish to finish interactive session with this image processing facilities management person, should just handle then and finish, and any image processing facility selecteed and that be not cancelled continues to work if in step 1750.
Graphical user interface
Figure 78 illustrates the operation of the graphical user interface part of this mainboard CPU software.As shown in the step 1778, a graphical user interface (GUI) engine explains by a user through a choice of location device, signal that mouse is imported for example, and produce and will be displayed on an object on the display monitor.In a preferred implementation of this VR/PC unit, this GUI engine 1778 uses the ability of well-known Windows95 operating system.In alternate embodiments, can imagine and use other GUI core, for example WindowsNT.The step 1780 to 1784 of Figure 78 illustrates the operation that this GUI engine 1778 is explained from the signal of this choice of location device.
In step 1780, this user's input receives and decodes.In step 1782, this decoded input signal data is translated into the message format of a standard, and in step 1784, these user interface software parts send a corresponding message or a plurality of message to the commander of this system.
The planning software parts
With reference now to Figure 79, describes and to relate to foundation, relate to the part of these planning software parts of a main options screen display especially.In step 1786, judge this user whether select to set up one vacation program operation.If just carry out and set up operation (step 1 788) this vacation.
In step 1790, judge whether this user wishes to set up the definition in " time on daytime " and " time at night " period.If an operation that was divided into day and night in 24 hours of forming one day just is performed (step 1792).
In step 1794, judge whether this user wishes to come a specific time block is carried out a planning function (for example, the time at night at workaday time on daytime or weekend).If just this selecteed time block is carried out a program operation (step 1796).
In step 1798, judge whether this user represents to have finished this program operation.If this user not expression has finished this program operation, this cycle of treatment turns back to step 1786.Otherwise, just finish this processing.
Figure 80 illustrates the processing in the module 1788 that is included in Figure 79.Setting up processing this vacation that illustrates among Figure 80 is with step 1802 beginning, in step 1802, provides a date screen display (seeing, for example Figure 152) to this user.
In step 1804, judge whether this user has selected a day from this date demonstration.If, just decoding this selecteed day (step 1806).Judge then whether this user wishes to be added in this tabulation vacation (step 1808) in selecteed sky.If this selecteed sky just is added to this tabulation vacation (step 1810).
In step 1812, judge whether this user wishes to cancel a decision that a selecteed sky is added to this tabulation vacation.If cancelled a selection, judge whether this user represents to have finished to set up session (step 1814) this vacation.If this session has been finished in expression, tabulation vacation that is produced during this session is used to substitute set vacation (step 1816) that exists previously, and finishes this planning session.Represent to have finished this session up to this user, this cycle of treatment is through step 1804,1808,1812 and 1814.If in step 1812, this user represents to wish to cancel the selection of having been done by this session, does not substitute original tabulation vacation that existed before this session vacation then with regard to finishing this session.
With reference now to Figure 81 and 82, describes by this processing that must comprise of the step 1792 among Figure 79.Shown in step 1818, control in two times that illustrate as Figure 81 that begin to be to use of this processing shown in Figure 82, in display box 1820, show current beginning (night) time and end (daytime) time.A upward arrow button (1822A), a following arrow button (1822B) and a time showing territory (1822C) are formed in each time control (1822,1824).Except these two controls, also have a cancel button (1826) and " finishing " button (1828).Use steady arm/cursor (1830) to handle this The whole control frame.After the demonstration of beginning, this processing just is in the undying circulation always, excites cancel button (1826) up to this user, shown in decision block 1832, perhaps excites this " to finish " button (1828), shown in decision block 1836.If this cursor (1830) excites this cancel button (1826), just finish this processing, and do not upgrade the distribute data as shown in processing block 1834.As excited this to finish button (1828), just read in the value in the display part (1822C) of this control and shown in processing block 1838 and 1842, upgrade this distribute data.If this cursor is used to select start time control (1822) and concluding time control (1824), can excite corresponding upward arrow button (1822A) or following arrow button (1822B) to increase or reduce this time by using this cursor (1830) then.With shown in the processing 1846 and 1850, the result of this increase or minimizing operation has produced a displayed value that is updated as decision block 1844 and 1848.
That Figure 83 illustrates is performed during planning is set up, allow to select this user pattern different from some to be selected for the processing of the pattern of planning.
In step 1854, judge whether this user has carried out planning to a file pattern.If this user has carried out planning to a file pattern, the time that is selected for the pre-planning archive operations just is set in the planning formation (step 1856) and picks up by this user parameter that select, this archive operations (step 1858).
In step 1860, judge whether this user has selected operation note pattern of planning.If be received (step 1862) and correlation time of this operator scheme be set in this planning formation by this this user-selected relevant operational parameter.
In step 1866, judge that whether this user plans that one or more video camera wants " off line ", does not promptly record (also being called " software is forbidden ").If just is set in this planning formation this correlation time (step 1868).
With reference now to Figure 84-86, continues further to discuss this processing shown in Figure 83.Especially, Figure 84 illustrates the details (" obtaining operating parameter ") of module 1862.Shown in the step 1870 of Figure 84, among Figure 84 illustrated in the explanation this processing be that each video camera that links to each other with this VR/PC unit is carried out a circulation.Step 1872 explanation for specific video camera, has been write down by this this user-selected logging mode.If selecteed pattern is " off line " (judging as step 1874), an expression representing the sort of effect then be used as an entry be inserted in a planning describe in (step 1876).
If in step 1874, find that selecteed pattern does not have off line, just obtain the selection (step 1880) of this user then to the spatial resolution parameter of the control data that will be recorded.Judge whether then and will write down this video camera according to an online mode of carrying out continuously or according to an event-driven mode.If online mode, it just is indicated in the suitable table entry (step 1876) then.
If in step 1880, judge and only use an event driven mode to write down this video camera, be exactly step 1882 then, in step 1882, judge whether this incident is to detect by external sensor or by being handled by the performed image analysis of this front-end electronics.If detect this incident by alarm sensor, be exactly step 1884 then, in step 1884, sign drives this alarm sensor or a plurality of sensor of record of the present invention.Otherwise, be step 1886 after the step 1882, in step 1886, select applied by this front-end electronics, be used to produce event driven this image analysis algorithm that writes down to particular camera.
Be step 1888 after the step 1886, in step 1888, receive the parameter of the execution that is used to retrain selecteed analytical algorithm.Looking different situation, after step 1888 or the step 1884, is again step 1876 perhaps, and in step 1876, these relevant data are used as an entry and are provided in a planning description or the table.
Figure 85 provides the example of the planning formation of a simplification.This planning formation is made up of entry 1890,1892 or the like.Each entry was begun by time and mode of operation head, was data or pointer that points to a relevant planning description or table of expression operation mode parameter then.Figure 86 has shown an example that planning is described or shown.In one first entry, shown in 1894, this system is shown as off line.In next entry, shown in 1896, planned an archive operations and represented and used a device that is designated as " peripheral driver 2 " to carry out this operation according to the mode of an iteration.One the 3rd entry is represented the time that this system starts working and has been comprised a description of the mode of operation that describes each video camera in detail.For example, 1902, expression video camera 1 will be used " in real time " mode and write down (not being free the delay), and use and the corresponding compression parameters of high quality image.So video camera 1 will be included in and use fixed intervals from wherein catch a camera sequence that writes down.In 1904, expression only when having detected incident by an alarm sensor, is just write down video camera 2.In 1906, expression only when when having detected incident by a performed motion video detection algorithm of front-end electronics, is just write down video camera 3.
In 1908, expression uses the speed of " time delay (a time lapse) " lower than the recording frequency of other video cameras to write down video camera 4, but the compression of high picture quality is arranged.
In 1910, expression uses real-time mode to write down video camera 5, but the image quality that video data compression provides is lower.The corresponding row (not showing) of describing also is provided for other video cameras (is 16 up to sum) that link to each other with this VR/PC unit.
Figure 87 is illustrated in processing performed when carrying out the pre-planning operator scheme that is stored in this system.In step 1912, judged whether the time of this planning of verification.For example, this timer may be every one minute just " stopping (go off) " once so that the change once planned with regard to verification of each minute.When this timer time then, just decode this current time (step 1914) and if a coupling arranged, just in this planning formation, search this time (step 1916).In to a performed cycle of treatment of each row in this planning formation (shown in step 1918), execution in step 1920-1924.In step 1920, it is capable to read this entry.In step 1922, set up a suitable command messages according to this entry is capable.Then, in step 1924, this message is sent to the commander of this system, carries out to transfer before carrying out by this relevant software part.
The machine analysis of video streaming content
With reference now to Figure 88-98, the software that is used to carry out image analysis algorithm is discussed.Ensuing discussion will be based on two aspects: by front-end electronics performed to the image analysis of " scene " video with by the video data after the match performed image analysis of this mainboard CPU to from this hard disk, regenerating.
Figure 88 represents all are provided at the summary of all identical aspect of image analysis algorithm in this VR/PC unit.1926 expressions relate to the aspect of this user interface, and the aspect that relates to this user interface can be divided into those aspects that are used to select an image analysis algorithm (1928) and those are used for a selecteed algorithm is provided with the aspect (1930) of parameter.This algorithm that module 1932 expressions are discussed is to be come " scene " video is carried out by this front-end electronics, is still come the vision signal that is reproduced is carried out by this mainboard CPU.Then, shown in 1934, the result of this algorithm is reported and/or is saved and/or taked specific action, and this output with this image analysis algorithm is relevant.
Figure 89 illustrate the selection that is included in an image analysis algorithm or " instrument " and set up in processing.Shown in step 1936, this user begins to be provided a screen display, and on this screen display, some instruments (analytical algorithm) can be selected by this user.For example, can be displayed on this screen with the corresponding corresponding icon of each available instrument.Perhaps, can show by name and come a menu that these available instruments are tabulated.In step 1939, judge whether this user has selected an instrument from this tool box then.If this user has selected an instrument from this tool box, just produce a new screen, perhaps use an expression to select the mode of this instrument to change selecteed screen (step 1940).Then, judge whether (step 1942) this user wishes to excite the use to this selecteed instrument.If this user does not wish to excite the use to this selecteed instrument, this cycle of treatment turns back to step 1936.But, if this user wishes to excite the use to this selecteed instrument, be exactly later on step 1944, in step 1944, this user represents that by video camera number or name this selecteed instrument will be applied to the video signal flow source on it.Be a circulation (shown in step 1946) then, this circulation is applied to each parameter that relates to selecteed instrument.This circulation is made up of to step 1952 step 1948.In step 1948, can be selected by this user, be displayed to this user about the option of this parameter.In step 1950, receive the input of this user about required option, in step 1952, be translated into the data that relate to this image analysis processing by the parameter setting that this user provided.
After this tool parameters circulation has been carried out each parameter, be exactly step 1954, in step 1954, the video whether this instrument that judgement is discussed will be applied to live video or be reproduced.If will use selecteed instrument to analyze live video, just produce a suitable command messages or command messages group (step 1956) then and be sent to this front-end electronics (step 1958) by this suitable command messages of the commander of this system or command messages group.On the other hand, if selecteed algorithm will be applied to the vision signal that is reproduced, just be written into the image analysis parts (step 1960) of this mainboard CPU then and represent that a suitable command messages of this selecteed algorithm and parameter is sent out (step 1962) through the commander of this system.
With reference now to Figure 90 A-90D, comes in the scope of " border intrusion instrument " example that characterising parameter is provided with.
Figure 155 is such one type screen display, and it can be provided for this user makes this user that the parameter of carrying out " border intrusion " image analysis tool can be set.
Forward Figure 90 A now to, in the step 1964 of a beginning, shown the drawing element (for example a frame or some line) of an expression by a border in a zone in the represented scene of this vision signal on a video image, the purpose of this image analysis algorithm is the entry that detected object enters into this appointed area.In this screen display of Figure 155, this parameter is represented by this block diagram shape element 1966.
With reference to figure 90A, step 1964 is a step 1968 later again.In step 1968, this user is allowed to use in essence with making in a traditional PC software is drawn bag and is used for dragging or mode that the mode of stretch a polygon or line is identical this boundary element 1966 that drags and/or stretch.When this user represents that having finished parameter is provided with (as detected in the step 1970), decide this line of this boundary element 1966 of expression or the end points (step 1972) of frame according to the related data tile (row and column) in this image plane space then.These end points are saved as the parameter of the position of this boundary element of expression then.
Another parameter that relates to this border instrument is the direction of passing this border.That is, may order this image analysis tool to detect and be on both direction, to pass this border, still only on a direction on two possible directions, pass this border.For this user is selected, can use an arrow (it can be the arrow of two statures, passes to be illustrated on any one direction) to represent through direction.Among Figure 155,1976 have represented unidirectional arrow.
Figure 90 B illustrates this processing that is used to select this through direction or direction that will be detected.In the step of a beginning, shown this through direction arrow 1976 (step 1978).Be step 1980 after the step 1978, in step 1980, judge whether this user indicates and changed through direction.If this user's indication has changed through direction, a through direction just is switched to next direction.For example, can use such circulation, for example go into, go out, twocouese goes into or the like to switch this through direction (step 1982).
As Figure 90 A, in this processing of Figure 90 B, also provide a step 1970, so that when this user represents that having finished parameter is provided with, this through direction parameter (step 1984) can be set.
Another relates to this border, and to invade parameter of testing tool be to find to pass through the size of the object on this border.For example, not given size or be not bigger that it is desirable ignoring tangible boundary crossings if seem to pass through the size of the object on this border.In order to stipulate this size of object, another drawing element frame (may be used contrast colors for example redness carry out shade) is shown, shown in 1986 among Figure 155.Like this, forward Figure 90 C to, this begins step 1988 must comprise the step that shows this dimension of object frame 1986.Step 1988 is a step 1990 later, in step 1990, judges whether this user represents to change this dimension of object frame.If this user represents to change this dimension of object frame, just the input from the user determines this new size setting (step 1992) and upgrades this size frame (step 1994) at this display screen.Again, Figure 90 C provides a step 1970 to judge whether to have finished this parameter setting.If finished this parameter setting, just decide final size setting (step 1996) according to the current state that is displayed on this size frame on this screen.Then will this final dimension of object setting translate into based on the unit (step 1998) of data tile and preserve that this is corresponding, based on the dimension of object parameter (step 2002) of data tile.
Another relates to this border, and to invade parameter of detection algorithm be " sensitivity ", and promptly one will be considered to represent to move rather than the signal contrast level of noise.Figure 90 D illustrates the setting of this sensitivity parameter, and the setting of this sensitivity parameter comprises a first step 2004, in step 2004, has shown a slider bar or similar picture element.This corresponding sensitivity slider bar element is represented by the label among Figure 155 2006.
Forward Figure 90 D again to, step 2004 is a step 2008 later.In step 2008, judge whether this user has represented that this sensitivity level changes.If just detect this new slider bar setting (step 2010).Shown in Figure 90 A-90C, come execution in step 1970 again, judge whether to have finished this parameter set handling.If, be exactly later on step 2012, in step 2012, the reading on this slider bar is translated into a vision signal contrast ratio, and the contrast proportion threshold value that this produced is used as this sensitivity parameter preservation (step 2014).
Figure 91 has shown a processing that is used to be written into this image analysis engine.Beginning (step 2016) is written into the equal practical parameter of any analysis tool.This parameter that is written in step 2016 for example may comprise, how expression translates into the GUI input information data of the data tile space and/or the velocity space.Then, in step 2018, the parameter that relates to speed determined, and the parameter that relates to speed is sent to velocity analysis parts (step 2020) of this image analysis engine.Decision relates to the parameter (step 2022) of detected motion vector cluster (clustering) then, and this parameter that relates to cluster is sent to cluster parts (step 2024) of this image analysis engine.Step 2024 is a step 2026 later, and in step 2026, decision relates to the parameter of the analysis of single video data data tile.The data tile analytical parameters that is produced then is sent to a data tile analysis part (step 2028) of this image analysis engine.Relate to the example that the parameter of testing tool is invaded on this border in order to provide, the parameter that relates to speed of this border instrument will be any detection of motion (module 2030).The parameter that relates to cluster of this border instrument is this object size frame, as (module 2032) measured with image plane data tile element.This related data tile analytical parameters of this border instrument is from this sensitivity this contrast ratio (module 2034) of coming out of deriving to be set.
Figure 92 illustrates by this front-end electronics a performed image analysis algorithm of live video data being carried out an initialized processing.At first, obtain one or more reference picture (preferably, being the live video image) (step 2036).Relevant parameter reference value is extracted (step 2038) from this reference picture.Step 2038 is a step 2040 later.In step 2040, the parameter reference value that this is extracted is stored in the storer of scene photo picture analysis module 846 (Figure 14), then, employed counter is set to zero (step 2042) in this image analysis engine, and a message is sent to the commander of this system parts of this mainboard software, represent this on-the site analysis engine preliminary work (step 2044).In step 2042, be set to this zero counter and be used to judge that when having enough data to use FFT handles.Invade on this border under the situation of testing tool, step 2038 preferably includes the average luminance of calculating along this boundary line, and a basic value is provided, and comes change detected (module 2046) according to this basic value.
With reference now to Figure 93 A-93E, describes and to be used to carry out the operation that this on-the site analysis module 846 of detection algorithm is invaded on this border.
At first, suppose that this on-the site analysis module uses in some traditional image sequence filtering algorithms of the data that produce the expression optical flow any one.In a preferred embodiment of the present invention, use time and space filtering to realize that the analytical algorithm that relates to motion is applied to being changed on the MCU mapping (enum) data sequence of changing in time, produces the estimated value of X-and Y-direction speed based on FFT.(this MCU mapping (enum) data that will be carried out fft analysis is and the identical data that are used to compress processing, or produces the data that are used for carrying out motion analysis especially according to those MCU " homogeny " criterions different with those criterions that are used to compress.) cluster analysis be applied to this velocity estimation, detects the rising edge and the negative edge of this moving object.The special example that relates to the algorithm of motion is that the check and analysis instrument is invaded on following this motion detection discussed and border.
Other analytical algorithms, example are " museum " and " optics " instrument as discussed below, must comprise that data tile of content with a current pictorial data field and a reference picture data fields connects a data tile and compares.Shown in Figure 93 A, obtain the data (step 2048) of expression velocity, and obtain the data (step 2052) of the cluster of this velocity of expression.In step 2054, obtain the data that relate in the analysis of the data tile of designated boundary.Then, shown in step 2056, judge whether to have sufficient time to finish this analysis.Finish this analysis if having insufficient time to, " analyzing an interrupted " message is sent to this front controller DSP1050 (step 2058).If suitable, this controller DSP may send an event report to this mainboard then, and then, this mainboard may be declared alarm conditions.
But,, be step 2060 after the step 2056 then if having sufficient time to finish this analysis.In step 2060, analyze this velocity data.If satisfy this velocity criterion (step 2062) that border of expression is invaded, just analyze this vector cluster data (step 2064) then.Then, judged whether to satisfy this cluster " object size " criterion (step 2066) that is used to find a border intrusion.If satisfied this cluster " object size " criterion that is used to find a border intrusion, be exactly later on step 2068, in step 2068, analyze the data of this expression data boundary tile itself.
Step 2068 is a step 2070 later.In step 2070, judge whether the analysis of this data boundary tile represents to have taken place a border intrusion.If representing to have taken place a border, the analysis of this data boundary tile invades, with regard to execution in step 2072.In step 2072, calculate this intrusion and detect a confidence factor of judging.Then, in step 2704, report generation and this confidence factor that this border is invaded to this front controller DSP1050.
On the other hand, then step 2058, perhaps in step 2062, any one finishes this border intrusion check and analysis when not judging in 2066 or 2070, and do not seek whether an intrusion has not taken place.
Figure 93 B illustrates by this controller DSP1050 " analyzing interrupted " message that is produced in step 2058 is responded the action of being taked.Shown in Figure 93 B, this controller DSP at first receives this and " analyzes " message (step 2076) interrupted, then a suitable message is formatd, to be received (step 2078) by this mainboard, and this message is connected the preceding commander of this system parts (step 2080) that forward this mainboard software to through the PCI between this front-end electronics and this mainboard.
Figure 93 C illustrates by this front controller DSP this message that is produced by this scene display analysis module in step 2074 is responded performed processing.Shown in Figure 93 C, this controller DSP receives this message (step 2082) of reporting the incident that is detected, then a suitable message is formatd, to report this incident (step 2084) to this mainboard, and this message is connected through above-mentioned PCI, send to the commander of this system (step 2086) of this mainboard software.
Figure 93 D uses general form to illustrate the judge module 2062,2066 of the processing among Figure 93 A and each this performed processing of 2070.The first step of Figure 93 D is this processing of Figure 93 D has all been carried out in an expression to each parameter a step 2087.In a step 2088, the threshold value than a higher level of this parameter is low to judge this parameter value.If then in step 2090, judge that this parameter value is whether than a more low level threshold value height of this parameter.
If the response to step 2088 and 2090 all is sure, the step 2092 of having arrived.In step 2092, an expression is set has satisfied the value of statistical indicant of this parameter rules, and return a confidence value.On the other hand, if in step 2088 and 2,090 one obtained a negative result, the step 2094 of having arrived.In step 2094, this is set indicates and represent that this parameter rules is not satisfied.
Figure 93 E illustrates the details (calculating confidence factor, Figure 93 A) of step 2072.In the processing shown in Figure 93, a confidence weighting repeated factor is applied to this speed confidence value (step 2096), a confidence weighting heavily is applied to this cluster and judges corresponding this confidence value (step 2098) then, and a confidence weighting heavily is applied to this data tile and handles confidence value (step 2102) then.In step 2104, this weight speed, cluster and data tile value are added, and obtain the confidence value of an integral body, and produce a message (step 2106) that comprises this addition result value.Figure 94 illustrates and is used for an analysis engine that is included in this mainboard CPU software is carried out initialization, an image analysis algorithm is applied to a processing of a video data stream that is reproduced.
In the shown first step of Figure 94, this first reference picture in this stream is found (step 2108) later at the point that will begin to analyze.Step 2108 is a step 2110 later, in step 2110, extracts and preserve this correlation parameter reference value.Then, in step 2112, this sign of this image analysis engine is initialized to suitable starting value, and in step 2114, and this analysis engine reports that to the commander of this system it has prepared to begin to carry out this analytical algorithm.
Figure 95 A-95C illustrates the operation performed to this video flowing that is reproduced of this image analysis engine.This image analysis engine that is used for analyzing the video data that is reproduced with above to get in touch described this live video analytical technology of Figure 93 A-93E similar.Represented as step 2116, this shown processing is a circulation that is applied to each the disparity map image field 1134 (Figure 44) in this video image field flow that is reproduced among Figure 95 A.Step 2118 is expressed as current disparity map image field, read reformed MCU mapping (enum) data, this mapping (enum) data is used as the input (step 2120) that a velocity analysis is handled then, and is used as the input (step 2122) that a cluster analysis is handled.Output according to from this velocity analysis and cluster analysis processing judges whether and need analyze (step 2124) to reformed MCU data tile data itself.If just carry out this analysis (step 2126) based on data tile data.Under any one situation, be step 2128 below, in step 21 28, estimate this image according to the result of this analyzing and processing.Judge whether to have satisfied this criterion (step 2130) then.If in step 2130, find that this criterion is not satisfied certainly, just do not seek then and whether detected interested image feature.Certainly being satisfied if find this criterion, is a step 2132 after the step 2130 then.In step 2132, detected interested feature and a confidence level is reported to the commander of this system.But, if not being high enough to report from this confidence level that this analysis produced has detected interested feature, and do not have lowly to the appearance of cancelling this feature certainly yet, further processing just takes place, shown in step 2134.
Figure 95 B illustrates the further details of step 2126 among Figure 95 A.Shown in Figure 95 B, reformed MCU data are read (step 2136), are used traditional jpeg decompression then and contract and handle decode (step 2138).
Need to judge whether then further to analyze these coefficient of frequency data (step 2140), if just this DCT coefficient is analyzed (step 2142).Then, in step 2144, need to judge whether point data (time domain data), if just handle and obtain this point data, and analyze this point data (step 2146) by carrying out transformation by reciprocal direction.
Figure 95 C illustrates this processing shown in the step 2134 among Figure 95 A.The first step of Figure 95 C is a step 2148, wherein the current set of the velocity that will from the optical flow analysis, be produced with compare according to the historical record of the velocity that image produced of front in this image sequence.
Be step 2150 after the step 2148, in step 2150, whether the analysis of judging the velocity historical record represents that this speed passed through this image analysis algorithm and wished a velocity vector value detecting.If just will be compared (step 2152) then by the group of picture that this optical flow is analyzed detected current object and front.In step 2154, judged whether to occur the object of the type that will be detected by this analytical algorithm then.If be exactly later on step 2156.In step 2156, carry out a historical analysis of data tile data, then in step 2158, judging current and pass by whether detected data tile feature represented to have occurred will be by the feature of the data tile that this current algorithm detected.If step 2158 has been made a sure judgement, then with regard to execution in step 2160.In step 2160, be that this algorithm output produces a confidence factor, and a judgement with " may be " is set indicates accordingly.Then, in step 2162, this that produces from each analysis portion branch of this algorithm puts that letter data is combined and weighting, just calculates a confidence factor (step 2164) that is conditioned below.Be step 2166 below, in step 2166, a suitable message that comprises " may a be " result is before forwarded to the commander of this system.
If in step 2150, any one makes a negative judgement in 2154 and 2158, and this processing of Figure 95 C is returned one do not occur will be by the conclusion (step 2168) of this feature that this algorithm detected then.
With reference now to Figure 96 and 97, a specific example of the operation of an image analysis algorithm that is applied to the video data that is reproduced is described.For this example, supposed that this analytical algorithm that will be employed is a type above-mentioned, that detect one " border " intrusion.Figure 96 has schematically shown an image sequence that is produced by a video camera (not showing), and this video camera provides a view that cash cassette obtained of seeing from above in the zone 2172 that is placed on a part sealing.
Figure drawing element 2174 an expressions border that distributed by a user, that be used to carry out a border intrusion analytical algorithm.Shape 2176-1 represents that to 2176-6 a motion object that is detected by this analytical algorithm is placed in the respective fields of analyzed sequence of video signals.Forward Figure 97 to, step 2178 expression of a beginning is to the just group of each in the analyzed sequence of video signals that is reproduced difference field execution processing subsequently.In order effectively to handle, when being applied to the video flowing that is reproduced, " reference " image is somebody's turn to do in the ignorance of the preferred implementation of this based drive image analysis algorithm, and as previously described, reformed MCU mapping (enum) data is operated.As a result, one " hole " or " seam " are arranged in the image sequence of each reference picture, but find, if in each interval between reference picture suitable, a large amount of difference images are arranged, it can't produce serious defective.Certainly, a large amount of difference images also produces a higher compression factor, and should under the situation of reverse regeneration image quality and convenience be traded off.As previously mentioned, this VR/PC unit preferred implementation produces 32 difference fields between every pair of reference field.
Continuation is with reference to Figure 97, this processing that is applied to every group of difference field comprises a step 2180, in step 2180, produce a reformed MCU mapping (enum) data array, and in step 2182, by using FFT (rapid fourier change) to handle or similarly handling, come from this MCU mapping (enum) data array, to form a velocity profile group (velocityprofile set).In step 2184, just constitute the velocity cluster of object then, and as Figure 95 A, each possible combination of analytic target and velocity, produce a "Yes", the judgement (step 2186) of "No" or " may be " about each object/velocity group.Step 2186 is a judge module 2188 later.In step 2188, it may judge that each object/velocity group has been marked as "No", and in this case, the back is a step 2190.Step 2190 expression finishes this analyzing and processing, and does not detect any intrusion to this border.
Each object/velocity group has been marked as "Yes" (represented as step 2192), and promptly each satisfies object/velocity group that this criterion on this border is invaded in expression, judges whether to satisfy this sensitivity (brightness) criterion in step 2194.If satisfy this sensitivity (brightness) criterion, just produce a confidence value (step 2196), and with this confidence level, and represent that the value of the output of this object and velocity analysis produces (step 2198) as output.Be marked as " may be " (that is, representing the intrusion on a border ambiguously, the intrusion on also indefinite expression neither one border) for each object/velocity group, just carry out a processing that starts from step 2202.Beginning produces the batten (step 2204) based on this speed history.An example of this batten is shown as the dotted line 2206 among Figure 96.Then, just a flex point of decision (step 2210) this batten (2208 expressions of Figure 96) judges then whether this flex point of this batten has passed through this border (step 2212).As can be seen, in the example shown in, this flex point 2208 of this batten 2206 has been passed through this border 2174 really therein from Figure 96.
In the judgement of step 2212 is under the situation of sure judgement, as previously described, and with regard to execution in step 2196 and 2198.In step 2212, perhaps in 2194 under any one situation that judges whether fixed judgement, with regard to execution in step 2190 (finish this processing, and do not detect intrusion).
Figure 98 illustrates after certain alarm event has taken place, and impels a preselected image analysis algorithm to be applied to the treatment step of a predetermined field video signal stream.In Figure 98, step 2214 expression has received by an external alarm sensor device or an alarm signal that is provided similarly has been provided.Step 2216 is judge modules, and specific image analysis algorithm of step 2216 expression is applied to a live video image stream that is produced from a video camera Y according to a continuous mode.The generation of step 2214 or step 2216 are sure judgements (detecting image feature), with regard to execution in step 2218.In step 2218, according to the parameter of having been preserved in advance (for example position of a boundary line), this preselected image analysis algorithm is applied to a live video image stream that is produced from a video camera Z.This performed in the step 2218 image analysis algorithm is possible, but not necessary, and is similar with applied this image analysis algorithm of step 2216.In addition, video camera Z can be counted as identical with video camera Y.In other words, using one first analytical algorithm to detect in the live video streams an interested feature may automatically cause one second analytical algorithm is applied to this identical live video streams.
The video data compression technology that substitutes
Figure 99 and 100 illustrates the processing that variation is carried out of being somebody's turn to do " stamp " compress technique that has been described according to the front.In essence, in the variation of Figure 99 and 100, be not " identical " MCU that in preservation, ignores simply in the difference image, but before JPEG coding, this difference image should " identical " MCU part can be converted into a complete black condition.
Forward Figure 99 to, an initialization step is performed (step 2220).Step 2220 is a step 2222 later, in step 2222, and next video data field that reception will be decoded.Step 2222 is a step 2224 later, and in step 2224, the lowest order of each point value is forced to get one and the inconsistent value of a stain value.Whether this is being a reference field in processed field just to judge (step 2226) below.If this is being a reference field in processed field just, with regard to execution in step 2228.In step 2228, all pieces are carried out JPEG compression (as the same in the described compress technique in front), and upgrade these reference statistical data that are used to make " homogeny " judgement.This cycle of treatment turns back to step 2222 then.
If in step 2226, this that discovery will be processed is a difference field, just picks up next data tile or MCU (step 2230) then, and calculated characteristics statistics (step 2232).The calculating of this statistics can be used verification plank Sampling techniques described above or other suitable technique, comprises the diagonal line sampling in four/part of data tile or data tile.
Be step 2234 after the step 2232, in step 2234, the feature and the fixed reference feature of this calculating compared, and in step 2236, make the judgement of " homogeny ".If in step 2236, find this current MCU or data tile and reference data tile or MCU " identical ", then all form the value (step 2238) that just is configured to represent a solid black color in MCU or the form arranged side by side at the point of processed MCU or data tile.
Step 2238 is a step 2240,2242 and 2246 later, and their expressions comprise the part of the JPEG coding techniques of orthogonal transformation, quantification and variable-length encoding.So the data tile of blackening or piece are carried out the JPEG coding.But,, so just this data tile or MCU are carried out the JPEG coding step, and at first do not make this data tile or MCU blackening if find that in step 2236 processed data tile or MCU are different with the reference statistical data.
Judge module shown in the step 2246 illustrates that this cycle of treatment turns back to step 2230, and up to treated last data tile or MCU or difference field, in this point, this cycle of treatment turns back to step 2222.Be displayed among Figure 100 with " black filling " corresponding decompression technique of stamp decompression technique of Figure 99.The step that begins among Figure 100 is a step 2248, step 2248 expression now whether need the to decode judgement of a reference field.If just all the data tiles in this are carried out jpeg decompression and contract, and use decoded data tile to upgrade and whole image plane corresponding display buffer position (step 2250).This cycle of treatment turns back to step 2248 then.
If in step 2248, find to decode now a difference field, then each data tile in this difference field or MCU are carried out a cycle of treatment (shown in step 2252).First step in this circulation is a step 2254, in step 2254 specific data tile or MCU is carried out jpeg decompression and contracts.Judge in step 2256 then whether whole data tile or MCU deceive.If, be exactly step 2258 below, in step 2258, do not upgrade and this data tile or the corresponding display buffer of MCU position.But, if it is not complete black finding this data tile or MCU in step 2256, just finished the processing (step 2260) of contracting of jpeg decompression to this piece then, and use with this piece accordingly " stamp " upgrade the display buffer (step 2262) of this image plane relevant position.
The decompression technique of Figure 100 can be summarized as: in the variance data field, black stamp is thrown away, and Hei stamp is not used to upgrade this image plane.Can see that getting in touch Figure 99 just now compares with for example getting in touch described initial " stamp " compress technique of Figure 44 with 100 described technology, lacks reformed MCU mapping (enum) data 1134.Like this, Figure 99 and 100 shown data layout and treatment technologies are simpler to a certain extent, but lack motion detection and back to the advantage that decompresses and brought by what this mapping (enum) data provided.In addition, " black filling " stamp technique provides a lower compression factor.In addition, the dynamic range of the compressed video data that technology provided of Figure 99 is littler than pure " stamp " technology, and this is because do not allow a compressed video data point to have a real black level value.
At user option demonstration/save options
With reference now to Figure 101 A and 101B, the processing of this user's selection of permission about the room and time resolution options of the preservation of incoming video signal stream described.
At first, will describe of this processing with reference to figure 101A and set up part.In step 2264, provide the room and time resolution options to this user.In step 2266, judge that whether this user is to having selected an off-gauge temporal resolution by a given video signal flow that video camera produced.The example of an off-gauge temporal resolution is " time delay " logging mode, should write down the field rate (temporal resolution) of this corresponding inlet flow than using standard resolution to come the field rate of recorded stream low by " time delay " logging mode.If the judgement in step 2266 is a sure judgement, send a suitable message (step 2268) so that the video camera stream of being considered is realized this off-gauge temporal resolution to this front-end electronics then.
Judge whether a given video camera stream has been asked an off-gauge spatial resolution in step 2270, the example of off-gauge spatial resolution is to use than system standard 240 horizontal line more or each video field of the video camera stream representing to be considered of row still less.If the judgement in step 2270 is a sure judgement, preserves suitable message of software part transmission to video then and realize selecteed off-gauge spatial resolution (standard 2276).
Figure 101 B represents to be performed the processing that realizes getting in touch the resolution options that Figure 101 A discussed.Especially, the shown processing list of Figure 101 B be shown in Figure 52 (" video preservation " software part) module 1306 (" pre-alarm? ") the step that entered when being a negative judgement of judgement.In the processing of Figure 101 B, at first judge whether to have received a non-standard resolution message (step 2274) about the field that is being saved.If not so, just directly enter the step 1308 of Figure 52 from the step 2274 of Figure 101 B.But, if the step in the step 2274 is a sure judgement, then for example, the line to be replaced of this that is saved is ignored (step 2276) from preserve, and produces the index data (step 2278) of data that expression is saved off-gauge time or spatial resolution then.
Figure 102 represents to be performed the processing that a parameter relevant with the compression of input signal stream is set.The shown first step of Figure 102 is a step 2280, and a screen display is provided in step 2280, provides the parameter relevant with squeeze operation that option is set to this user.This option is usually at image quality with effectively use between the preservation capacity of VR/PC unit and trade off.This option may comprise the useful bit rate of quantization operation that is provided with in the JPEG chip 848 (Figure 14), be increased in the number of each the difference field at interval between the reference field, and/or be adjusted in a used data tile same threshold of data tile that judges whether to keep or abandon in the difference field.Should be appreciated that this option that is provided for this user should be can carry out different settings at different cameras.
Judged whether to select the non-standard interval between the reference field in the step 2282.In addition, in step 2284, judge whether to have selected an off-gauge same threshold.Any one judgement in step 2282 and 2284 is under the sure situation, to be later on step 2286, in step 2286, sends a suitable message that realizes this off-gauge compression parameters to this front-end electronics.Although be not displayed among Figure 102, change has taken place if this user represents the quantization bit rate, the message about this influence also is sent to this front-end electronics.
The application of image analysis tool in distributed I VIM system
With reference now to Figure 103 A, describe to 103C and to relate to from a remote location, for example this locality or host node are to the processing of this VR/PC unit downloading images flow analysis algorithm.
The shown processing of Figure 103 A is from this external device (ED), and for example the angle of this host node or this remote node provides.In step 2286, judge whether a long-range VR/PC unit has asked to send to this VR/PC unit (step 2286) with the corresponding program of image analysis algorithm from this host node or local node with one.If, be exactly step 2288 below, in step 2288, judge whether will to be used for retraining the parameter that this algorithm is carried out with in this request VR/PC unit, being provided with.Step 2290 is a step 2292 later, if perhaps this parameter is not set in this host node or the local node, step 2288 is exactly a step 2292 later.Be downloaded to the VR/PC unit of this request in the step 2292 with corresponding program of this requested analytical algorithm and the parameter (depending on the circumstances) that is set up.
Figure 103 B provides the additional detail that relates to step 2290 among Figure 103 A.Especially, in the step 2294 of Figure 103 B, represent that a reference picture is uploaded to host node or the local node that this algorithm parameter will be set up therein by the VR/PC unit from this request.This host node or local node use this reference picture that is updated that this algorithm parameter (step 2296) is set then.
Figure 103 C has illustrated when a host node or local node carry out work to come when two or more initiate search based on analysis tool in away from the VR/PC unit of this node apparatus the processing that is taken place in this node.Step 2302 is the shown first steps of Figure 103 C.In step 2302, this node apparatus is initiated and communicating by letter of carrying out between a VR/PC unit of one first remote location, impels a reference picture to be uploaded to remote-control device by a VR/PC unit.In this node apparatus, select an image analysis algorithm then, and the reference picture that uses this quilt to upload is provided with the parameter (step 2304) of the execution of this algorithm of constraint.Step 2304 is a step 2306 later, and in step 2306, time parameter (start and end time) is prescribed the time of representing the performed consequential database search of the video data in the VR/PC unit.Step 2306 is a step 2308 later.In the step 2308, a data message (perhaps clearer and more definite a series of data-messages) is sent to this first long-range VR/PC unit and downloaded this selecteed analytical algorithm, the parameter that is set up and interested time range this node apparatus from this node apparatus.This time be engraved in the search that begins to carry out video database in the VR/PC unit based on algorithm.
Be that step 2310 is to step 2316 after the step 2308, except step 2310 is carried out the 2nd VR/PC unit that is positioned at a position different with the position of a VR/PC unit to step 2316, step 2310 is identical to step 2308 with step 2302 to step 2316.Identical with the front, a reference picture is uploaded (step 2310), in the selection of this node apparatus execution analysis algorithm and the setting of parameter, the setting of the time range of also being correlated with (step 2312 and 2314), and message is sent to the 2nd VR/PC unit and downloads this selecteed algorithm, and, excite in the performed video database search (step 2316) in the 2nd VR/PC unit with the parameter that comprises time parameter.Final step among Figure 103 C is 2318, and this node apparatus waits for that the VR/PC unit is reported in the result of the performed corresponding search in each VR/PC unit in step 2318.In this distributed I VIM system, another example that can design is that a dynamic video pictorial data sequence is uploaded to a host node or local node from a VR/PC unit, an image analysis algorithm is applied to the image sequence of being uploaded in this node.Can see that need very high transmission bandwidth and/or a large amount of time because this video data is sent to this node from this VR/PC unit, this example is always undesirable.
Figure 104 is illustrated in the processing of the interested video information of automatic transmission that is produced when an image analysis algorithm that is applied to an on-the-spot input video stream detects an interested feature.Suppose that the processing among Figure 104 is performed in a VR/PC unit.Beginning in the processing of Figure 104, judges whether an analytical algorithm being applied to an inlet flow that is produced by video camera X detects this algorithm and wish the feature (step 2320) that detects.If this VR/PC cell operation is come automatically and will be sent to an external device (ED) from the video data field of input video camera X signal, for example this locality or host node device (step 2322).
Intelligent response to incident
Figure 105 illustrated in a VR/PC unit, be performed, come to change the processing of a camera record sequence when an on-the-spot input video image being flow to row image analyzing and testing to interested feature.
The beginning step of Figure 105 is a step 2324, in step 2324, judges whether to detect an interested feature in the incoming video signal stream from one first video camera.When detecting such feature, be exactly later on step 2326, in step 2326, different with first a video camera and video camera that also be not included in the records series till now is added in this records series.For example, two video cameras may produce the view of observing identical zone from different angles.First video camera may use a motion detection analytical algorithm that is applied to from the inlet flow of this first video camera to carry out work normally.When detecting motion, second video camera that normally is in " off line " (not writing down) is added to this records series, so that catch this motion event from two angles.Although not expression among Figure 105, the response of step 2326 also comprise that permanent preservation is produced by second video camera and appear at a vision signal in the pre-alarm impact damper as the piece 1726 and 1728 with Figure 75.
The feature that Figure 106 has illustrated in software to be realized, by this software, image analysis algorithm detects an image feature to be impelled and excites an other image analysis algorithm.Among Figure 106 one beginning module 2328 is identical with module 2324 among Figure 105.Wishing detected feature if detect it by module 2328 these represented image analysis algorithms, is step 2330 after the step 2328 then.In step 2330, a predetermined image analysis algorithm and preserve parameter in advance and from storer, retrieved.Step 2330 is a step 2332 later.In step 2332, this analytical algorithm that is retrieved is sent to this front-end electronics, to be applied to by the input video that identical video camera the was produced image stream that uses this algorithm in 2328, mentioned to monitor, perhaps another input video stream, perhaps two video flowings.This additional algorithm that in step 2330, is retrieved can be with step 2328 in applied algorithm have identical type, perhaps different in essence.
Figure 107 has illustrated to impel and has responded to detecting alarm conditions, changes the processing of video data compression parameter.From Figure 108 as can be seen, at first in step 2334, judge whether to detect alarm conditions.Can be by input from an alarm sensor device, by using an image analysis algorithm to an analysis that live video streams is performed, perhaps a user by this system imports a signal and declares that alarm conditions detect this alarm conditions.Under any situation, when detecting alarm conditions, step 2334 is a step 2336 later.In step 2336, description of retrieval from storer, a video camera or a plurality of video camera relevant with the alarm conditions that are detected are specified in this description, and represent the compression algorithm parameter that is updated that will be employed because of the alarm conditions that are detected.Typically, the compression parameters that excites of this alarm can be a kind of parameter that can improve image quality.Like this, with the available bit number of quantification that increases in the JPEG encoding process, to reduce the number of each the difference image at interval between the reference picture, and/or reduce and be used for " identical " threshold value that difference field data tile and reference data tile are compared.
Step 2336 is a step 2338 later.In step 2338, the compression parameters that this that is retrieved in step 2336 is updated and be sent to front-end electronics according to the order that the compression parameters that this is updated compresses appointed camera video stream.
How Figure 108 has illustrated that an image analysis algorithm detected image feature that use is applied to an on-the-spot input video stream responds, and changes the example of the apparent field of a video camera.Suppose among Figure 108 that as previously described, a border is invaded detection algorithm and is applied to the live video streams that video camera produced that is designated as video camera X by.In step 2340, judge whether to have detected a border and invade.If be exactly later on step 2342.In step 2342, carry out a zoom operations by handling from the input video stream of video camera X, so that be exaggerated with the corresponding part in appointed border in this image plane.Make in such a way, it may be most important information that a major part of image plane can be used in the image stream, promptly on appointed border or the borderline feature adjacent with appointed border.
Figure 109 has illustrated that an image analysis algorithm that is applied to an on-the-spot input video stream is detected an image feature to respond, and changes the example of the visual field of a video camera automatically.Shown beginning step 2344 is identical with step 2324 among Figure 105 among Figure 109.If in step 2344, find to have detected interested feature, then with regard to execution in step 2346.In step 2346, produce a predetermined alarm output and/or produce a signal and cause and in step 2344, monitor from a predetermined motion of the video camera of wherein video flowing and/or the predetermined motion of a different video camera.Can use an intended target algorithm to carry out a predetermined camera motion or a plurality of motion, as U.S. Patent number 5,526,041 is described.
Figure 110 has illustrated two different characteristics that detect an input video stream are responded and has taked the processing of an athletic performance or a plurality of actions.By using, judge in an input video image stream one first feature whether occurred in the step 2348 with first image analysis algorithm.If in step 2348, judge first analytical algorithm and detected this predetermined characteristic, be exactly step 2350 then, in step 2350, in identical input video stream, whether detect one second predetermined characteristic by using one second analytical algorithm to judge.If be exactly later on step 2352.Shown in Figure 110, only when step 2348 and 2350 has all been made a sure judgement, just enter step 2352.
One or more performance in the step 2352 expression following actions: the displayed image (for example, by increasing vertical resolution or temporal resolution) that changes input video stream; Select wherein will preserve a preservation medium (for example, to an external device (ED), this locality of example or host node send the data of this video flowing of expression) of an input video stream; Send a signal and cause the video camera of this input video stream of generation or an automatic target search or other motions of a different video camera; And/or select an additional image analysis algorithm that will be applied to the input video stream of different video stream.The shown processing of Figure 110 can be considered to use a logical combination of two analysis tools, particularly in this case, and one and combination of two instruments.Other logical combinations that further can the design analysis instrument also can be used for detecting an event condition, perhaps trigger a predetermined response, or the like.For example, when in " click " two or more instruments any one is detected, just lay claim to an event condition.This can constitute one or combination of two or more instruments.(another understanding of sentence from the front also can design the logical combination more than two analysis tools.) in addition, can also adopt other logical operators, for example or non-, with non-, XOR, or the like and comprise the complex logic combination of two or more operational symbols.
Can provide a screen display (not showing) to allow required analysis tool logical combination of structure from the selection of instrument or logical operator to this user.The video camera that this user also can be allowed to select this analysis tool will be applied on it flows or a plurality of stream, and is allowed to be provided with the parameter relevant with this selecteed instrument.
Figure 111 has illustrated a processing, in this is handled, when the video camera that produces the video flowing that will be employed this algorithm during in motion; Perhaps when alarm conditions relevant, forbid using an on-the-spot image analysis algorithm with this video camera.In Figure 111, step 2354 expression judges whether to have occurred alarm conditions relevant with a particular camera (being designated as video camera X).These alarm conditions can be passed through an alarm sensor device, and are detected by the scene photo picture analysis, perhaps import to be initiated by the user.Whether step 2356 expression judges video camera X in motion, and this judgement can be based on whether receive a camera motion signal from this camera control unit 528 (Fig. 1).
Continuation if any one in step 2354 and 2356 made a sure judgement, just enters step 2358 with reference to Figure 111.In step 2358, an appropriate information is sent to this front-end electronics, forbids using an image analysis algorithm to the live video streams that is produced by video camera X.In the place that detects alarm conditions relevant with video camera X, forbid video camera X video flowing is carried out on-the site analysis, be favourable to avoid unsuitable prolongation alarm conditions.In addition, when video camera X in when motion, forbidding that it is desirable that video camera X video flowing is carried out on-the site analysis, may be impossible because produce about the significant analysis by the video flowing that video camera produced that is moving, and this essence with image analysis algorithm is relevant.Figure 112 has illustrated and has been designed to handle another feature that a video camera that moves is producing the situation of relevant video flowing.Among Figure 112, first step is 2360, and step 2360 is identical with the step 2356 of Figure 111.If in step 2360, judge a video camera X and move, be exactly step 2362 then.In step 2362, message is sent to this front-end electronics and adds an expression (for example, in the head of the video data field of video camera X video flowing a bit value being set) and illustrate that the video data field of being considered is to produce during in motion at video camera.In regeneration during, write down image analysis algorithm after can using these data to forbid one, perhaps as being carried out a prompting of special processing by an image analysis algorithm by a motion cameras institute video captured data fields.Substitute as one of step 2362, in a data should " being in the motion " prompting can during recording processing, be added to mainboard CPU, and/or be added to video flowing is recorded in hard disk during in the index data that produced.
Figure 112 A has illustrated and has used this video camera to forbid using subsequently an image analysis algorithm in the prompting of motion.The first step that Figure 112 A shows is a step 2364.In step 2364, the parameter setting that has judged whether to select an image analysis algorithm and be used for this algorithm application is flowed to a video image that is reproduced from this hard disk.If, then from this hard disk the retrieval will be analyzed this video flowing (step 2366).Step 2366 back is a step 2368.In step 2368, judge that a data that is accompanied by the video data field that is reproduced are illustrated in when producing the video data field, whether this video camera that produces the video data field is in motion.If step 2368 is a step 2370 later.In step 2370, forbid the execution of selecteed image analysis algorithm, be created in during the selecteed video flowing video camera then at a message report (step 2372) of motion.On the other hand, if in step 2368, find the motion sign,, and report result's (step 2374 and 2376) of this algorithm according to the disclosed common practice in front then with regard to the application image analytical algorithm.
Should be appreciated that a movement representation is included in index data rather than the stature data, in this case, step 2368 must check whether this index data has movement representation.
Figure 113 has illustrated a processing, wherein when detecting alarm conditions, increases an overall rate of catching and preserving the video data field.
Among Figure 113, step 2378 judges whether to detect alarm conditions.The detection of these alarm conditions can be based on a signal that is received from an alarm sensor device, or by an image analysis algorithm being applied to live video image stream, or excite an alarm signal by a user of this system.If in step 2378, find alarm conditions to have occurred, be exactly step 2380 then.In step 2380, this VR/PC unit, particularly the front end electronic section on it are by one second pattern of coming the capture video data field to use one higher comprehensive speed from one first mode switch of using a first rate to catch and preserve the video data field.For example, can design when not having alarm conditions, this VR/PC unit uses the speed of 30 of per seconds to catch and preserve.For example, the comprehensive speed of 30 of this per seconds can be distributed to each video camera that links to each other with this VR/PC unit successively.But, when detecting alarm conditions, catch speed and be added to 45 of per seconds for this comprehensive.Can catch and preserve speed according to what the sequence identical with the front distributed 45 of this per seconds, perhaps preferred, 15 of per seconds or 30 s' speed can be assigned to one or two interested especially video cameras, and the script holder with remainder records time slot allocation to a series of other video cameras then.By a sequence that video camera produced that alarm arranged by the series of fields weave in of being discussed with the top Figure 76 of contact that other video cameras produced.
Figure 114 has illustrated and respectively different compression parameters has been applied to from the step of the different input video streams that video camera received.Step 2382 is the shown first steps of Figure 114, and it is represented when each video camera is chosen in the camera record sequence that will be recorded, to each video camera execution step subsequently
Rapid 2384 and 2386.In step 2384, the compression algorithm parameter of the next video camera that this processing retrieval will be recorded, the difference of this compression algorithm parameter decision between data tile of a difference video data field and reference field to what extent can be considered to this difference field data tile and this reference field data tile does not have " difference ".In step 2386, use the compression parameters of in step 2384, being retrieved that this compression algorithm is applied to this difference field data.
Figure 115 represents a processing, and the different mode of mode of using the detection of alarm conditions of this processing to impel a display buffer to be used when not having alarm conditions is upgraded.First step among Figure 115 is a step 2388, and this step is identical with the step 2378 of Figure 113.Alarm conditions occurred if step 2388 is judged, upgraded this display buffer then so that upgrade whole display buffer (step 2390) at interval in each demonstration.On the other hand, if do not find that alarm conditions are with regard to execution in step 2392.In step 2392,, only upgrade a part of display buffer in each display cycle.For example, the technology that can use an interlacing to upgrade, upgrade with even number line in one first display cycle thus and put corresponding display buffer position, upgrade remaining position (odd-numbered line) in the next display cycle then, and carried out the alternately renewal of this even number and odd-numbered line display buffer position the display cycle subsequently.
The VR/PC unit is carried out major function simultaneously
Figure 116 illustrates that the VR/PC unit solves performed record in this VR/PC unit, the processing of the conflict between playback and the file function.At record, playback and between the grandfather cycle quantity of handled data be dynamically to change.For example, exist a time neither to carry out playback, also do not carry out file.Perhaps will be by playback, perhaps the quantity of the video information that will be filed changes.In addition, the captive comprehensive speed in the speed of the video data that generation will be write down and video field, to the obtainable compression degree of input video stream, this user-selected resolution is relevant with other factors with compression parameters and can change.
VR/PC disclosed herein unit is used for writing down simultaneously, playback and file exist.As used herein, term " simultaneously " should upward and from virtual angle be understood from literal.The record that carries out simultaneously on literal as one and the example of playback operation, can consider a kind of like this situation, video display monitor is showing a signal of a video image stream that is reproduced from this hard disk, and, be written in this hard disk with the video data that video camera produced that this VR/PC unit links to each other by one or more in the identical moment of strictness." virtual " can realize with the multi-job operation of this mainboard CPU simultaneously, and by this operation, independently record, playback and file control thread are simultaneously in operation.The person skilled in art is appreciated that simultaneously logout, playback and file thread comprise that the processing cycle of being carried out by this CPU carries out time division multiplex and realize record, playback and file thread in other operations.By this multi-job operation, in the very short time, for example, in one second the sub-fraction, it is all corresponding movable to repeat with record, playback and these three of file threads.
Certainly, be limited a preset time with interior number by the performed processing cycle of this mainboard CPU.In addition, data are written to or also are limited from the speed that this hard disk reads.So there is a kind of situation, write down simultaneously, playback and archive operations limited by the access rate of the processing power of CPU or hard disk.Figure 116 has illustrated how this VR/PC unit manages a preferred implementation of this restriction.
The first step of Figure 116 is a step 2394, in step 2394, with regard to while logout, playback and archive operations.Step 2394 is a step 2396 later, in step 2396, judges whether record, playback and the file thread of operation have caused overload to this processor (and/or hard disk access bandwidth) simultaneously by three.If no, this processing circulates simply and turns back to step 2394.But, transshipped if in step 2396, find this processing power, be step 2398 then, in step 2398, only logout and playback operation simultaneously.Like this, any one priority is low in the priority ratio of file record and the playback.Step 2398 is a step 2402 later.In step 2402, judge the record of operation simultaneously and the overload of the ability (and/or hard disk access bandwidth) whether playback operation has caused this processor.If no, judge whether in step 2404 that then enough processing poweies and hard disk bandwidth support file and record and playback.If this processing turns back to step 2394.Otherwise this processing turns back to step 2398.
Caused this processor overload or surpassed the ability that this hard disk is visited that this playback operation just is stopped then, only executive logging operation (step 2406) if in step 2402, find this record and playback.Like this, as can be seen, the priority ratio playback of record and the priority of file are all high.
Step 2406 is a step 2408 later.In step 2408, judge whether this processor has the not use capacity of a predetermined quantity, if having, is exactly step 2398, and logout and playback simultaneously again.But, if the not use capacity of predetermined quantity do not occur, the recording operation in the operating procedure 2406 only just then.
Logout and archive operations and do not move under the situation of playback operation at the same time, the understanding to the flow graph of Figure 116 has become simply to a certain extent.But, should be appreciated that always than the priority height of playback and file, the priority of playback is always than the priority height that files for the priority of record.
The processing that Figure 117 A has been performed when logout and archive operations at the same time have been described.First step 2410 according to Figure 117 A, data (are for example copied to dismountable recording medium of being used to file from a disc driver (appointed disc driver B), digital audiotape), and the video flowing of input is recorded to another disc driver, the driver A of appointment simultaneously.Step 2410 is a step 2412 later, in step 2412, judges whether all video datas are copied to this file medium from disc driver B.If no, just continue step 2410.But, in case finished this data disk driver B is copied out, just begin to copy data to this file medium from disc driver A, and use disc driver B to write down inputting video data stream (step 2414).In step 2416, judge whether to copy all video datas to this file medium from disc driver A.If no, just continue step 2414.But in case finished the file of this disc driver A, this processing just turns back to step 2410, and disc driver B is filed and carry out field notes on disc driver A.
This processing of carrying out in Figure 117 A can be summarized as follows: disk from the moment that file switches to record be finishing of filing.Figure 117 B provides an alternative technology, wherein continues data recording field on first disk, even after all video datas on another disc driver are filed.The beginning step of the processing of Figure 117 B is a step 2418, and wherein it is identical with the step 2410 of Figure 117 A.Among Figure 117 B, step 2418 is a step 2420 later.Step 2420 is identical with step 2412 among Figure 117 A.Promptly judge at all video datas on the disc driver B whether be copied on this file record medium.If no, just proceed step 2418.If but finished copy to disc driver B, be exactly step 2422 then.In step 2422, no longer carry out archive operations, but continue inputting video data is recorded on the disc driver A.Step 2422 is a step 2424 later.In step 2424, judged whether to use the preservation capacity (perhaps substitute, judge whether to have used a predetermined portions of recording capacity) of whole magnetic disk driver A.If no, continue step 2424.But,, be exactly step 2426 then if disc driver A has expired or reached the predetermined quantity degree.Step 2426 is identical with the step 2414 of Figure 117 A, and the expression field notes is switched to disc driver B from disc driver A, and the video data that begins to be recorded copies on this file medium from disc driver A.In step 2428, judge whether to have finished file to disc driver A.If no, just continue step 2426, if still finished file, be exactly step 2430 then to disc driver A.In step 2430, no longer carry out and file, but continue data recording field to disc driver B.In step 2432, judge whether disc driver B is full.If no, continue video data recording to disc driver B, otherwise just enter step 2418 again.Be that field notes is switched back disc driver A, begin again disc driver B is filed.
For Figure 117 A and Figure 117 B, supposed before the file of finishing another disc driver, will never reach the full recording capacity of a disc driver.If situation is not like this, just be designed to field notes is switched to the disc driver that is being filed from full disc driver.Should be appreciated that, when record that contact Figure 117 A and 117B discussed and archive operations are carried out, can carry out playback operation.
Before filing fully, another disk quilt forbidden that the technology that a disk is filled is provided among Figure 118 for one.To Figure 118, supposed a disc driver has been carried out record, simultaneously another disc driver is filed.In step 2434, the time of finishing archive operations is estimated, and considered the quantity of speed of filing and the data that will be filed.In step 2436, the depleted time of capacity of the disk that is used to write down is estimated.For example, the speed that is recorded according to the preservation capacity that is not used and the data of remainder is estimated.
In step 2438, judge before another disc driver is filled, whether will finish archive operations.If at first finished file, cycle of treatment turns back to step 2434 then.But if before file is finished, the capacity of another disk will be filled, and just reduce to message of this front-end electronics transmission to be used for the comprehensive speed (step 2440) of capture video data field to preserve.Make in such a way, reduced preservation speed, thereby allow to finish file another disk.
The user friendly feature of VR/PC unit
Figure 119 illustrated produce on the hard disk that will be stored in the VR/PC unit, be used to preserve about the front by a processing of the database of the index information of the file tape of VR/PC unit record.First step among Figure 119 is a step 2442, and when filing, this step 2442 request adds up the index messages of a file tape.In step 2444, expression continues the accumulative total to the index information of file tape, up to having ejected the file tape.At this moment, the index information by the ejection tape that add up is added to and is kept in the VR/PC unit in the archival database at least one hard disk (step 2446).This user is allowed to visit this file tape library, so that this VR/PC unit can help this user management to be stored in video data on this file tape.
Figure 120 has illustrated a processing, handles this VR/PC unit by this and carries out work and come the cause of the fault in the video camera that is connected to this VR/PC unit is carried out automatic diagnosis.The service life of typical video camera is limited, and its performance of increase in time reduces.The typical problem that the video camera that wears out runs into is can't focus on and " halation ", i.e. the output of some some positions generations is brighter than reality.This processing shown in Figure 120 make this VR/PC unit can from motion tracking and the diagnosis video camera characteristic reduction in time.
First step among Figure 120 is 2448, in step 2448, at first is connected to the just later of this VR/PC unit at this video camera, perhaps in the near future, catches an image that is produced by this video camera.Step 2448 is a step 2450 later.In step 2450, to carrying out an analysis of statistical data with the corresponding data of captive image in step 2448, produce when analysis subsequently by this video camera catch visual the time employed one group of foundation characteristic statistics line.Preferably, this foundation characteristic statistics line comprises the statistics of the color distribution of the statistics of high fdrequency component of one or two presentation image signals and presentation image data.This foundation characteristic statistics line is stored in the hard disk in this VR/PC unit then.
Step 2452 is illustrated in after the periodic generation foundation characteristic statistics line, carries out the automatic diagnosis to video camera.After beginning that video camera has been installed, for example, can use fixing interval, for example weekly or every month, carry out this automatic diagnosis and handle.The first step of automatic diagnosis is a step 2454.In step 2454, catch the current image that produces by this video camera.In step 2456, produce and the corresponding statistics of this foundation characteristic statistics line from this current pictorial data, and this statistics and basic statistics line are compared then.Shown in step 2458, judge according to the statistics of this current image and the comparative result of basic statistics line whether this video camera continues to have satisfactory performance then.For example, compare with basic high fdrequency component statistics line, the remarkable decline of the high fdrequency component of current picture intelligence may represent that this video camera no longer includes gratifying focusing.Similarly, the remarkable skew of the distribution of color data may be illustrated in the pickup device of video camera excessive halation.For fear of the fault of the discovery video camera of mistake, be proposed to be used in the captive image of diagnosis and under the light condition identical, produced with the light condition that produces the ground line image.For example, by under controlled light condition, taking ground line and image subsequently at night, just can accomplish top this point.
If in step 2458, find in visual statistics, to have the remarkable change of a not satisfied camera performance of expression, take suitable step then, for example show an alarm (step 2460), represent the work that this video camera can not be correct.Although it is only a single camera to be carried out automatic diagnosis that Figure 120 has illustrated, be appreciated that, by the sequential of catching of catching basic statistics line and automatic statistical picture is subsequently carried out suitable variation, the processing of Figure 120 can be applied to all video cameras that link to each other with the VR/PC unit.
With reference now to Figure 121, describes a processing, will be used to provide a status information data presented to a user by the generation of this processing VR/PC unit.The example that a kind of like this state shows is provided among Figure 149.
The processing of Figure 121 in step 2462, judges whether this user asks to want display system status information from step 2462.For example, can import the request (Fig. 6) that this state shows by exciting the status button 646 that is provided on the front panel.
Continuation is with reference to Figure 121, if asked a state request to show, step 2462 is a step 2464 later.In step 2464, this mainboard CPU judges that this disc driver or a plurality of disc driver that are included in the VR/PC unit have the recording capacity that how much is not used.In step 2466, judge the recording capacity what have be not used on the recording medium (digital audiotape) that is installed in the inner file DAT driver then.Below in step 2468, judge the number of the alarm event condition that has write down but do not browsed by the user.Step 2468 is a step 2470 later.Step 2470 produces number viewed but that do not have deleted alarm event to report.Step 2470 is a step 2472 later.In step 2472, this mainboard CPU produces the number of the video camera be connected to this VR/PC unit, in step 2474, the number of the video camera that is arranged in the activation record sequence is counted.In step 2476, all data that step 2464 to 2474 is caught are used to produce status information display screen (Figure 149) at last.
Figure 122 has illustrated and has allowed a user easily from watching a video image stream that is reproduced that is produced in a preset time by one first video camera to move to watching by the feature of a different video camera in the VR/PC unit of the video image stream that is reproduced that the identical time produced.The first step of Figure 122 is a step 2478.In step 2478, by one first video camera in the past the image stream that produced of a special time by regeneration from this hard disk and be displayed on this display monitor.The example of a playback display format, for example in step 2478, be provided, be displayed among Figure 12, the screen display of Figure 12 comprises a video image viewing area 2479 of the image that is reproduced that demonstration is produced by a video camera (being assumed to video camera 1).Example hereto, suppose be displayed on visual stream in this zone 2479 be by video camera 1 10 of mornings of the previous day produce.
With reference to Figure 122, step 2478 is a step 2480 later again.In step 2480, judged whether to select a different video camera to carry out playback.Flow (step 2478) by video camera 1 at the image that produced the day before yesterday if no, just continue regeneration.If but except with video camera 1 corresponding button this user also excited in other the video camera selector button 650 (Fig. 6) one, in the step 2480, select another video camera (video camera 2) then.In this case, step 2480 is a step 2482 later, this mainboard CPU carries out work, to search, to regenerate and to show the video image stream that is produced in the time (10 points of the previous day) identical with the time of the current playback video that is shown of video camera 1 generation by video camera 2 (selecteed video camera).Step 2482 is a step 2484 later, in step 2484, this VR/PC unit impels this display monitor to move to a separator screen display mode (among Figure 12 not show), shows simultaneously that in this pattern video camera 1 and video camera 2 video image that 10 institutes produce respectively in the morning of the previous day flows.
Should be noted that the virtual video camera selector button that can use mouse to excite to be displayed among Figure 12 carries out video camera detected in step 2480 and select, it can be used as an alternative that excites corresponding front panel video camera selector button 650.
Figure 123 has schematically illustrated the operation of VR/PC unit while executive logging and playback operation.Figure 123 has schematically illustrated the control thread 2486 and 2488 of operation simultaneously.Thread 2486 relates to video camera produced the on-the-spot incoming video signal stream of catching and writing down by being connected to the VR/PC unit, and thread 2488 relates to receiving user's request and asking to retrieve with playback according to this user and is stored in video data stream on the hard disk in the VR/PC unit.Be provided for the multitask characteristic of the operating system software of this mainboard CPU by use, realize that simultaneously it is possible independently controlling thread.
Shown thread 2486 comprises step 2490 and 2492.In step 2490, be used a time-multiplexed mode by the video image stream that video camera produced that is connected to the VR/PC unit and catch continuously, in step 2492, with the video data field sequential recording that produced on this hard disk.
Shown playback thread 2488 comprises step 2494 and 2496.In step 2494, this user initiates a request, request come the playback front by a specific video camera one specific that the time produced, be recorded in a video data stream on this hard disk.In step 2496, from this hard disk, retrieved and be displayed on this display monitor with the corresponding video data of this requested stream.For example, can provide a demonstration that shows according to the shown form of Figure 12.Should be appreciated that,, proceed the record of on-the-spot inputting video data stream and can not interrupt even the requested video data stream that is reproduced is displayed in the viewing area 2479 of screen display of Figure 12.
Figure 124 has used general formal specification and has shown two different features that are reproduced video image stream simultaneously.The first step of Figure 124 is a step 2502, and step 2502 is corresponding with the step 2478 of Figure 122.In step 2504, judge whether this user asks to regenerate an additional video flowing and show this additional video streams when step 2502 shows stream then.A particular example of step 2504 is the steps 2480 among Figure 122, in step 2480, in the stream that this user's request is shown in being created in step 2502, shows simultaneously by the different stream that video camera produced.Issuable another request of step 2504 is to ask to show simultaneously by the identical video camera that is created in the image stream that is reproduced in the step 2502 to flow at the different video images that is recorded that the moment produced.
Under any situation, when additional streams of request playback in step 2504, be exactly later on step 2506, in step 2506, receive or produce the parameter (time of video camera and generation) of requested stream.In step 2508, image stream that is shown in step 2502 and additional requested stream are used a separator screen form and show simultaneously then.
With reference now to Figure 125, describes a processing, handle, relate to that the data that are recorded in a video data field on the hard disk are recorded on the hard disk identical with the hard disk that records the video data field and one independently on the hard disk by this.Figure 125 comprises a first step 2510, in step 2510, receives a live video data fields sequence.Next procedure is a step 2512, and step 2512 expression produces and the corresponding index data in received video data field.Then, step 2512 is step 2514 and 2516 later, and these two steps are carried out simultaneously.In step 2514, inputting video data field and the index data that is produced in step 2512 are recorded on the hard disk together.In step 2516, identical index data is recorded on the hard disk different with the hard disk that records the video data field.
By one " shade (shadow) " group index data being recorded in independently on the hard disk drive, can independently search this index data on the hard disk drive at this, and can not stop ongoing, needs to visit record, playback or the archive operations that has write down the hard disk drive of video data field on it.
Figure 126 has described the VR/PC unit and has used than " forever " field rate that is assigned to this video flowing a high field rate to come an incoming video signal stream is alarmed the operation that impact damper is preserved in advance.First step among Figure 126 is a step 2518.In step 2518, the form that inputting video data stream was received and was used a video data field sequence is hunted down.To this example, suppose and use the speed of 3 of rough per seconds to catch this video data stream.
In step 2520, use a lower field rate, i.e. the speed of one of per second will in the captive video data field more selecteed main " the permanent recording zone " that is recorded in a hard disk drive in step 2518.(should be appreciated that, in should preferred implementation, only the video data on should " forever " recording section in this hard disk drive be filed; And preferably, this pre-alarm padded coaming is not filed) unless it is at first transferred to this " forever " part of this hard disk drive.Simultaneously, in step 2422, all captive data fields are recorded in the circular buffer zone of this hard disk drive, provide one with catch the recording rate that speed equates, promptly be 3 of per seconds in this example.
Step 2522 is a step 2524 later, in step 2524, judges whether to detect alarm conditions.If, be exactly step 2526 below, in step 2526, a pointer that defines this circular buffer zone is moved preserves forever to the data of 3 of this per seconds that (alternatively, the data of 3 of this per seconds can be copied to from this circular buffer zone and be used for permanent this main region of preserving the step 2526.)
In step 2528, use in the whole field with interior continuation that speed writes down (supposition is 3 of per seconds) detecting a later predetermined time period of these alarm conditions.
Should be appreciated that, the above-mentioned discussion of Figure 126 given field rate only be exemplary, and can change, it is mainly put and be that the temporal resolution (field rate) that is provided is bigger than the temporal resolution that is provided in step 2522 and step 2528 in step 2520.
Figure 127 represents a vague generalization example of this played-back image analysis examples described above here.In the first step of Figure 127, promptly in the step 2530, video data and/or accordingly all data from a recording medium of a for example hard disk, be retrieved out.Then, in step 2532, the data that the VR/PC element analysis is retrieved.For example, one or more image analysis algorithm that can use the front or describe later on.Substitute, can use other image analysis algorithm, comprise the motion that relates to the target that in image stream, is expressed or other algorithms of acceleration.But in step 2532, the machine analysis that is used is not limited to detect image feature.For example, this index data can be by investigation to judge that a certain given time in the past is which video camera stream has been recorded.
In step 2534, be stored in the data on this hard disk or other recording mediums, index data or similar data are changed or add, and are illustrated in the result of the machine analysis that is performed in the step 2532.
Should be realized that, the processing that is provided among Figure 127 generally considered should the machine analytical applications to be stored on should recording medium should video database, upgrade the result that index data or a data represent that this machine is analyzed then.With reference now to Figure 128 (a) and Figure 128 (b), an example of the referred image processing facility in front is discussed.Especially, these figure relate to a feature that allows this user to improve in image plane the image contrast in the selecteed part.
Figure 128 (a) has shown in the front end analog board electronic installation that may be provided in Figure 113, the next amplifier 808 ' that substitutes a gain-variable that is displayed on each amplifier 808 among Figure 13.
As Figure 128 (a) 2536 shown in, the gain of this amplifier 808 ' can be by control signal control from this front controller DSP-1.
Figure 128 (b) has illustrated and has realized that contrast optionally improves the processing of feature.In step 2538, select the part degree of comparing of image plane to improve.In step 2540, a suitable message is sent to this front-end electronics then, impels the amplifier 808' of each gain-variable to carry out work, with the image contrast (dynamic range) that increases chosen part in this image plane.
Figure 161 is provided for an example this user, that allow to select the display screen that part degree of comparing improves in the image plane.Represented rectangle drawing element is to be selected to carry out the zone that image improves in the image plane in 2542.
The example of user interface screen
Figure 11 is provided at an example that shows in the IVIM system by a screen display format of a present field video signal that produces in the video camera that is connected to the VR/PC unit.This screen display format of Figure 11 comprises the top 2546 of screen and the bottom 2548 of screen.Top 2546 comprises a video image stream display window 2550.Line up a horizontal array six switch regions 2552 that can excite with mouse are provided below this live video display window 2550.As can be seen from Figure 11, the switch region 2552 that can excite with mouse top has symbol description " full frame " (full screen) respectively, " planning is opened " (opening the pre-planning operator scheme), " facility ", " file ", " foundation " and " help ", the display formats that two of exciting this " full frame " switch region to impel to be shown in Figure 11 part shows are used as on-the-spot visual display window with whole screen area a display format is substituted." planning is opened " switch allows this user to impel this VR/PC unit to enter the pre-planning operator scheme." facility " switch region allows the specific system features of this user capture, for example sends data, the report that perhaps produces system activity to other devices." file " switch region allows this user capture to relate to by the feature of the performed file function in this VR/PC unit." foundation " switch region allows this user to enter the pattern that this VR/PC unit is configured." help " switch region allows the explanatory text of user capture context-sensitive to show.
The LED viewing area 2554 of a simulation is provided in the position, bottom right of upper area 2546.A symbol description " record " is near this simulation LED2554.This simulation LED2554 is corresponding with the LED654 that is provided on the physics front panel (Fig. 6).Preferably, when this VR/PC unit was writing down field video signal, the LED2554 of this simulation was shown as first state (for example, redness), and when not writing down, it is shown as a different state (for example, black or Dark grey).
If the front panel layout that the lower area 2548 of the screen display of Figure 11 and Fig. 6 is shown is compared, can observe the corresponding layout of the layout with this front panel that has of this lower area 2548.Especially, the switch region that can excite with mouse be provided at the zone 2548 in the switch 638,640,642,644,646,648,650,658 of front panel (Fig. 6) and 660 corresponding position, relevant position.Have symbol description " 1 " to arrive " 16 " respectively on the video camera selector switch zone in this zone 2548, and the video camera selector switch 650 consistent with the number on the physics front panel is corresponding.In addition, 2556 of Figure 11 represented " alarm " switch regions are corresponding with the switch that indicates " alarm " 658 on this front panel.
The right of lower area 2548 have be aligned to one laterally be 2 vertically be 3 array and on position and function with switch 638,640,642,644,646 and 648 corresponding six switch regions of front panel.Used by 2558 represented single rectangles with these front panel switch 638 corresponding switch regions to be covered, this single rectangle represent by exciting switch 638 or above have the switch region of this single rectangle to set up a single window display format.Similarly, a covering of one 2 * 2 windowed format of expression is provided at and switch 640 corresponding switch regions, a covering of one 3 * 3 windowed format of expression is provided at the switch 642 corresponding switch regions with front panel, and a covering of one 4 * 4 windowed format of expression is provided at the switch 644 corresponding switch regions with front panel.All above-mentioned switch regions all can be excited by the operation mouse in the lower area 2548 of Figure 11, and its effect is with physically the effect by the respective switch button on this front panel is identical.Also be included in this zone 2548 be above one of symbol description " broadcast " circle, that can excite with mouse basically switch region 2560 is arranged.This circle switch region 2560 is corresponding with the round switch 660 of the propelling slowly of this front panel on position and function.Slowly advance round switch similar to this, there is the symbol description of arrow switch region 2560.In the switch region 2560 arrow the zone by being operated, so that this switch region 2560 turns clockwise or is rotated counterclockwise with mouse.The effect simulation of the rotation that can excite with mouse of switch region 2560 manual this front panel of rotation advance the effect that comes and goes switch 660 slowly.LED652 that the simulation front is discussed with reference to this front panel (Fig. 6) and 656 simulation LED viewing area have also been comprised on position and function on the lower area 2548 of the screen display of Figure 11.
Another notable feature of the screen display format of Figure 11 is the covering (overlay) that is provided at the alphanumeric character in the visual display window 2550.This shown covering comprises a date and time information and a symbol description " scene " among Figure 11, and this symbol description " scene " is clear and definite represents to be provided to the reader in the window 2550 is a field video signal.A video camera sign covers the lower left corner that also is provided at window 2550.
Figure 12 has illustrated an employed display format when the VR/PC unit uses a playback mode to carry out work.Can perhaps enter this playback mode by the broadcast/time-out zone that excites advancing slowly of this front panel to come and go on the switch 660 by symbol description " broadcast " with click center of shown round switch region 2560 in Figure 11.The screen display format of Figure 12 screen display format with Figure 11 basically is identical, and the difference between these two forms only is discussed.Should be appreciated that at first video image stream expression one image that be recorded previously and that be reproduced at present that is shown flows, rather than an image scene, that produced at present stream in the display window 2479 of Figure 12.
Identical among switch region feature among Figure 12 and simulation LED zone and Figure 11, switch region " planning is opened " except Figure 11, " facility ", " file " and " foundation " " withdrawed from " by being masked as in the display format of Figure 12, and the switch region of " search " and " instrument " substitutes." withdrawing from " switch region among Figure 12 excite an on-the-spot display mode that this system is turned back to form with Figure 11.Exciting of " search " switch region among Figure 12 shows produce a menu screen that allows this user to select in the video database function of searching.Should allow this user capture image processing facility in " instrument " switch region.
Figure 129 represents and can show by the search Dialog Box screens that " search " switch region that excites on the screen display of Figure 12 to be called by this user.The principal character that the search dialog box of Figure 129 shows is to select to have respectively symbol description " date ", " video camera ", the mark dialog box of " alarm " and " instrument ".In the shown particular display format example, " date " dialog box is topmost in Figure 129.Can excite other mark dialog boxes by clicking corresponding mark.
This date dialog box has three model selection settings that can excite with mouse, is represented as " image that all are recorded " respectively, and " ... between image " and " image wherein ".First selects the video database search of realization without any time restriction.Second selects to provide the beginning that limits this searchs period and stops to limit.The 3rd selects this search is restricted in the given delay before the current time with during this current time.
Be to be used to explicit identification to find " field as a result " with the information of the image of search criteria coupling below this dialog box zone.On the right in this dialog box zone is five switch regions that are aligned to a vertical row, has symbol description " to search " respectively, " stopping ", " broadcast ", " closing " and " help ".Being right after below these switch regions is an icon area that has wherein shown " searchlight " icon.When carrying out one when searching, use the mode of an oscillatory scanning to move this searchlight icon 2562, represent that to the user search carries out.
Excite " searching " seek button to impel and carry out selecteed search.Excite " stopping " button to impel and stop a search of carrying out.Excite " broadcast " switch region to impel the playback of VR/PC unit and selected corresponding video flowing in the field that is produced.Excite " closing " switch region that the playback screen that this user turns back to Figure 12 is shown.In Figure 129, can observe and select " image that all are recorded ".
Figure 130 is and the similar display screen of the display screen of Figure 129, but expression select " ... between image " option rather than " image that all are recorded " option.Also show to allow the user to be provided with to limit a Pop-up dat frame on a date in this search period among Figure 130.On the left side of this Pop-up dat frame is the setting of the time among the user can regulate a day.
Figure 131 has illustrated that the search dialog box that is provided shows when selecting " alarm " dialog box.On the left side of this alarm dialog box, this user can select to relate to the search criteria of the alarm event that is detected by the external alarm sensor device.Can whether detect the predetermined characteristic that this image analysis algorithm will detect according to the image analysis algorithm that carries out work when producing this video data this user of the right of this alarm dialog box, select search criteria.
Figure 132 has illustrated the search dialog box display screen that is provided when selecting the video camera dialog box.In the shown example of Figure 132, excite a drop-down video camera to tabulate and allow this user to stipulate the video camera stream that to be searched.
Should be noted that the field that is produced is included in a tabulation of found four mapping (enum) datas part in the search of a front in Figure 132." broadcast " switch region on the right of the demonstration 132 is allowed the video data part of this user's playback list in the field that is produced by bright apparent (comparing with Figure 129 to 131).
In Figure 133, again the option date/the time dialog box.In addition, be listed in produced the field in the first video data part selected, and if this user excited " broadcast " switch region with regard to this first video data part of playback.
Figure 134 excites to the user that " foundation " switch region responds among Figure 11, and what produced sets up the screen of options curtain.This sets up the screen of options curtain provides four to set up option to this user, and each sets up option can be accessed by a corresponding switch region.These options are " name " (name are applied to video camera and alarm input and output), " record " (allowing this user to select to write down option), " safety " is (if be authorized to, allow this user that this VR/PC unit is carried out the configuration of safeguard construction), " login " (, allowing this user's increase or deletion to be allowed to sign in to the user of this VR/PC unit) if be authorized to.This screen display that impels in addition that is provided turns back to " closing " switch region of display format of Figure 11 and " help " switch region of customization.
Figure 135 is to an example of the record option dialog box that excites " record " switch region among Figure 134 to respond to be shown.This dialog box that is shown in Figure 135 allows this user option 3 Class Options: logging mode, recording quality, and recording rate.Two possible logging modes are " linearities " and " circle ".In linear model, when the preservation capacity of hard disk drive runs out, just stop at the record on this hard disk drive.In circular logging mode, when reaching the end of this hard disk storage area, next inputting video data is written to the beginning of this hard disk storage area.
The recording quality option is " normally ", " high fidelity " and " expansion ".In these three quality options each is corresponding with the different quantization bit speed of being carried out by JPEG chip 848 (Figure 14).Continuation is with reference to Figure 135, and the data byte that " high fidelity " is provided with each video data field of average employed expression is provided with employed many than " normally ", so that the image quality that is enhanced to be provided.The data byte that " expansion " is provided with each video data field of average employed expression is provided with employed lacking than " normally ", preserve capacity with the more efficient use hard disk, but image quality only reduces a bit.
Shown in Figure 136, realize the recording rate option by a drop-down list.It is corresponding that speed is caught in this rate options and different fields, and be illustrated in corresponding rate options and preserve effective record that capacity provided by hard disk and postpone (2,6,12 or 24 hours).Although the option in Figure 135 and 136 is not expressed, also can one video camera stream connect a video camera stream speed and quality option be set.In addition, quality settings can be used to regulate compression front end " block similarity " threshold value and/or reference picture speed, rather than regulates the JPEG that is driven at present by the quality settings option and quantize to be provided with, and perhaps can be used for carrying out above-mentioned adjusting simultaneously.Can also design each quality and rate options, show that the sample video image sequence represents the image quality that can obtain to this user from each quality and speed are provided with.
Figure 137 is to exciting the dialog boxes for login that " login " switch region responds and is provided among Figure 134.The dialog boxes for login of Figure 137 requires this user's use to import people's identification number (PIN) one by one at the virtual digit keyboard on the right of dialog box.A backspace switch region 2564 also is provided.When having excited in this dummy keyboard one digital, one is displayed in the zone below the symbol description " input PIN number: " with the corresponding asterisk of each numeral.An input function is realized in " OK " switch region, reads the numeral that is transfused to constantly at this and judges whether it is that an authorized users is attempted login.
If the user that current login enters is authorized to reset the security feature of this unit, excite " safety " switch region in Figure 134 will impel the screen display that in Figure 144, is shown to be shown then.Shown main option is to increase a new user (" increase " switch region) in the demonstration of Figure 144, changes a user's who has existed access characteristic (" editor " switch region) and deletes a user who has existed (" deletion ").In the zone on the left side of the demonstration of Figure 144, a tabulation that existed, authorized users is arranged.
The access privileges that excites " editor " switch region among Figure 144 to allow visit one first example to be displayed among Figure 138 shows.Scroll bar in a vertical direction of the lower right-most portion of Figure 138 allows this user to see the setting of various feature access privileges by rolling.The option that is provided with that Figure 138 is shown comprises the ability (" enabling " below " planning ") of covering pre-planning operator scheme and the pre-planning operator scheme is configured (" configuration " below " planning ").
Figure 139 has shown and has comprised and relate to another view that feature access privileges that system sets up the option of priority is provided with option.Shown priority and the option of setting up that relates to video camera that is used to select to be applied to the analytical algorithm of input video stream set up among Figure 139, live video display format option is corresponding with the security of system feature.
Figure 140 is the demonstration identical with the demonstration of Figure 139, but represents that the security feature access privileges has been given a specific user.In addition, " video camera " left side has been given this identical user by bright apparent zone in front with these privileges of " analysis tool " feature privilege expression.
Figure 141 is illustrated in the identical demonstration with respect to another scrolling position of this feature access privileges.Shown privilege relates to selection logging mode option (" configuration " below " record ") among Figure 141, accessing operation system software file (" record " below " maintenances "), visit are stored in the video data (" enabling " below " playback ") on this hard disk and visit the vision signal (" VCR transcribes " that " playback " is following) that is stored on the video tape recorder that is not shown, traditional outside is connected.
Shown further feature access privileges among Figure 142, i.e. ability that the event handling pattern is configured (" configuration " below " incident ") and visit is about the ability (" report " below " incident ") of the report of detected incident.
Figure 143 has further shown the feature access privileges.These features relate to retrieval be stored in one on the file record medium data (" files " following " enabling ") and be stored in preservation and retrieval (" recovery " and " storehouse ") on the hard disk, that the content of file record medium carried out indexed data.
Figure 145 represents exciting " name " switch region among Figure 134 to respond the screen display that is shown.In this display screen of Figure 145, can visit three dialog boxes i.e. " video camera ", " alarm input " and " alarm output ".In the specific demonstration that Figure 145 shows, selected this " video camera " dialog box.This video camera dialog box provides the name zone in 16 video cameras each.Can use a keyboard (not being shown) that is connected to this VR/PC unit, perhaps be displayed on this display monitor and can not come the numeric-alphabetic name of input word by " dummy keyboard " (being shown) that this mouse excites.
Figure 146 provide can be accessed in Figure 145 " alarm output " dialog box.This alarm output dialog box allows the alphanumeric of 16 alarm output signals is specified.
Figure 147 provide also can be accessed in Figure 145 (perhaps Figure 146) " alarm input " dialog box.In Figure 147, name can be transfused to the respective external alarm sensor devices that identify 16 inputs of generation alarm detection signals.
The screen display that Figure 148 is shown exciting Fig. 1 " facility " switch region to respond.The option that is provided for this user in the demonstration in Figure 148 is to an external device (ED), for example this locality or host node send data (comprising video data), generation will and produce signal by the report of a printer (not showing) operation that print, that relate to this VR/PC unit and be operatively connected to video camera this VR/PC unit, transportable.
Figure 149 is to exciting a display screen that responds and be shown with " state " Figure 12 switch region Figure 11.The status data that is shown among Figure 149 is produced by the processing that top contact Figure 121 is discussed.Figure 149 video data, these data are illustrated in remainder available on this hard disk or a plurality of hard disk and preserve capacity, available remainder is preserved capacity on the detachable file record medium, the number of alarm event that be detected and that do not browsed by this user, viewed but, be connected to the number of video camera of this VR/PC unit and the number of the video camera that is being recorded from the video flowing that wherein comes not by the number of the alarm event of from this alarm event file, deleting.
Figure 150 is a display screen that is provided to be used to define a pre-planning live video display format.In the display screen of Figure 150, can be respectively 2 * 2,3 * 3 by the dialog box of this user capture, 4 * 4 and " customization " video display window form.This 2 * 2 dialog box is shown as and activates among Figure 150.The right literal " weekend " at the top of selecteed dialog box represent this user be defined in time on daytime of being defined as weekend during with the interior pre-planning operator scheme that will be automatically performed.The left side of this dialog box is some icons, and each icon is corresponding with a corresponding video camera that is connected to this VR/PC unit.On the right of this selecteed dialog box is one 2 * 2 empty frame array of four video display windows in the display format of representing be established.Shown in Figure 151, cursor of available use 2566 is drawn a camera icon, so that this icon is placed in the frame.This video camera is placed in this frame expression should corresponding camera video flow distribution shown in the display window accordingly.In the particular example of Figure 151, the designated left upper end that is presented at this 2 * 2 display format of " stopping label " video camera stream.
Figure 152 sets up the display screen that operating period is provided in planning, will be taken as vacation to define which calendar day.On the left side of the demonstration of Figure 152, provide a calendar to show and allowed this user to select the specific sky of the specific moon of a year.The right in the demonstration of Figure 152 provides a tabulation on the date that is designated as vacation.Excite " increase " switch region among Figure 152 to impel that a selecteed date will be added in this tabulation vacation in this date shows.
Image analysis tool is set up screen
Figure 153 represents to be provided this user, allows this user that the parameter of an image analysis algorithm that is designated as " optical tool " is set.This algorithm be designed to detect the selecteed brightness in the interested screen change.Screen display among Figure 153 comprises a visual display window 2568 that has wherein shown a static video.This static video is used to define the parameter of this optical tool.What be shown in this image is a display element frame 2570, and this display element frame 2570 has shown the zone in this image plane that this algorithm will operate on it.By using the mode of in a traditional computing machine drawing software bag, handling similar graphic element to make mouse-based operation, this frame 2570 can be dragged to the another location this image plane from a position, and its size can be reduced or be increased and shape can be changed.Use a mode brighter to be presented at image area in this frame 2570, so that highlighted demonstration is carried out in the zone in this frame 2570 than the remaining part of this figure.A circular point 2572 is arranged in this frame 2570.If excited a some brightness detected characteristics of this analytical algorithm, the size of the luminance point that the size Expressing of this point 2572 will be detected.Preferably, this point is shown as a solid color, and is for example red.
So be used to select the feature of this algorithm and the virtual push button of parameter and the right of the demonstration that switch is provided at Figure 153 be set.In 2574, this user is allowed to fixed position (" sign ") or deletes (" wiping ") by these frame 2570 defined active regions.In 2576, this user is allowed to select in three operator schemes of this algorithm: a big increase of sensed luminance (" secretly arrive bright "), a big minimizing of sensed luminance (" bright to dark ") or detect a bright luminous point (" putting light (spot light) ") in this active region.First option detects this VR/PC unit and when opens light, and second option detects this VR/PC unit and when closes black out, and the 3rd option is used to make this VR/PC unit to detect the appearance of a flash of light in the dark area.
Slider bar in 2578 is controlled the size of the point 2572 of a parameter setting that is used for this light operator scheme.When this slider bar is operated to the left side, just reduce the size of point 2572.When this slider bar is operated to the right, just impel this spot size to be increased.
2580 provide the sensitivity of this optical tool that option is set.The sensitivity parameter of this optical tool is configured for judging whether having taken place the threshold value from dark to bright or change from bright to dark, perhaps judge will be detected in this optical mode this light-emitting zone whether be sufficiently bright constitute will be detected an incident.If this user has excited " default " switch region, use then by the designator of this VR/PC unit and be thought of as optimum threshold value setting.This user can also regulate this sensitivity degree and operate slider bar in 2580.This slider bar moved to left just reduce sensitivity, this has reduced the danger of fault alarm, but has increased the danger of not noticing important incident.This slider bar is slided into the right has increased sensitivity, has reduced the danger of not noting critical event thus, but has increased the danger of fault alarm.Should allow dialog box of this user capture in " ADV " switch region, various sensitivity parameters do not have tied and can be provided with respectively in this dialog box.
In visual display window 2568, there is a symbol description to identify the type of the analytical algorithm that is being established and the type that will be employed this video camera of this algorithm.Be to excite the switch region of this algorithm application to selecteed video image stream below this window 2568, perhaps the switch region of this analytical algorithm is selected in cancellation.
Figure 154 provides and has allowed the screen display of this user's setting about the parameter of a motion detection analytical algorithm.Set up screen classes seemingly with the optical tool that is displayed among Figure 153, this motion detection of Figure 154 is set up screen and is comprised a visual display window 2568 and a graphic element frame 2570, and this graphic element frame 2570 has defined a zone of the image plane that wherein will be employed this motion detection algorithm.The upper right portion of the demonstration among Figure 154, represented by label 2582, provide the feature that to use cursor to excite to allow this user to excite, perhaps deexcitation or remove one or more and one or more graphic image element frame 2570 corresponding active regions.Demonstration among Figure 154 also comprises and a corresponding sensitivity control area 2580, the sensitivity control area of Figure 153.As for the motion detection instrument, this sensitivity control provides the threshold value of the factor of the quantity of detected motion in this active region and/or intensity level.
Figure 155 provides and has set up screen display, invades the check and analysis algorithm to be used for above-mentioned border.Set up screen display with other instruments of being discussed just now, the demonstration of Figure 155 comprises a visual display window 2568.The front has been got in touch the Processing Algorithm that shows among Figure 90 A-90C and has been mentioned other features that show among Figure 155, comprises the graphic element frame 1966 of representing the border, through direction arrow 1976 and object size frame 1986.The control that is provided in 2584 allows this user to come mark or wipes and this frame 1966 corresponding marks.In whether this this direction arrow 1976 of control selection in 2586 points to respect to this border, point to outer or both direction.2588 slider bar is controlled the size of this object box 1986.This slider bar 2588 moved to left just reduce the size of this object box, operate the size that this slider bar just increases this object box in the opposite direction.Should be appreciated that,, can reduce or increase the size of this object box by the direct control mouse as an alternative mode, as in traditional computing machine drawing software bag to the picture element operated.
The front has been got in touch Figure 90 D and has been mentioned sensitivity control 2006.The sensitivity factor that can control 2006 may comprise the contrast ratio and with regard to the appearance of one or more motion, the confidence level of the size of the position of motion object and motion object.
Although be not displayed among Figure 155, can also be designed to allow this user that the speed on this border of object crosses is arranged to a parameter.Can also provide to be imported the graphic element that a set speed is vibrated by the user.For example, the form of expression of this graphic element can be similar to the baton of a music beat device.This VR/PC unit can be programmed for the detection image edge, detect the distant view of the degree of depth of presentation image scene, and the data tile space measurement that comes governing speed is to consider the degree of depth when estimating the speed of an object.A slider bar control that is used to be provided with this speed parameter can provide with this " beat " element, and perhaps substituting should " beat ".
Figure 156 be called " museum's instrument " analytical algorithm set up screen.The purpose of this museum's instrument algorithm is to allow to detect automatically removing of an object, and for example a width of cloth is drawn, a jewelry or similar.As other instruments are set up in the screen, this screen that Figure 156 shows comprises a visual display window 2568.Graphic image element frame 2590 has defined a zone will being monitored by this analytical algorithm.The width of this frame 2590 and height or both can be changed, to change monitored zone.Unless this monitored zone is very little, otherwise a preferred implementation of this museum's instrument algorithm need define some by colored rectangles 2592 represented " focus ".When focus 2592 is provided, in fact only need to monitor part with the corresponding image plane of this focus.Great zone is arranged by defined monitored regional occupied relevant in the number of the focus that is provided and this image plane by this frame 2590.In a preferred implementation, this user can reduce the number of focus, but can not increase its number.The purpose of only monitoring focus and not monitoring whole monitored zone is to save storer.In a preferred implementation of this museum's instrument, come removing of monitoring target by noting the difference between the content of the corresponding data tile in the content of the data tile on this focus and the reference picture.
Control in 2594 allows this user to come mark or wipes and this frame 2590 corresponding monitored zones.Control in 2594 allows this user to come mark or wipes focus.This user is allowed to reset the position of focus in monitored zone by using a cursor to draw focus.
Control in 2598 allows this user definition thinking how long this monitored zone must be missed or be closed before an incident has taken place.
In 2602, provide sensitivity to control for example variation of colourity, the number of the focus that is closed or the similar factor have been set.
Figure 157 is to use a screen display that is used to produce the demonstration of tabulating a vacation among a Figure 152 form, alternative.In the form shown in Figure 157, use a pull-down unit usually to realize this date choice box, and should " be set vacation " dialog box be to comprise " generally ", " working day ", in some dialog boxes at " weekend " and " vacation " one.
Figure 158 provide can be accessed in Figure 157 " working day " dialog box.Figure 158 plans the recording status of 16 video cameras being connected to this VR/PC unit.Generally, the form that shows among Figure 158 is a bar chart, and its horizontal bar is used for each demonstration of 16 video cameras.The bar of different colours is represented different operation modes this video camera, that be selected for the appropriate section of 24 hours periods.Can use cursor to operate this, stretching or to compress this, its effect be increase or reduce the corresponding operating pattern by planning period.To each video camera can with the operator scheme option be " activation " (writing down), " alarm " (in incident only a condition being detected of record) and " forbidding " (not writing down video camera flows).A Pop-up element in 2604 uses the form of numeral to describe by selecteed the time that element is represented.Pop-up element 2604 shown in Figure 158 is corresponding with the bar of video camera 7, as selecting shown in the announcer 2606.The form of " weekend " and " vacation " dialog box and Figure 158's is similar.
Figure 159 has shown and the corresponding dialog box of " generally " dialog box that can see in Figure 157 and 158.The control of among Figure 159 2608 allows this user to select between the planning option of fixing planning option and customization.In 2610, when this user takes place " time on daytime " and " time at night " the definition on working day if being allowed to.2610 control is that one of the demonstration that is provided among Figure 81 is substituted.
Continuation allows this user definition start time and the concluding time at weekend with reference to Figure 159 in 2612 controls that are provided, and the control 2614 allows this user definition start time and the concluding time of vacation.
Figure 160 is a variation of the display screen of Figure 159.From the control 2608 of Figure 160 as can be seen, " weekend " and be not selected for " vacation " planning of customization.As a result, be presented among Figure 159 this weekend and vacation dialog box in Figure 160, be not used as option and provide.
Figure 161 provides a screen display, allows this user to select and be provided with a parameter of an image processing facility.This display screen of Figure 161 comprises a visual display window 2616 that is used to show a image that will be processed.Referred above frame 2542 is capped in the part of this window 2616, and has defined this part that wherein will be employed the image plane of a selecteed treatment facility.Convergent-divergent control 2618 is provided at a upper right portion of this display screen.The control of this convergent-divergent allows a user to being dwindled and is amplified by the defined part that should image of this frame 2542, and/or mobile this zone that will be scaled.The switch region that is masked as " 1: 1 " is one with this image recovery does not have scaled state.The switch region that is masked as " improvement " resembles set of diagrams to improve to handle and is applied to this scaled zone, and this picture group resembles to improve to handle and comprises outstanding high fdrequency component, increases contrast, normalization color and intensity distributions, the non-linear interpolation of point, rather than linear interpolation.Be that the facility that allows a user to select from other image processing facilities except that zoom function is selected control 2620 below control 2618.For example, can be used control 2620 facilities of selecting is that the selectivity contrast that earlier in respect of figures 128 (a) and Figure 128 (b) describe is improved facility.Other image processing facility options that can be provided for this user comprise adjusting brightness, " sharpness " (i.e. degree of the high fdrequency component of outstanding pictorial data) and color and/or intensity distributions.The switch region of a horizontal array is provided in the bottom of this display screen.A switch region that is masked as " flooding " switches to a flooding form with the screen format of Figure 161.Image of retrieval from a floppy disk that is inserted into this VR/PC unit is impelled in the switch region that is masked as " being written into ", so that this image that is retrieved can be displayed in this window 2616.This image that is displayed in this window 2616 is impelled in the switch region that is masked as " preservation ", comprises any change that processed facility applies, and is written in this floppy disk.
" printing " switch region impels the image that is displayed in this window 2616 to be output through a printer as a hard copy, perhaps is sent out as a fax.The switch region that is masked as " recovery " is removed any by change that this treatment facility applied.
Figure 162 allows a user to realize abandoning a screen display with a processing of the corresponding video data of specific part of this image plane.The screen display of Figure 162 comprises a visual display window 2622 that is used to show a video image.This user for example can form, and 2624 and 2626 represented polygon drawing elements define in this image plane, its video data does not need the zone that is saved.2628 provide the control that relates to selecteed zone.This user's appointment is responded by polygon diagram 2624 and 2626 represented zones, this this front-end electronics of mainboard cpu command, be not sent out by the appropriate section of an image plane in the vision signal that this selecteed video camera produced and preserve, more effectively utilized the preservation capacity of this hard disk thus.Illustrated feature allows this user to select to be judged as uninterested zone in the image plane for a specific video camera image stream among Figure 162.This uninterested zone can be to be similar to zone represented among Figure 162, they replace in essence, perhaps (for example may include frequent motion, a highway in the background, the perhaps tree of a wind) zone, although if this frequent motion is not dropped, seem to produce the data bulk of a large amount of difference video data fields, they can not provide important information.
Figure 163 has illustrated a live video display format of the live video display format that is similar to Figure 11, except this live video display format window 2550 has been divided into the subwindow of 42 * 2 structures.In the particular example of this shown 2 * 2 structure of Figure 163, each in two different on-the-spot input signals is displayed on two in 4 windows.But, can be designed in the form of Figure 163, show four different live video streams simultaneously.In addition, shown in 2630, provide a bright color box (for example using red) to represent to write down this image stream of being considered in the upper left of each window.Should be appreciated that,, do not occur this announcer 2630 then if an image stream that is displayed in the subwindow is not recorded.Can also design other forms of record and select announcer 2630, comprise a circular announcer of the lower boundary that for example is positioned at this display window.
Figure 164 responds to " file " switch region in the demonstration that excites Figure 11 (perhaps Figure 163), a screen display that is produced.With reference to Figure 164, this user is provided a selection with " file " and " search " corresponding two dialog boxes.The specific example of this display screen that Figure 164 is shown shows the situation that " file " dialog box is activated.This file dialog box allows this user to select in four archive operations patterns: " backstage ", archive operations was carried out (preferably continuously on the backstage when wherein other feature operations in the VR/PC unit took place, the priority ratio record of this file background process or the priority of playback operation are low, shown in the above-mentioned discussion of Figure 116); " noiseless backstage ", it is identical with " backstage ", except the data that relate to the file space are not shown (Figure 149) shown come out by state; " fast special-purpose ", it is wherein not write down or a pattern of playback operation, so that this mainboard CPU handles and the disk access resource is dedicated to apace video data from this diskcopy to this medium that files; " file is closed " wherein do not carry out archive operations.
This file dialog box represents that also which device is being used to the function of filing; Under this situation, it is the inner DAT driver of installing.Shown those are similar among switch region in a vertical row on the left side of the screen display of Figure 164 and Figure 129-133, but they are can not be operated when having selected the file dialog box.In addition, these switch regions can come together to use with following " search " dialog box that will be referred.Be provided in the form of Figure 164 is icon 2562 similar " searchlight " icons of being discussed with the top Figure 129 of contact.In addition, search result field is provided at the lower part of this screen display.
Figure 165 provides the alternate version of the demonstration of a Figure 164.The difference of Figure 165 and Figure 164 is that Figure 165 provides two additional dialog boxes that are marked as " storehouse " and " customization " respectively.In addition, the file dialog box of Figure 165 allows this user to select in some inside or external record medium drive assembly.The selection that is provided in Figure 165 is an inner magneto optical driver device of installing.
Figure 166 provides a screen display that can obtain by " search " dialog box that is chosen among Figure 164.This shown search dialog box allows this user definition will be to being stored in date and time or alarm parameter on the current file record medium that is inserted into, that interested video data is used in searching among Figure 166.
Figure 167 is a screen display that is provided when selecting " storehouse " dialog box in Figure 165.The purpose of this storehouse dialog box is a contact file medium database that Figure 119 discussed, that be recorded previously above the visit.This user may ask to browse the corresponding data of file medium that are recorded previously with all, perhaps only browses those and is recorded in a user-defined time durations with interior data.The display field that is provided at the bottom of this screen display is to be used to show requested information.Should allow this user to be increased to this database in " increase " switch region with a corresponding item of addition record file medium.Should allow this user that an item is removed from this database in " deletion " switch region.Should allow this user to select to relate to the demonstration of the detail data of an item in this database in " details " switch region.Should allow this user to come from this record file medium database, to remove all data in " resetting " switch region.
" customization " dialog box that shown display screen has illustrated in the screen display in Figure 165 to 167, can use this user among Figure 168.In the dialog box of this " customization ", this user is allowed to upgrade the tabulation that has been installed in a file medium actuator unit in the inside, VR/PC unit or that be connected to this VR/PC unit.In the type of the file medium actuator unit that is mounted or may be mounted, the screen display of Figure 168 has shown following: a digital video disks (DVD) driver, an inner DAT driver, a magneto optical driver, so-called " jazz " driver (dismountable hard disk) and a DAT disk conveying belt.Certainly, can be designed in these type of device any one or all be connected to this VR/PC unit.
Analysis tool is selected screen
Figure 169 is a screen display to selecting Figure 129 to respond, be provided to any one " instrument " dialog box among Figure 133.Purpose that should " instrument " dialog box be allow a user come to a corresponding video flowing that is recorded of given video camera, selection will be applied to an image analysis algorithm of this image stream.Should " instrument " dialog box comprise a drop-down list 2632 that is used to select a video camera, with another drop-down list 2634, allow this user to select to be applied to an image analysis algorithm of video flowing that produce, that be recorded by selecteed video camera.A switch region that is flagged as " foundation " allows this user to propose to set up screen display with the corresponding instrument of selecteed analytical algorithm.The instrument of the foregoing has discussed is set up the example of screen display, and the example that this instrument is set up screen display is displayed among Figure 153-156.In the lower part of Figure 169, the expression front is shown the information that analysis tool is assigned to the video camera stream that is recorded.
Figure 170 shown ought be drop-down this drop-down menu below " instrument that will use: " of Figure 169.As can be seen, this menu provides several options about the selection of image analysis algorithm to this user, comprises " tool movement " " border instrument ", " museum's instrument " and " optical tool " from 2636.In addition, this user may be chosen in and not use any analysis tool in this search.Like this, this screen display of Figure 170 allows this user to select to be used for easily from a plurality of image analysis algorithms that are saved previously to search.One as the tabulation of the speech of analytical algorithm option shown in Figure 170 substitutes, and can also be designed to user's icon that this user can select to represent the analytical algorithm that is saved previously.
As can be seen, this pull-down menu also is provided for this user and allows this user to select to be applied to an analytical algorithm of on-the-spot input video stream from Figure 171 and 172.In this shown screen display of Figure 171, a video camera connects a video camera ground and distributes this analytical algorithm, and in Figure 172, provide even meticulousr " granularity ", the selection of a particular analysis algorithm is applied to a specific video camera soon thus, and is applied to a specific period in future.
Figure 173 is an example to a hard copy print screen that excites " printing " switch region that is shown in Figure 161 to respond to be produced.Among Figure 173 this shown hard copy printout comprise may be a colour and window 2616 that may represent to be displayed on Figure 161 in a convergent-divergent of this image and/or an image 2640 of other processing sections.In addition, shown in 2642, provide and represented that this image is produced and be printed the time of output, the heading message of this eikongen or the like.
The front is mentioned, and this preferred video data file form also will be preserved video data and produced simultaneously voice data on disk.The input that one or more microphone (not being shown) or other sources of audio information can be used as this a VR/PC unit or an one revision connects.After being quantized, this audio-frequency information will be stored on the disk, and what be saved simultaneously also has a relevant video data.In this VR/PC unit will have band or come a connected loudspeaker (not being shown) as a peripheral hardware, with the audio-frequency information that uses the form that can listen to regenerate and from this disk, be retrieved out.
Also can be designed to, in this VR/PC unit, provide audio content analytical algorithm (" instrument ") to detect the incident that to listen, for example open the door/close the door footsteps or the like.This audio instrument may be used as " scene " or operate as " search " instrument, promptly is applied to the voice data that is retrieved out and is applied to this video described above " instrument " similarly from disk.For example, can use a live audio instrument or a plurality of instrument to come the detection incident and excite alarm conditions.Can also use audio instrument each other logical combination and the logical combination between this audio instrument and the video frequency tool.As an example, under the detection of the light that lacks a video frequency tool, an audio instrument detects footsteps and will excite alarm conditions.Detect an audio event and can trigger the application of a video frequency tool, and/or detect the application that a Video Events can trigger an audio instrument.
Preferably, can use with mode like top " video frequency tool case " feature class that in Figure 170-172 and 153-156, illustrates and select and dispose audio instrument.Parameter that the user can select, that retrain the execution of an audio instrument comprises sound size, sound frequency, direction or the like.Because it is low that audio surveillance has expense, the data reserve capacity is less relatively, comprehensive and with features such as the brightness of environment is irrelevant, so monitor to video monitor it is a more valuable implementation, perhaps can substitution video monitor in many application sound intermediate frequencies.Simultaneously, the value that is saved data of using information management thought disclosed herein can improve audio surveillance widely and be produced.
Should be appreciated that, above disclosed embodiments of the present invention be to be used for schematically being not by way of limitation.

Claims (15)

1. analyze the method that a series of frames of video information detect the change of described frames of video information for one, the method comprising the steps of:
In an image plane, show step by an image of a represented target of described frames of video information;
Operate a graphic image element with the zone of selection, thereby specify the step of monitored object corresponding to the image plane of this target;
Select the step of the part of described appointed target; And
In the described frames of video information of continuous review, judge with the corresponding part of the described chosen part of described designated target whether described designated target rests on the step in a precalculated position in the described frames of video information.
2. method as claimed in claim 1, the step of the described target of wherein said appointment comprise a border that defines a described target.
3. method as claimed in claim 2, the described chosen part of wherein said designated target are the parts away from the described target at the described boundary position place of described target.
4. video data save set comprises:
Be used for receiving the device of video data of an image of an image plane of expression;
Be used to select not need to preserve the drop device of a part of the image plane of its data; With
Save set responds to described drop device, be used for preserving the video data of described reception, not with the corresponding so a part of video data of the described chosen part of described image plane;
Display device is used to show and the corresponding image of being represented by the video data of described reception of an image;
In the video data of described reception, be not stored in described this save set with the corresponding part of the described chosen part of described image plane.
5. video data save set as claimed in claim 4, wherein said drop device comprise and are used for the plotting unit of picture element of described part that on the described image that is shown by described display device expression of stack does not need to preserve the image plane of its data.
6. video data save set as claimed in claim 5, wherein said plotting unit comprise a mouse.
7. video data save set as claimed in claim 4 further comprises a video camera that is used to produce the video data that is received by described receiving trap.
8. video data save set as claimed in claim 4, wherein said save set comprises the recording medium of the part of the video data that a record on it is not selected by described drop device, described recording medium is from by a hard disk, chooses in the group of a magneto-optic disk and a compact disc rocordable composition.
9. preserve the method for video data, comprise step:
Receive the video data of an image in image plane of expression;
Not need to select to preserve the part of the image plane of its data; With
On a recording medium, only preserve those not with the part of the corresponding receiving video data of described chosen part of described image plane.
10. method as claimed in claim 9, wherein said selection step comprise that a pattern primitive that operates on the display screen usually represents the chosen part of this image plane.
11. an apparatus for processing of video signals, this apparatus for processing of video signals comprises:
Be used for receiving the device of a vision signal of an image that is illustrated in an image plane;
Be used to select the part of image plane to improve the selecting arrangement of contrast; With
Improve device, link to each other with described receiving trap, described selecting arrangement is responded, be used to expand a dynamic range with the part of the corresponding described receiving video signals of described chosen part of image plane, and be used to export one and optionally improve vision signal.
12. apparatus for processing of video signals as claim 11, further comprise the display device that is used to show a corresponding image of image of representing by described receiving video signals, wherein said selecting arrangement comprises plotting unit, and described plotting unit is used on the described image shown by described display device the picture element of described part that expression of stack selects to improve the image plane of contrast.
13. as the apparatus for processing of video signals of claim 11, wherein said selecting arrangement comprises the device that links to each other with described receiving trap, is used to detect a contrast metric of described receiving video signals.
14. apparatus for processing of video signals as claim 11, it is a simulating signal that wherein said selectivity is improved vision signal, and further comprise the A/D conversion equipment, be used for that described selectivity is improved vision signal and be converted to described selectivity and improve the corresponding video data of vision signal.
15., further comprise save set as the apparatus for processing of video signals of claim 14, link to each other with described A/D conversion equipment, be used to preserve described video data.
CN99123164A 1996-10-31 1999-10-26 Intelligent management system for video frequency information Expired - Fee Related CN1110779C (en)

Applications Claiming Priority (21)

Application Number Priority Date Filing Date Title
US08/741,715 1996-10-31
US08/729,620 1996-10-31
US08/740,628 1996-10-31
US08/740,627 US5884042A (en) 1996-10-31 1996-10-31 Data identification in an intelligent video information management system
US08/729,620 US5822542A (en) 1996-10-31 1996-10-31 Electronic and structural components of an intelligent video information management apparatus
US08/741,914 US5828848A (en) 1996-10-31 1996-10-31 Method and apparatus for compression and decompression of video data streams
US08/740,628 US5917958A (en) 1996-10-31 1996-10-31 Distributed video data base with remote searching for image data features
US08/741,983 1996-10-31
US08/740,627 1996-10-31
US08/741,983 US5854902A (en) 1996-10-31 1996-10-31 Video data capture and formatting in intelligent video information management system
US08/742,017 1996-10-31
US08/742,015 1996-10-31
US08/742,015 US5909548A (en) 1996-10-31 1996-10-31 Apparatus for alerting human operator to status conditions of intelligent video information management system
US08/740,651 US5862342A (en) 1996-10-31 1996-10-31 Intelligent video information management system with information archiving capabilities
US08/741,650 US6031573A (en) 1996-10-31 1996-10-31 Intelligent video information management system performing multiple functions in parallel
US08/741,650 1996-10-31
US08/741,982 1996-10-31
US08/741,982 US5875305A (en) 1996-10-31 1996-10-31 Video information management system which provides intelligent responses to video data content features
US08/740,651 1996-10-31
US08/741,914 1996-10-31
US08/741,715 US5875304A (en) 1996-10-31 1996-10-31 User-settable features of an intelligent video information management system

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
CNB971992215A Division CN1145122C (en) 1996-10-31 1997-10-01 Intelligent video information management system

Publications (2)

Publication Number Publication Date
CN1271146A CN1271146A (en) 2000-10-25
CN1110779C true CN1110779C (en) 2003-06-04

Family

ID=27581314

Family Applications (3)

Application Number Title Priority Date Filing Date
CN99123164A Expired - Fee Related CN1110779C (en) 1996-10-31 1999-10-26 Intelligent management system for video frequency information
CN99123611A Expired - Fee Related CN1109323C (en) 1996-10-31 1999-10-28 Intelligent management system for video and audio information
CN99123610A Expired - Fee Related CN1110780C (en) 1996-10-31 1999-10-28 Intelligent management system for video and audio information

Family Applications After (2)

Application Number Title Priority Date Filing Date
CN99123611A Expired - Fee Related CN1109323C (en) 1996-10-31 1999-10-28 Intelligent management system for video and audio information
CN99123610A Expired - Fee Related CN1110780C (en) 1996-10-31 1999-10-28 Intelligent management system for video and audio information

Country Status (1)

Country Link
CN (3) CN1110779C (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100988257B1 (en) * 2002-09-06 2010-10-18 소니 주식회사 Information processing device and a computer-readable recording medium recording thereon a program
CN102264003B (en) * 2010-05-27 2013-10-09 华为软件技术有限公司 Realization method for lowering code rate as well as relevant equipment and system
CN104754421A (en) * 2014-02-26 2015-07-01 苏州乐聚一堂电子科技有限公司 Interactive beat effect system and interactive beat effect processing method
US11670147B2 (en) * 2016-02-26 2023-06-06 Iomniscient Pty Ltd Method and apparatus for conducting surveillance

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4308559A (en) * 1979-05-14 1981-12-29 Peter Schiff Switching apparatus for closed circuit television monitoring systems
US4857912A (en) * 1988-07-27 1989-08-15 The United States Of America As Represented By The Secretary Of The Navy Intelligent security assessment system
US4918523A (en) * 1987-10-05 1990-04-17 Intel Corporation Digital video formatting and transmission system and method
CN2063256U (en) * 1990-03-12 1990-10-03 庄志荣 High-resolution display controller for chinese characters and graphics
CN1046651A (en) * 1989-04-20 1990-10-31 汤姆森消费电子有限公司 The centering of Inset picture in many pictures system
US5099322A (en) * 1990-02-27 1992-03-24 Texas Instruments Incorporated Scene change detection system and method
US5119205A (en) * 1963-03-11 1992-06-02 Lemelson Jerome H Methods and apparatus for scanning and analyzing selected images areas
CN1062791A (en) * 1991-11-30 1992-07-15 浙江大学 Movement objective orbit image analytical method
CN1066548A (en) * 1991-05-06 1992-11-25 汤姆森消费电子有限公司 Television receiver with " picture-in-picture " and Nonlinear Processing
US5229862A (en) * 1990-11-16 1993-07-20 Sony Corporation Apparatus and method for recording image signals to allow reproduction at more than one data rate

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR2648594B1 (en) * 1989-06-15 1993-03-05 Omega Electronics Sa APPARATUS FOR TIMING COURSES
KR0145007B1 (en) * 1992-07-31 1998-07-15 강진구 Video library system and realization method thereof
US5463565A (en) * 1993-10-29 1995-10-31 Time Warner Entertainment Co., L.P. Data block format for software carrier and player therefor
KR0129792B1 (en) * 1993-11-16 1998-04-11 배순훈 Cctv camera for zoom type

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5119205A (en) * 1963-03-11 1992-06-02 Lemelson Jerome H Methods and apparatus for scanning and analyzing selected images areas
US4308559A (en) * 1979-05-14 1981-12-29 Peter Schiff Switching apparatus for closed circuit television monitoring systems
US4918523A (en) * 1987-10-05 1990-04-17 Intel Corporation Digital video formatting and transmission system and method
US4857912A (en) * 1988-07-27 1989-08-15 The United States Of America As Represented By The Secretary Of The Navy Intelligent security assessment system
CN1046651A (en) * 1989-04-20 1990-10-31 汤姆森消费电子有限公司 The centering of Inset picture in many pictures system
US5099322A (en) * 1990-02-27 1992-03-24 Texas Instruments Incorporated Scene change detection system and method
CN2063256U (en) * 1990-03-12 1990-10-03 庄志荣 High-resolution display controller for chinese characters and graphics
US5229862A (en) * 1990-11-16 1993-07-20 Sony Corporation Apparatus and method for recording image signals to allow reproduction at more than one data rate
CN1066548A (en) * 1991-05-06 1992-11-25 汤姆森消费电子有限公司 Television receiver with " picture-in-picture " and Nonlinear Processing
CN1062791A (en) * 1991-11-30 1992-07-15 浙江大学 Movement objective orbit image analytical method

Also Published As

Publication number Publication date
CN1259721A (en) 2000-07-12
CN1109323C (en) 2003-05-21
CN1271146A (en) 2000-10-25
CN1110780C (en) 2003-06-04
CN1259720A (en) 2000-07-12

Similar Documents

Publication Publication Date Title
CN1123857C (en) Intelligent management system for video and audio information
US8166420B2 (en) Electronic apparatus and image display control method of the electronic apparatus
CN1741178B (en) Reproducing apparatus
CN100409228C (en) Stand alone printer with hardware / software interfaces for sharing multimedia processing
CN1777950A (en) Reproduction apparatus, program, integrated circuit
CN101094397A (en) Video monitor system, and video monitoring program
US20100037179A1 (en) Display control apparatus, display control method and program
JP2008192013A (en) Display image control unit and its control method
CN1110779C (en) Intelligent management system for video frequency information
CN1110778C (en) Intelligent management system for video and audio information
JP4279812B2 (en) Digital data playback device
JP2009140453A (en) Information processor and method, and program
JP2007165983A (en) Metadata automatic generating apparatus, metadata automatic generating method, metadata automatic generating program, and recording medium for recording program
JP4609711B2 (en) Image processing apparatus and method, and program
JP2002175298A (en) Data management system, data management method and program
CN1645355A (en) Printer and network printing system which have embedding function for printing time-base media
JP2001216770A (en) Information processing device, method, and recording medium
JP2004180340A (en) Video recording apparatus, video reproducing apparatus, video recording method, and video reproducing method
JP2001215875A (en) Device and method for processing information, and recording medium
TH2723A3 (en) "Digital image recording security system"
TH2723C3 (en) "Digital image recording security system"

Legal Events

Date Code Title Description
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C06 Publication
PB01 Publication
C14 Grant of patent or utility model
GR01 Patent grant
C19 Lapse of patent right due to non-payment of the annual fee
CF01 Termination of patent right due to non-payment of annual fee