Background technology
Video monitoring is the important component part of safety-protection system, can reduce the cost that society keeps the peace effectively by video monitoring, and can reduce the factor of artificial disturbance, greatly improves the efficient of monitoring and public security.Development along with the communication technology and digital technology, the video monitoring industry is also taking place to change, simulation monitoring, digital supervision have been experienced in the video monitoring technical development, be at present from the transition stage of traditional DVR (Digital VideoRecorder, digital video recorder) mode to the network monitoring system development.Technology such as the front end video acquisition of video monitoring system, video transmission, videograph, control, demonstration all are among the fast development, and growing demand has also been brought new opportunities and challenges for the development of video monitoring technology.
The development trend of current video monitoring safety-protection system mainly shows as: 1, digitlization, networking, intellectuality, integrated; 2, progressively establish and improve the security protection system of scientific and technical innovation; 3, greatly develop 4C technology (communication/computer/control/multimedia display technology) and combined with intelligent building system and intelligent transportation system; 4, network technology and multiple security and guard technology are implemented integrated; 5, create and improve comprehensive safety service network at different levels, social security Dynamic Monitoring System, city pre-alarming system and safety urban construction.
Video monitoring system mainly contains functions such as quick candid photograph, real-time storage, taking turn supervision, program management and alarm linkage output.At present, the candid photograph technology of using in the video monitoring system mainly is divided into two classes: a class is to adopt external trigger mechanism control video camera such as induction coil to capture; An other class is to take out from the storage hard disk according to the image that motion detection technique is taken video camera to analyze, or based on the application (being mainly used in traffic tests the speed) of a kind of Measuring Object movement velocity of image motion detection technique.
The scheme specificity that adopts relay induction control camera shutter to capture is more intense, generally needs to adopt special circuit to control, and sensing device need be placed in the applied environment.This mode mainly is applicable to velocity variations ratio capturing system faster, thereby is mainly used in the overspeed snapping in the traffic control system.This candid photograph mode need be based on special digital camera, so capturing image separates with video monitoring image, capture image and can not carry out playback and centralized stores in real time, and system often needs to bury underground other auxiliary equipment such as induction coil, cost is than higher.
Second kind of scheme of capturing according to motion detection technique is a kind of application in the traditional video surveillance system, utilizes software to realize.This scheme is primarily aimed at video image that DVR plays in real time and captures and be converted into static images, or the video record of DVR hard disk stored is accessed broadcasts of decoding, and candid photograph is converted into static images as required.The image that adopts this scheme to capture has often passed through encoder compresses, and certain loss is arranged on image fidelity.When image change when very fast, the static images of candid photograph is second-rate.And a kind of often fixed camera of this traditional video surveillance system, can only catch the image change in the fixed range, can not follow the tracks of candid photograph to mobile object; Perhaps need under artificial the interference, rotate by control (passing through the monitor staff) camera, therefore not be suitable for the real-time monitoring of extensive supervisory control system.
Summary of the invention
Main purpose of the present invention provides a kind of video monitoring equipment, make and capture image and video monitoring image is separated and the mode of utilizing motion detection technique to capture makes the problem that picture quality is relatively poor to solve in the prior art mode of carrying out real-time grasp shoot, guarantee to capture the image after picture quality and convenient management are captured based on special digital camera.
Another object of the present invention provides a kind of video frequency monitoring method, with the real-time grasp shoot of realization to mobile object, and the quality of assurance image, simplify capturing the management of back image.
For this reason, the invention provides following technical scheme:
A kind of video monitoring equipment, described equipment comprises:
Video acquisition unit is used to gather video data, obtains the real-time video monitoring image;
System control unit is used for other unit transmitting control commands to equipment;
Image is captured and processing unit, and the real-time video monitoring image that the candid photograph strategy that is used for issuing according to system control unit obtains video acquisition unit is captured and encoding process;
Network Transport Element, be used for according to the order of system control unit image is captured and processing unit processes after video data be sent to network.
Preferably, described equipment also comprises:
Motion detection unit, link to each other with image candid photograph and processing unit with video acquisition unit respectively, be used for carrying out locomotion evaluation and detection, and testing result is sent to system control unit according to the real-time video monitoring image that the order of system control unit is obtained video acquisition unit;
The video acquisition control unit is used for according to collection and the input to video of the order control of video collecting unit of system control unit.
Described video acquisition unit comprises:
Camera is used to obtain the real-time video monitoring image of monitoring target;
The Cloud Terrace links to each other with the video acquisition control unit, is used for fixing camera, and adjusts the angle of camera automatically according to the control of video acquisition control unit.
Preferably, described equipment also comprises:
User interface section links to each other with described system control unit, is used to import user instruction, sets and captures strategy and monitoring parameter.
Described Network Transport Element by multicast or mode of unicast image is captured and processing unit processes after video data send in the network storage equipment.
Preferably, described Network Transport Element is integrated with internet small computer system interface, and utilizes this interface and the described network storage equipment to carry out data interaction.
A kind of video frequency monitoring method, described method comprises step:
A, obtain the real-time video monitoring image;
B, the predetermined candid photograph strategy of basis are captured and encoding process described real-time video monitoring image;
C, the picture captured and the real-time video monitoring image behind the coding are sent in the network.
Preferably, between steps A and step B, also comprise:
The present image that video acquisition unit is obtained carries out locomotion evaluation and detection, and the mode of real-time video monitoring image is obtained in adjustment according to testing result.
Alternatively, described predetermined candid photograph strategy comprises:
Carry out the dynamic image locking according to following the tracks of the object size variation; And/or
Catch the sensitizing range in the scanning area; And/or
Catch and the park scan zone in the two mobile object; And/or
A plurality of mobile objects in the selected locking scanning area.
Alternatively, described step C is specially:
By multicast or mode of unicast the picture of candid photograph and the real-time video monitoring image behind the coding are sent in the network storage equipment.
By above technical scheme provided by the invention as can be seen, the present invention is directed to the demand of in the security protection application monitored object being carried out real-time grasp shoot, video monitoring equipment of the present invention is at first gathered video data by video acquisition unit, obtain the real-time video monitoring image, this real-time video monitoring image is captured and encoding process according to predetermined candid photograph strategy by image candid photograph and processing unit then, realized finishing in the video monitoring encoder function of candid photograph.Because the original video monitoring image that the present invention adopts un-encoded to handle is captured, thus the quality that has guaranteed to capture image.Further, before the original video monitoring image is captured, it is carried out locomotion evaluation and detection, and capture and follow the tracks of processing according to the associated components in the testing result control of video collecting unit, to catch and to lock monitored object more clearly, improve monitoring effect.On the other hand, the present invention also utilizes the characteristics of Network Video Surveillance encoder, and image after capturing and the data behind the coding are saved in the network storage equipment, has realized centralized stores and management.
The present invention not only can be applied to systems such as social security dynamic monitoring, city early warning, and can be applied in the traffic control system, makes it not need to bury underground other auxiliary equipment such as induction coil, reduces cost.
Embodiment
Core of the present invention is to gather video data by video acquisition unit, obtain the real-time video monitoring image, this real-time video monitoring image is captured and encoding process according to predetermined candid photograph strategy by image candid photograph and processing unit then, realized finishing in the video monitoring encoder function of candid photograph.Further, before the original video monitoring image is captured, it is carried out locomotion evaluation and detection, and capture and follow the tracks of processing according to the associated components in the testing result control of video collecting unit, to catch and to lock monitored object more clearly, improve monitoring effect.In addition, the present invention also utilizes the characteristics of Network Video Surveillance encoder, and image after capturing and the data behind the coding are saved in the network storage equipment, and it is carried out centralized stores and management.
In order to make those skilled in the art person understand the present invention program better, the present invention is described in further detail below in conjunction with drawings and embodiments.
With reference to Fig. 1, Fig. 1 shows the theory diagram of present device first embodiment:
This equipment comprises: video acquisition unit S01, image candid photograph and processing unit S02, Network Transport Element S03, system control unit S04, user interface section S05.
Wherein, system control unit S04 is used for other unit transmitting control commands to equipment; User interface section S05 links to each other with system control unit S04, is used to import user instruction, sets and captures strategy and monitoring parameter; Video acquisition unit S01 is used to gather video data, obtains the real-time video monitoring image; The real-time video monitoring image that the candid photograph strategy that image is captured and processing unit S03 is used for issuing according to system control unit obtains video acquisition unit is captured and encoding process; Memory cell S03 is used for the picture of memory image processing and motion detection unit candid photograph and the real-time video monitoring image behind the coding.
System control unit S04 is the core control part of video monitoring equipment of the present invention, is connecting each critical piece of video monitoring equipment.The user passes through the user interface input instruction, by system control unit other unit transmitting control commands and configuration order to terminal equipment, such as, send candid photograph strategy etc. to image candid photograph and processing unit S02.Certainly, in not needing the monitoring environment that often manually is provided with and adjusts, also user interface can be set, but with these demanded storage in system control unit S04, make it according to other unit in the fixing mode control appliance.
Video acquisition unit comprises common components such as camera and The Cloud Terrace, the vision signal in the guarded region is carried out processing such as data acquisition, A/D conversion.The data that obtain from video acquisition unit are initial data of un-encoded, and are therefore more true to nature than the data that obtain after encoding through image processing.
Therefore, present device is carried out buffer memory with the original video data of video acquisition input unit output, the image that these original video data form is captured according to the candid photograph strategy that system control unit S04 issues by image candid photograph and processing unit S02, when the satisfied candid photograph strategy of image change requires, original image is preserved.Capturing strategy can be one or more following combinations:
Carry out the dynamic image locking according to following the tracks of the object size variation; Catch the sensitizing range in the scanning area; Catch and the park scan zone in the two mobile object; A plurality of mobile objects in the selected locking scanning area etc.
Image is captured and processing unit S02 also needs original video data is carried out encoding process simultaneously, that is to say that the data of its output not only comprise the data flow of the picture after the candid photograph, but also comprises real-time monitor video data flow.These data flow send among the network storage equipment S10 by Network Transport Element S03.
Network Transport Element S03 is used for realizing the Network Transmission and the storage of video data, different with the DVR of traditional video surveillance, this unit sends video data and receive by modes such as multicast or clean cultures, such as, play in real time, perhaps send in the network storage equipment and store.Simultaneously, this unit also is integrated with iSCSI (Internet Small Computer System Interface, a kind of SCSI host-host protocol) module based on the internet, by the iSCSI agreement video data is stored and read in the IP network storage system, can also receive the control command of Network Management Equipment simultaneously.
The present technique field personnel know, the iSCSI agreement is a hottest current network storage agreement, can solve the SCSI agreement the number of devices that connects be connected apart from aspect restriction.The user can be as the storage resources that uses any position on the local hard disk accesses network, also the storage resources of original old dispersion can be integrated into a big storage pool.Growth along with the storage application, storage networking often needs to cross over far distance, because iSCSI is in the advantage of aspects such as flow congestion controlling mechanism, discovery and addressing mechanism, overtime retransmission mechanism, security mechanism, it can be applicable in safety-protection system the storage to video monitoring data well.
As seen, video monitoring equipment of the present invention has not only guaranteed to capture the quality of image, but also the image after capturing can be saved in the network storage equipment, has made things convenient for the storage and the management of these data.
In order to trace and monitor object better, and obtain video image more clearly.Motion detection unit and video acquisition control unit can also be set in video monitoring equipment of the present invention.
The theory diagram of present device second embodiment as shown in Figure 2:
Wherein, motion detection unit S06 captures with image with video acquisition unit S01 respectively and processing unit S02 links to each other, be used for carrying out locomotion evaluation and detection, and testing result is sent to system control unit according to the real-time video monitoring image that the order of system control unit S04 is obtained video acquisition unit.Such as, in the equipment application process, the user is provided with own interested and responsive guarded region by system control unit, and regulates some threshold values of carrying out motion detection.Motion detection unit S06 monitors and follows the tracks of guarded region according to these crucial monitoring parameters of user's input.The user can be provided with a plurality of monitoring sensitizing range, and for entering these regional moving object, motion detection unit S06 can monitor and assess by certain method for testing motion, then these testing results is returned to system control unit.
Video acquisition control unit S07 links to each other with system control unit S04 with video acquisition unit S01 respectively, is used for according to collection and the input to video of the order control of video collecting unit of system control unit.System control unit S04 can send control commands corresponding to video acquisition control unit S07 according to the testing result that motion detection unit S06 returns, and video acquisition control unit S07 controls the associated components among the video acquisition unit S01 according to these orders.
The front is mentioned, and video acquisition unit comprises common components such as camera, photoflash lamp and The Cloud Terrace, the vision signal in the guarded region is carried out processing such as data acquisition, A/D conversion.The Cloud Terrace is the fastening of installation, fixing camera, is divided into usually fixing and two kinds of electric platforms.Fixedly The Cloud Terrace is applicable to the situation that monitoring range is little, can adjust the level of camera and the angle of pitching install camera on fixing The Cloud Terrace after; Electric platform is applicable to that it can enlarge the monitoring range of camera to carrying out scanning monitoring on a large scale.Electric platform is realized by two operating motors that usually motor accepts the signal of self-controller accurately to move the location.Under the effect of control signal, but both automatic scan monitor area of the camera on the The Cloud Terrace also can be traced and monitored object under the operator on duty's of Surveillance center manipulation.
In present device, video acquisition control unit S07 is according to the order of system control unit S04, can focus or control such as automatic mobile The Cloud Terrace to camera, thereby make it carry out detail scans to monitored object.Such as, adjust The Cloud Terrace motion and attitude according to the high priority monitoring objective movement velocity and the direction of motion, focus on apart from far and near and monitoring priority according to it etc.
The relativeness that motion detection unit S06 can further judge mobile object and guarded region according to the variation of focusing back image (for example, be just to enter or leave), system control unit is captured strategy according to this testing result adjustment, focuses on and follows the tracks of and capture such as taking for the target that has just entered; Then take to reduce monitoring priority, abandon tracking etc. for the target of leaving.
Present device can be applied in the systems such as social security dynamic monitoring, city early warning, traffic administration.
With reference to Fig. 3, Fig. 3 shows the flow chart of a preferred embodiment of the inventive method, may further comprise the steps:
Step 301: obtain the real-time video monitoring image.
Step 302: present image is carried out locomotion evaluation and detection.
Such as, can adopt background subtraction (Background Subtraction) and time difference (TemporalDifference) method that the motion object is detected.The original analog vision signal is converted to YUV (the another kind of method for expressing in image pixel space of ITU 656 standards (International Telecommunications Union about broadbands such as optical fiber, cables on the some of the recommendations standard of digital data transmission interface) through video acquisition unit A/D, YCbCr, luminance component Y and chromatic component Cb, Cr, because human eye is relatively more responsive for luminance component Y, so often use the Y of this pixel space to carry out motion detection and estimation etc., be fit to image compression and Network Transmission) form, detect the main luminance component (Y) the most responsive to human eye perceives and handle.(x is y) with the corresponding T of background surveyed area for each picture element during to moment T in the surveyed area
iPicture element is got poor:
Mx,y(T
i-T)=|Yx,y(T)-Yx,y(T
i)| (1)
Wherein, Yx, y (T) are T picture element (x, luma component values y) (Y), Yx, y (T constantly
i) be T
iMoment picture element (x, luma component values y) (Y).
T is to T in the then whole surveyed area D
iDifference constantly is:
M
D(T
i-T)=∑
(V(x,y)∈D)Mx,y(T
i-T) (2)
If M
D(T
i-T)>and F, wherein F is predefined detection threshold (can define adjusting by the keeper), then sends the warning that motion detection surpasses threshold value to system.
Locomotion evaluation also is an estimation.At present the method for locomotion evaluation commonly used has overall motion estimation, the estimation based on picture element, block-based estimation and based on the estimation in zone.That block matching motion is estimated is simple because of algorithm, be convenient to hardware realizes being used widely.
(u v) judges, if motion vector surpasses the surveyed area D that pre-sets, then can control by sending movement instruction to The Cloud Terrace according to the motion vector Mv of estimation.It is concrete that kinematic parameter can (u v) reaches the time difference (T between the picture frame according to motion vector Mv
i-T) to calculate, the direction of motion also can (u v) comes to determine by motion vector Mv.Thereby reach the effect of capturing from motion tracking.
Step 303: judge whether motion detection result surpasses preset threshold.If surpass, then enter step 304; Otherwise, enter step 310.
Step 304: select suitable candid photograph strategy.
The front is mentioned, and capturing strategy can be one or more following combinations:
Carry out the dynamic image locking according to following the tracks of the object size variation; Catch the sensitizing range in the scanning area; Catch and the park scan zone in the two mobile object; A plurality of mobile objects in the selected locking scanning area etc.Can select for use according to applied environment, and can in monitor procedure, select.
Step 305: judge whether the needs candid photograph according to the monitoring strategies of selecting.If desired, then enter step 306; Otherwise, enter step 309.
Step 306: locomotion evaluation and testing result are done further divisional processing and analysis.
Because, need watch-dog quick lock in monitored object in case satisfy the candid photograph condition.But in order to obtain the better pictures quality, can also do further divisional processing and analysis to locomotion evaluation and testing result, obtaining better to capture parameter, such as, parameters such as the angle of camera, focal length.
Step 307: judge whether that needing to adjust The Cloud Terrace, focal length etc. dynamically captures parameter.If desired, then enter step 308; Otherwise, enter step 309.
Step 308: adjust and dynamically capture parameter.Then, enter step 309.
Step 309: the raw image data in the buffer memory is carried out special processing, that is to say, capture required monitoring image.
Step 310: the raw image data in the buffer memory is carried out conventional encoding process, and carry out follow-up flow processs such as the network storage.Then, return step 301, continue monitored object is monitored in real time.
The video data behind the coding and the view data of candid photograph can be sent in the network by modes such as multicast or clean cultures respectively, play in real time or store.But also can be undertaken reading these data of storage alternately by the iSCSI agreement and the network storage equipment.
As seen, utilize the present invention, not only can guarantee to capture the quality of image, but also can be to storage in image after capturing and the real time video data adfluxion and management.
Though described the present invention by embodiment, those of ordinary skills know, the present invention has many distortion and variation and do not break away from spirit of the present invention, wish that appended claim comprises these distortion and variation and do not break away from spirit of the present invention.