CN101299812B - Method, system for analyzing, storing video as well as method, system for searching video - Google Patents
Method, system for analyzing, storing video as well as method, system for searching video Download PDFInfo
- Publication number
- CN101299812B CN101299812B CN 200810115569 CN200810115569A CN101299812B CN 101299812 B CN101299812 B CN 101299812B CN 200810115569 CN200810115569 CN 200810115569 CN 200810115569 A CN200810115569 A CN 200810115569A CN 101299812 B CN101299812 B CN 101299812B
- Authority
- CN
- China
- Prior art keywords
- moving target
- video
- information
- module
- image
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000001514 detection method Methods 0.000 claims abstract description 74
- 238000004458 analytical method Methods 0.000 claims abstract description 39
- 239000012634 fragment Substances 0.000 claims description 56
- 239000000284 extract Substances 0.000 claims description 18
- 238000000605 extraction Methods 0.000 claims description 16
- 230000002123 temporal effect Effects 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 7
- 230000013011 mating Effects 0.000 claims description 2
- 238000012544 monitoring process Methods 0.000 abstract description 4
- 238000012545 processing Methods 0.000 abstract description 2
- 230000003068 static effect Effects 0.000 description 11
- 238000005516 engineering process Methods 0.000 description 5
- 238000004321 preservation Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 3
- 238000004040 coloring Methods 0.000 description 2
- 230000002950 deficient Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 239000012467 final product Substances 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 238000010223 real-time analysis Methods 0.000 description 1
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention discloses a video analysis and storage method, a system and a video detection method and a system, respectively saving the storage space of the video image and increasing the efficiency of the video search, wherein the analysis and storage method and the system analyze the video image and store the motion segment containing the motion target, while the video detection method and the system analyze the video image and store motion segment containing the motion target, and the related motion target information and the motion segment information, search the stored motion segment according to the information and the index term input by the users. The invention can be widely applied in the video processing system, particularly the video monitoring system.
Description
Technical field
The present invention relates to image processing techniques, relate in particular to a kind of video analysis and storage means, system, and a kind of video retrieval method, system.
Background technology
The mode of existing video monitoring system store video images, generally be under camera is recorded real-time video, intactly be kept in the database.For general video surveillance applications scene, such as warehouse, factory building etc., in some periods such as nights, the situation that the scene appearance of being monitored changes is less, seldom has moving object to get in these monitored spaces.For the application scenarios that adopts static camera to monitor, the scene picture generally all is static, that is to say, image can keep same picture for a long time and can not produce comparatively obvious variation.
Prior art all is saved in video image all the time in the memory devices such as hard disk, video tape in real time, and in order to application such as later retrieval Hui Cha, but this needs bigger memory space.Though development along with memory technology; Jumbo memory device can address this problem preferably; But in view of view data itself just has the more defective of the byte of taking; Even if therefore regardless of expense adopts jumbo memory device, also be difficult to preserve all video images in longer a period of time.
Prior art carry out video content retrieve back look in, the mode of looking into is returned in many employings manually, also promptly adopt to play the video of being preserved, and video is carried out artificial judgment, looks into the mode inefficiency and occurs careless omission easily for this time.Return when looking into manually; The speed of video playback can not be too fast; Otherwise a target object sudden strain of a muscle is probably promptly crossed and is not noted, so one section video of 10 hours, and normal speed is returned the words of looking into from the beginning to the end just needs about 10 hours; And significant probably video content has only several seconds or tens seconds, so efficient is very low.And return for manual work and to look into, to return and look into personnel and watch for a long time in the video process, attentiveness can not keep long-time to be concentrated, and is easy to produce tired, looks into even if therefore return by normal speed, also occurs careless omission probably and misses relevant information.
Also there is same problem in video image for other application scenarioss that adopt static camera to take obtain.
Summary of the invention
Technical problem to be solved by this invention provides the method and system of a kind of video analysis and storage, to practice thrift the memory space of video image.
In order to solve the problems of the technologies described above, the present invention at first provides a kind of video analysis and storage system, and comprise detection module, preserve module and memory module, wherein:
Said detection module carries out motion detection to video image, judges that whether including the object that is kept in motion in the video image is moving target;
Said preservation module links to each other with said detection module, and the sequence of video images that contains moving target in the said video image is saved in said memory module;
Said memory module links to each other with said preservation module, stores said sequence of video images.
In aforesaid video analysis and the storage system, said detection module can further carry out target detection to said moving target, obtains the type of said moving target.
Further, said detection module carries out target detection to said moving target, can comprise carrying out the detection of people's face, human detection and/or vehicle detection.
The present invention also provides a kind of video analysis and storage means, comprises step:
Video image is carried out motion detection, judge that whether including the object that is kept in motion in the video image is moving target;
Preserve the sequence of video images that contains moving target in the said video image.
Aforesaid video analysis and storage means may further include said moving target is carried out target detection, obtain the step of the type of said moving target.
Further, said moving target is carried out target detection, can comprise said moving target is carried out the detection of people's face, human detection and/or vehicle detection.
In aforesaid video analysis and the storage means, preserve the step that contains the sequence of video images of moving target in the said video image, can comprise:
When judging said video image and containing moving target, begin to store said video image, continue said video image is carried out motion detection;
When judging said video image and remaining static, stop to store said video image;
Preserve the sequence of video images that contains moving target in the said video image.
Another technical problem to be solved by this invention provides a kind of video retrieval method and system, to improve the efficient of video frequency searching.
In order to solve this technical problem, the present invention at first provides a kind of video frequency search system, comprises detection module, tracking module, characteristics analysis module, memory module and retrieval module, wherein:
Said detection module carries out motion detection and target detection to the video image of importing, and catching the object that is kept in motion in the said video image is moving target;
Said tracking module links to each other with said detection module, after said detection module captures said moving target, discerns and follow the tracks of said moving target, returns to inactive state up to detecting said video image, stops to follow the tracks of said moving target;
Said characteristics analysis module links to each other with said tracking module, is used to obtain the moving target information of said moving target, and the sequence of video images that contains said moving target is the sports segment information;
Said memory module links to each other with said characteristics analysis module, is used to preserve said moving target information and sports segment information;
Said retrieval module links to each other with said memory module, moving target information and/or the sports segment information preserved is retrieved the output result for retrieval according to the search terms of user's input.
In the aforesaid video frequency search system; Said characteristics analysis module can be obtained the type and the image of moving target from said tracking module; To characteristic information and/or the movable information that obtains said moving target after the graphical analysis, can obtain the temporal information and/or the identification information of said motion fragment simultaneously from said tracking module.
Further; When said characteristics analysis module is obtained a plurality of moving target from said tracking module; Can select the image of part or all of moving target to resolve; Extract the characteristic information and/or the movable information of these moving targets, and obtain the temporal information and/or the identification information of the corresponding motion fragment of each moving target of selection.
And this system can also comprise video extraction module and video display module, wherein:
Said retrieval module further sends to said video extraction module with corresponding sports segment information when retrieving the motion fragment of mating with said search terms;
Said video extraction module; Link to each other with said detection module; Be used for according to the video extraction one or more image of said sports segment information from said motion fragment; Perhaps from said motion fragment, extract one or more snippets video, then the image or the video that extract are sent to said video display module;
Said video display module links to each other with said video extraction module, is used to show the video that the image received or broadcast are received.
And the result for retrieval of said retrieval module output can be sports segment information and/or the corresponding moving target information that matches according to search terms.
In the aforesaid video frequency search system, said characteristics analysis module can further be given said tracking module with said moving target feedback information, to improve the tracking efficient of said tracking module.
In the aforesaid video frequency search system, said system can also comprise moving target database and moving target identification module, wherein:
Said moving target database is preserved the characteristic information of moving target;
Said moving target identification module links to each other with said detection module, is used for obtaining from said detection module the image of the moving target of respective type, and discerns according to the characteristic information of this moving target in the said moving target database, and the output recognition result.
The present invention also provides a kind of video retrieval method, comprises step:
Video image to input carries out motion detection, and catching the object that is kept in motion in the said video image is moving target;
Through said moving target being carried out target detection and follow the tracks of, obtain and preserve the moving target information of said moving target and comprise the sports segment information of the i.e. motion fragment of sequence of video images of said moving target;
After said video image detects and finishes, said moving target information and/or the sports segment information preserved are retrieved according to the search terms of user's input, and the output result for retrieval.
In the aforesaid video retrieval method; Said moving target information can comprise one or more in type, characteristic information and the movable information of moving target, and said sports segment information can be the temporal information of motion fragment and in the identification information one or more.
Further; When in tracing process, comprising a plurality of moving target; Can select part or all of moving target; Extract the moving target information of these moving targets, to each moving target of selecting, with the motion fragment of the sequence of video images that has this moving target as this moving target correspondence.
And; Said result for retrieval can be one or more snippets video in the motion fragment that matches according to search terms; Perhaps can be the one or more image in the motion fragment that matches according to search terms, perhaps can be sports segment information and/or the corresponding moving target information that matches according to search terms.
And, be people's face like detected moving target, then can extract the image and the characteristic information thereof of this people's face, and can discern automatically this people's face according to the characteristic information of people's face in the face database of configuration, and the output recognition result; And/or
Like detected moving target is vehicle, then can extract the image and the characteristic information thereof of this vehicle, and can discern automatically this vehicle according to the characteristic information of vehicle in the vehicle data storehouse of configuration, and the output recognition result.
Compared with prior art, video analysis provided by the invention and storage means, system through preserving the motion fragment, have effectively saved memory space, have practiced thrift cost.Video retrieval method provided by the invention, system return when looking at video frequency searching, through retrieval motion fragment, have accelerated video frequency searching significantly and have returned the speed of looking into, and have improved operating efficiency and retrieval rate, have reduced wrong report and have failed to report possibility.
Description of drawings
Fig. 1 is the module map of embodiment of the invention video analysis and storage system;
Fig. 2 is the flow chart of embodiment of the invention video analysis and storage means;
Fig. 3 is the module map of embodiment of the invention video frequency search system;
Fig. 4 is the flow chart of embodiment of the invention video retrieval method.
Embodiment
Specify execution mode of the present invention below in conjunction with accompanying drawing and embodiment, how the application technology means solve technical problem to the present invention whereby, and the implementation procedure of reaching technique effect can make much of and implement according to this.
In the video system that static camera is taken, it is comparatively common that the video image that obtains and store can not produce this situation of significant change for a long time.Fig. 1 shows the embodiment of a kind of video analysis and storage system, and this embodiment comprises detection module 110, preserves module 120 and memory module 130, wherein:
Preserve module 120; Link to each other with detection module 110; Judge at detection module 110 that to contain moving target in the video image also be to begin in the moment that image gets into motion state; From then on the video image that constantly begins is saved in the memory module 130, detects up to detection module 110 that not comprise moving target in the video image also be image when getting into inactive state, stop in memory module 130, preserving video image;
In the video image, begin when returning to inactive state, to finish when image gets into motion state from detecting, the consecutive image sequence that this section is contained movable information is referred to as a motion fragment;
Above-mentioned detection module 110 detects the situation that does not comprise moving target in the video image; Rest in the coverage of camera such as this moving target and keep static; Perhaps this moving target coverage of camera or the like of moving out of; Also promptly preserve module 120 when preserving video image, detection module 110 detects when not comprising the object that is kept in motion in the video image, preserves module 120 and promptly stops to preserve.Because the mode that the motion detection of video image adopts consecutive frame to compare more, so these situations are when being reflected in the video image, and this video image is to return to inactive state from motion state.
This analysis and storage system shown in Figure 1; Can be used to analyze and store the video of before having recorded; The real-time storage that can also be used for video system, also promptly this storage system video that camera is taken carries out real-time analysis and judgement, then storing moving fragment optionally.
Fig. 2 shows the flow process of video analysis and storage means, comprises the steps:
Step 210 is carried out motion detection to video image, judges whether include moving target in the video image, judges promptly that also image is kept in motion or remains static; Can carry out target detection to moving target, the type that identifies target is such as people or vehicle or the like;
Step 220 when getting into motion state from judging video image, is preserved video image, and is continued video image is carried out motion detection, judges whether image has returned to inactive state;
Relative prior art, the video analysis of present embodiment and the method and system of storage have effectively been saved memory space, have practiced thrift hardware cost.For the memory space of fixed size, the present invention can store more useful informations than prior art.Video return look in, directly read each motion fragment and carry out analysis and judgement and get final product, the user also need not to stand and watches the discomfort that video image brought that remains static for a long time, has improved video and has returned operating efficiency and the accuracy rate of looking into.Because the movable information in the video is all preserved, therefore also reduced and failed to report possibility.
It is an inevitable demand of present watch-dog development that the automation of video monitoring system is handled.Existing all kinds of video monitoring system; After accident takes place, generally can only come video returned and look into and analyze through video playback, very expend user time, and be prone to cause human fatigue work; Be prone to miss important information, cause mistake to fail to report defectives such as phenomenon is many, video recording data analysis difficulty.Fig. 3 shows a kind of embodiment of video frequency search system, and this video frequency search system can be used for the retrieval of recorded video, also can be used for the real-time retrieval of supervisory control system.Referring to Fig. 3, in conjunction with the retrieval of recorded video is described, this video frequency search system comprises:
Detection module 310; Video image to having recorded carries out motion detection; Judge whether include moving target in the video image, judge promptly that also image is kept in motion or remains static, and when in judging image, comprising moving target (at this moment; Claim that video image has got into motion state), catch the moving target that causes image generation movable information;
Tracking module 320 links to each other with detection module 310, after detection module 310 captures moving target; Moving target is carried out target detection; Discerning the type of this moving target, is human body or vehicle or the like such as identifying this moving target, and begins to follow the tracks of this moving target in the video image; Detect at detection module 310 that not comprise in the video image that moving target or moving target remain static also be image when returning to inactive state, stop to follow the tracks of;
Characteristics analysis module 330; Link to each other with tracking module 320; Obtain the type and the image of moving target from tracking module 320; To characteristic information and the movable information that obtains moving target after the graphical analysis, also obtain the information of the i.e. motion fragment of sequence of video images that comprises moving target simultaneously, and above-mentioned information is write memory module 350 from tracking module 320;
Above moving target information can only be obtained partial information wherein such as the type of moving target, the characteristic information of moving target and the movable information of moving target etc.Wherein, the moving target type can be like people, vehicle etc.The characteristic information of moving target can be the overall dimension information like human body, the colouring information of the clothes that human body is worn, perhaps people's face overall dimension and colour of skin information, the perhaps overall dimension of automobile and colouring information or the like.The movable information of moving target can be scope of activities, position, the direction of motion, movement velocity like moving target or the like.The sports segment information can be like numbering of motion fragment temporal information (like the time started or the concluding time of motion fragment), sports segment identification such as motion fragment or the like.
In addition; When comprising a plurality of moving target in certain tracing process, can analyze the image of each moving target, extract the information of each moving target respectively; As the motion fragment, each motion fragment in time maybe be all or part of overlapping with the sequence of video images that has this moving target.Also can only select wherein one or more moving targets (can be according to size, whether in certain particular range activity wait select) information, only the video image of these one or more moving targets is analyzed the extraction with above-mentioned information.
All characteristics are deposited in the database, utilize database to set up index, and the characteristic of the moving target that characteristics analysis module 330 is obtained sends to memory 350 for these characteristics;
Memory module 350 links to each other with characteristics analysis module 330, is used to preserve one or more sports segment informations and corresponding moving target information thereof;
Retrieval module 360; Link to each other with memory module 350; Be used for retrieving at the database of memory module 350, sports segment information that retrieves and/or moving target information are sent to video display module 370 as result for retrieval according to the search terms of user's input.Search terms can comprise type, the color of intending the retrieval moving target, and perhaps moving target stays duration or the like in scene;
Video display module 370 links to each other with retrieval module 360, the video that is used to show the image of motion fragment or plays the motion fragment.
This video frequency search system can also comprise a video extraction module 380; This video extraction module 380 links to each other with retrieval module 360 and video display module 370; Sports segment information in the testing result that retrieves according to retrieval module 360 extracts the one or more image in this motion fragment video, extracts one or more snippets video in the perhaps said motion fragment; Then with the one or more image in this motion fragment, perhaps one or more snippets video sends to video display module 370 and shows or play.
Afterwards, the temporal information of the motion fragment that this video frequency search system can will be checked according to the user extracts the video of corresponding sports fragment from video.Certainly, system's video of directly extracting the corresponding sports fragment according to the temporal information of the motion fragment in the retrieval module result for retrieval supplies the user to check also to be fine.
This retrieval module can only provide sports segment information and/or moving target information in result for retrieval, the user transfers corresponding motion fragment voluntarily and checks directly according to this information.This moment, system only provided query function, and above-mentioned video display module 370 and video extraction module 380 can be set.
In another embodiment; Retrieval module 360 is when returning result for retrieval; This result for retrieval can also be sports segment information and the corresponding moving target information thereof that satisfies search condition, and the user confirms whether will transfer and check corresponding motion fragment video through video extraction module 380 according to these information.If retrieval is less than information matching then point out user search less than qualified information.The user can determine whether to check complete motion fragment video according to result for retrieval.
In addition; Detection module 310 detects after the moving target; Other functional units that can also coupling system carry out that further other are judged, also comprise the database of preserving the moving target characteristic information and the moving target identification module that is used to discern moving target such as system, and the moving target that detects and follow the tracks of like detection module 310 is people's face; Then can the image and the characteristic information thereof of this people's face be sent to a face recognition module; This face recognition module is passed through the face database in the inquiry system, the automatic identification of people's face that realization captures, personage's face characteristic information such as identity under the acquisition people face.If the moving target that detection module 310 detects and follows the tracks of is a vehicle; Then image and the characteristic information thereof with this vehicle sends to a vehicle identification module; This vehicle identification module is through the vehicle data storehouse in the inquiry system; The automatic identification of the vehicle that realization captures, vehicle characteristic informations such as the license plate number of acquisition vehicle, vehicle, car owner.
The moving target information that characteristics analysis module 330 is obtained; Can also feed back to tracking module 320 to improve the tracking efficient of tracking module 320; Such as overall dimension information, positional information and the direction of motion information according to moving target, the following range that can dwindle tracking module 320.
Fig. 4 shows a kind of flow process of the video retrieval method based on Fig. 3 video frequency search system, to be that example describes to recorded video, comprises the steps: equally
Step 410 is carried out motion detection to the video image of importing, and obtains the moving target that is kept in motion in the video image;
Step 420 through moving target being carried out target detection and follow the tracks of, is obtained in type, characteristic information and the movable information of moving target information such as moving target one or more, and with the corresponding sports segment information of moving target information; Wherein the type of moving target is such as people's face, human body or vehicle or the like; The moving target information of being obtained can be used for the tracking to this target, to dwindle following range, improves and follows the tracks of efficient;
Step 430 is preserved moving target information and corresponding sports segment information in database, and the information that can be preservation is set up index to improve retrieval rate;
When in tracing process, comprising a plurality of moving target; Then select part or all of moving target; Extract the moving target information of these moving targets, to each moving target of selecting, with the motion fragment of the sequence of video images that has this moving target as this moving target correspondence.
Step 440 after the motion detection of recorded video finishes, is retrieved the information of preserving in the database according to the search terms of user's input;
Can improve recall precision through the content of adjustment search terms; Such as when carrying out the retrieval of people's face; Because the colourity of people's face is within the narrower chromaticity range of scope; Therefore a chromaticity range more approaching with human body complexion can further be set in search terms, only in meeting the moving region of this chromaticity range, carry out the retrieval of people's face in the time of retrieval, can increase substantially people's face effectiveness of retrieval;
Except the information of preserving in the database being retrieved, also can retrieve the whole video image according to the search terms of user's input according to the search terms of user's input;
Above-mentioned retrieval can be carried out on the backstage, when result for retrieval is shown, can export information through modes such as sound, image brightness variations, and to User Alarms, prompting detects the testing result that meets search terms.
Can also obtain and preserve the motion fragment that contains the moving target that is kept in motion through moving target is followed the tracks of, after obtaining result for retrieval, can from the motion fragment of preserving, extract video or image.
Video retrieval method shown in Figure 4; After obtaining result for retrieval, through the alarm user, so the extraction and the search operaqtion of motion detection, characteristic information can be carried out on the backstage fully; Only last result for retrieval being presented to the user makes further judgement and gets final product; Retrieving does not need user's participation, and the user is unlikely to produce tired, has reduced the possibility of failing to report wrong report.In addition; When retrieving to the motion fragment; At first high-rate fitration is fallen not contain the video image of movable information and is kept significant motion fragment; And only the motion fragment is retrieved automatically according to the search terms of user's input on this basis, so retrieval rate is much larger than the original frame per second of video.
It is big to the present invention is directed to present supervisory control system video content memory space; Return technical problems such as looking into, retrieve difficulty; Through technology such as motion detection and the detection of people's face, human detection, vehicle detection, a kind of video analysis and storage system and method and a kind of video frequency search system and method have been realized.Video analysis wherein and storage system and method are through high speed, intellectual analysis; The motion fragment that the user is concerned about is preserved in order to application such as retrievals; Improved operating factor of memory space; Solved the problem that data space expands preferably, this time span that also means the video image that can store simultaneously significantly increases, as increased to 1 year from one month.Video frequency search system wherein and method can have been analyzed the original video of long-time section within a short period of time according to the characteristic information that from the motion fragment, obtains.Search terms according to user's input is retrieved video image, has improved the detection recognition efficiency greatly, has also accelerated the speed of video analysis significantly, has improved operating efficiency and retrieval rate.
Though the execution mode that the present invention disclosed as above, the execution mode that described content just adopts for the ease of understanding the present invention is not in order to limit the present invention.Technical staff under any the present invention in the technical field; Under the prerequisite of spirit that does not break away from the present invention and disclosed and scope; Can do any modification and variation what implement in form and on the details; But scope of patent protection of the present invention still must be as the criterion with the scope that appending claims was defined.
Claims (10)
1. a video frequency search system is characterized in that, comprises detection module, tracking module, characteristics analysis module, memory module and retrieval module, wherein:
Said detection module carries out motion detection and target detection to the video image of importing, and catching the object that is kept in motion in the said video image is moving target;
Said tracking module links to each other with said detection module, after said detection module captures said moving target, discerns and follow the tracks of said moving target, returns to inactive state up to detecting said video image, stops to follow the tracks of said moving target;
Said characteristics analysis module; Link to each other with said tracking module; Be used to obtain the moving target information of said moving target; And the sports segment information that contains the i.e. motion fragment of sequence of video images of said moving target, said moving target information comprises the type of moving target, the characteristic information of moving target and the movable information of moving target, said sports segment information comprises motion fragment temporal information and sports segment identification;
Said memory module links to each other with said characteristics analysis module, is used to preserve said moving target information and sports segment information;
Said retrieval module links to each other with said memory module, moving target information and/or the sports segment information preserved is retrieved the output result for retrieval according to the search terms of user's input.
2. video frequency search system as claimed in claim 1 is characterized in that,
When said characteristics analysis module is obtained a plurality of moving target from said tracking module; Select the image of part or all of moving target to resolve; Extract the characteristic information and the movable information of these moving targets, and obtain the temporal information and the identification information of the corresponding motion fragment of each moving target of selection.
3. according to claim 1 or claim 2 video frequency search system is characterized in that this system also comprises video extraction module and video display module, wherein:
Said retrieval module further sends to said video extraction module with corresponding sports segment information when retrieving the motion fragment of mating with said search terms;
Said video extraction module; Link to each other with said detection module; Be used for according to the video extraction one or more image of said sports segment information from said motion fragment; Perhaps from said motion fragment, extract one or more snippets video, then the image or the video that extract are sent to said video display module;
Said video display module links to each other with said video extraction module, is used to show the video that the image received or broadcast are received.
4. according to claim 1 or claim 2 video frequency search system is characterized in that,
The result for retrieval of said retrieval module output is sports segment information and/or the corresponding moving target information that matches according to search terms.
5. video frequency search system as claimed in claim 1 is characterized in that,
Said characteristics analysis module is further given said tracking module with said moving target feedback information, to improve the tracking efficient of said tracking module.
6. video frequency search system as claimed in claim 1 is characterized in that, said system also comprises moving target database and moving target identification module, wherein:
Said moving target database is preserved the characteristic information of moving target;
Said moving target identification module links to each other with said detection module, is used for obtaining from said detection module the image of the moving target of respective type, and discerns according to the characteristic information of this moving target in the said moving target database, and the output recognition result.
7. a video retrieval method is characterized in that, comprises step:
Video image to input carries out motion detection, and catching the object that is kept in motion in the said video image is moving target;
Through said moving target being carried out target detection and following the tracks of; Obtain and preserve the moving target information of said moving target and comprise the sports segment information of the i.e. motion fragment of sequence of video images of said moving target; All characteristics are deposited in the database; Said moving target information comprises the type of moving target, the characteristic information of moving target and the movable information of moving target, and said sports segment information comprises motion fragment temporal information and sports segment identification;
After said video image detects and finishes, said moving target information and/or the sports segment information preserved are retrieved according to the search terms of user's input, and the output result for retrieval.
8. video retrieval method as claimed in claim 7; It is characterized in that; When in tracing process, comprising a plurality of moving target, select part or all of moving target, extract the moving target information of these moving targets; To each moving target of selecting, with the motion fragment of the sequence of video images that has this moving target as this moving target correspondence.
9. like claim 7 or 8 described video retrieval methods, it is characterized in that,
Said result for retrieval is one or more snippets video in the motion fragment that matches according to search terms; Or the one or more image in the motion fragment that matches according to search terms, or the sports segment information and/or the corresponding moving target information that match according to search terms.
10. like claim 7 or 8 described video retrieval methods, it is characterized in that,
Like detected moving target is people's face, then extracts the image and the characteristic information thereof of this people's face, and according to the characteristic information of people's face in the face database of configuration this people's face is discerned automatically, and the output recognition result; And/or
Like detected moving target is vehicle, then extracts the image and the characteristic information thereof of this vehicle, and according to the characteristic information of vehicle in the vehicle data storehouse of configuration this vehicle is discerned automatically, and the output recognition result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200810115569 CN101299812B (en) | 2008-06-25 | 2008-06-25 | Method, system for analyzing, storing video as well as method, system for searching video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200810115569 CN101299812B (en) | 2008-06-25 | 2008-06-25 | Method, system for analyzing, storing video as well as method, system for searching video |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101299812A CN101299812A (en) | 2008-11-05 |
| CN101299812B true CN101299812B (en) | 2012-12-05 |
Family
ID=40079480
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200810115569 Active CN101299812B (en) | 2008-06-25 | 2008-06-25 | Method, system for analyzing, storing video as well as method, system for searching video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101299812B (en) |
Families Citing this family (54)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101763388B (en) * | 2008-12-25 | 2013-03-27 | 北京中星微电子有限公司 | Method for searching video, system therefor and device therefor as well as video storing method and system thereof |
| CN101753997B (en) * | 2009-12-22 | 2015-01-21 | 北京中星微电子有限公司 | Processing method and processing device for video data of video monitoring system |
| CN102081671A (en) * | 2011-01-25 | 2011-06-01 | 北京中星微电子有限公司 | Method and device for generating scalable video file |
| CN102148963A (en) * | 2011-04-06 | 2011-08-10 | 上海玄烨电子系统工程有限公司 | Method and system for facing digital high-definition network video monitoring based on cloud storage |
| CN103020624B (en) * | 2011-09-23 | 2015-12-02 | 杭州海康威视系统技术有限公司 | All-purpose road monitor video smart tags, retrieval back method and device thereof |
| CN103188468A (en) * | 2011-12-30 | 2013-07-03 | 支录奎 | Study on useful video recognition video tape recorder |
| CN102625081B (en) * | 2012-02-20 | 2014-12-03 | 华焦宝 | Method and system for tracking target in visit place |
| CN102665036B (en) * | 2012-05-10 | 2015-01-07 | 江苏友上科技实业有限公司 | High-definition three-in-one network video recorder system |
| CN102752582A (en) * | 2012-06-19 | 2012-10-24 | 杭州海康威视系统技术有限公司 | Intelligent video analysis system and method |
| JP2014053749A (en) * | 2012-09-06 | 2014-03-20 | Sony Corp | Image processor, image processing method, and program |
| CN103778159A (en) * | 2012-10-26 | 2014-05-07 | 苏州千视通信科技有限公司 | Video detection process based on video abstraction and video retrieval |
| CN102902819B (en) * | 2012-10-30 | 2015-10-14 | 浙江宇视科技有限公司 | A kind of Intelligent video analysis method and device |
| CN103838750A (en) * | 2012-11-23 | 2014-06-04 | 苏州千视通信科技有限公司 | Video investigation technique for obtaining vehicle features based on video retrieval abstract |
| CN103945171B (en) * | 2013-01-17 | 2017-10-10 | 上海快视信息技术有限公司 | A kind of criminal investigation video analytic system and method |
| CN104143074A (en) * | 2013-05-07 | 2014-11-12 | 李东舸 | Method and equipment for generating motion feature codes on the basis of motion feature information |
| CN104144520A (en) | 2013-05-07 | 2014-11-12 | 李东舸 | Method, device and system for establishing connection between devices |
| CN104142939B (en) | 2013-05-07 | 2019-07-02 | 杭州智棱科技有限公司 | A kind of method and apparatus based on body dynamics information matching characteristic code |
| CN104239309A (en) * | 2013-06-08 | 2014-12-24 | 华为技术有限公司 | Video analysis retrieval service side, system and method |
| CN103327306A (en) * | 2013-06-14 | 2013-09-25 | 广东威创视讯科技股份有限公司 | Method and device for storing video surveillance image |
| WO2015061964A1 (en) * | 2013-10-29 | 2015-05-07 | Hua Zhong University Of Science Technology | Simulataneous metadata extraction of moving objects |
| CN104751636A (en) * | 2013-12-27 | 2015-07-01 | 贺江涛 | Video analysis-based unlicensed taxi statistical system and method |
| KR102126370B1 (en) * | 2014-11-24 | 2020-07-09 | 한국전자통신연구원 | Apparatus and method for analyzing motion |
| CN104506934A (en) * | 2014-12-25 | 2015-04-08 | 合肥寰景信息技术有限公司 | Video processing method capable of saving storage space |
| CN104980681A (en) * | 2015-06-15 | 2015-10-14 | 联想(北京)有限公司 | Video acquisition method and video acquisition device |
| CN104883548B (en) * | 2015-06-16 | 2018-08-14 | 金鹏电子信息机器有限公司 | Monitor video face captures processing method and its system |
| CN105007434B (en) * | 2015-07-24 | 2018-06-08 | 北京紫荆视通科技有限公司 | Processing method, processing unit and the video matrix of video data |
| CN105357475A (en) * | 2015-10-28 | 2016-02-24 | 小米科技有限责任公司 | Video playing method and device |
| CN105578135A (en) * | 2015-12-15 | 2016-05-11 | 浙江宇视科技有限公司 | Video on-demand method and device |
| CN105744199A (en) * | 2015-12-16 | 2016-07-06 | 上海大学 | Method for automatically recording dynamic objects in monitoring system |
| CN106910211A (en) * | 2015-12-21 | 2017-06-30 | 中国石油天然气股份有限公司 | Multi-maneuvering-target tracking method under complex environment |
| CN105704457A (en) * | 2016-03-09 | 2016-06-22 | 国网浙江省电力公司湖州供电公司 | Target state identification and online monitoring equipment for important channel scene during extra-high-voltage power transmission process |
| CN105931313A (en) * | 2016-04-13 | 2016-09-07 | 北京小米移动软件有限公司 | Recording method and device of automobile data recorder |
| CN106202333A (en) * | 2016-07-02 | 2016-12-07 | 北京工业大学 | A kind of method of store in a warehouse Video processing and quick-searching |
| CN106446002A (en) * | 2016-08-01 | 2017-02-22 | 三峡大学 | Moving target-based video retrieval method for track in map |
| CN107690058A (en) * | 2016-08-04 | 2018-02-13 | 惠州学院 | A kind of multiple target video frequency following system and method |
| CN106372607A (en) * | 2016-09-05 | 2017-02-01 | 努比亚技术有限公司 | Method for reading pictures from videos and mobile terminal |
| CN106210674A (en) * | 2016-09-22 | 2016-12-07 | 江苏理工学院 | Personnel-oriented monitoring video data file rapid processing method and system |
| CN106709424B (en) * | 2016-11-19 | 2022-11-11 | 广东中科人人智能科技有限公司 | Optimized monitoring video storage system |
| CN106530331A (en) * | 2016-11-23 | 2017-03-22 | 北京锐安科技有限公司 | Video monitoring system and method |
| CN107103615A (en) * | 2017-04-05 | 2017-08-29 | 合肥酷睿网络科技有限公司 | A kind of monitor video target lock-on tracing system and track lock method |
| CN107395968A (en) * | 2017-07-26 | 2017-11-24 | Tcl移动通信科技（宁波）有限公司 | Mobile terminal and its video recording operation detection process method and storage medium |
| CN107666573A (en) * | 2017-10-13 | 2018-02-06 | 北京奇虎科技有限公司 | The method for recording of object video and device, computing device under camera scene |
| CN108156406A (en) * | 2017-11-15 | 2018-06-12 | 珠海格力电器股份有限公司 | Information processing method and device of automobile data recorder |
| CN107944382B (en) * | 2017-11-20 | 2019-07-12 | 北京旷视科技有限公司 | Method for tracking target, device and electronic equipment |
| CN108391088A (en) * | 2018-03-12 | 2018-08-10 | 北京环境特性研究所 | A kind of safety defense monitoring system and method |
| CN109271554B (en) * | 2018-09-07 | 2022-02-11 | 江西省云眼大视界科技有限公司 | Intelligent video identification system and application thereof |
| CN111126115B (en) * | 2018-11-01 | 2024-06-07 | 顺丰科技有限公司 | Violent sorting behavior identification method and device |
| CN109361904A (en) * | 2018-11-21 | 2019-02-19 | 佛山市云团科技有限公司 | A kind of monitor video slice storage method and system |
| CN109657624A (en) * | 2018-12-21 | 2019-04-19 | 秒针信息技术有限公司 | Monitoring method, the device and system of target object |
| CN109871815B (en) * | 2019-02-26 | 2021-07-20 | 新华三技术有限公司 | Method and device for inquiring monitoring information |
| CN110083742B (en) * | 2019-04-29 | 2022-12-06 | 腾讯科技（深圳）有限公司 | Video query method and device |
| CN111866444A (en) * | 2019-04-29 | 2020-10-30 | 杭州海康威视数字技术股份有限公司 | Video data storage method and device |
| CN112818172B (en) * | 2021-01-18 | 2023-04-18 | 浙江大华技术股份有限公司 | Video retrieval method, electronic device and storage medium |
| CN117834942A (en) * | 2023-12-15 | 2024-04-05 | 山东爱特云翔信息技术有限公司 | Mobile detection video recording method based on ViBe algorithm |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1801930A (en) * | 2005-12-06 | 2006-07-12 | 南望信息产业集团有限公司 | Dubious static object detecting method based on video content analysis |
| CN1968384A (en) * | 2005-11-18 | 2007-05-23 | 建兴电子科技股份有限公司 | Digital A/V optical disc camcorder |
| CN101071447A (en) * | 2006-03-06 | 2007-11-14 | 索尼株式会社 | Search system, image-capturing apparatus, data storage apparatus, information processing apparatus |
-
2008
- 2008-06-25 CN CN 200810115569 patent/CN101299812B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1968384A (en) * | 2005-11-18 | 2007-05-23 | 建兴电子科技股份有限公司 | Digital A/V optical disc camcorder |
| CN1801930A (en) * | 2005-12-06 | 2006-07-12 | 南望信息产业集团有限公司 | Dubious static object detecting method based on video content analysis |
| CN101071447A (en) * | 2006-03-06 | 2007-11-14 | 索尼株式会社 | Search system, image-capturing apparatus, data storage apparatus, information processing apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101299812A (en) | 2008-11-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101299812B (en) | Method, system for analyzing, storing video as well as method, system for searching video | |
| CN101763388B (en) | Method for searching video, system therefor and device therefor as well as video storing method and system thereof | |
| CN106203458B (en) | Crowd video analysis method and system | |
| Bashir et al. | Performance evaluation of object detection and tracking systems | |
| US20070294207A1 (en) | People searches by multisensor event correlation | |
| CN107958220B (en) | Face library compression processing method based on face recognition and intelligent device thereof | |
| US20040161133A1 (en) | System and method for video content analysis-based detection, surveillance and alarm management | |
| US10535375B2 (en) | Information processing system, information processing method, and recording medium | |
| US20060279630A1 (en) | Method and apparatus for total situational awareness and monitoring | |
| CN110290346B (en) | Bidding video acquisition method based on intelligent video analysis | |
| US20030033318A1 (en) | Instantly indexed databases for multimedia content analysis and retrieval | |
| CN102881100A (en) | Video-analysis-based antitheft monitoring method for physical store | |
| CN108540750A (en) | Based on monitor video and the associated method, apparatus of electronic device identification and system | |
| CN105279496B (en) | A kind of method and apparatus of recognition of face | |
| CN106355154B (en) | Method for detecting frequent passing of people in surveillance video | |
| US20190042088A1 (en) | A method of data analysis and extraction for embedded security devices | |
| CN108540752B (en) | Method, device and system for identifying target object in video monitoring | |
| CN103916626A (en) | Monitoring video information providing method and device and video monitoring system | |
| CN108074363A (en) | Perimeter protection monitoring identifying system based on depth camera | |
| CN111259789A (en) | Face recognition intelligent security monitoring management method and system | |
| CN111191498A (en) | Behavior recognition method and related product | |
| CN104980695A (en) | Coordination of object location data with video data | |
| CN112019928B (en) | Video playback method and device and electronic equipment | |
| CN108540748A (en) | Monitor video and the associated method, apparatus of electronic device identification and system | |
| CN114627403A (en) | Video index determining method, video playing method and computer equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20160516 Address after: 519031 Guangdong city of Zhuhai province Hengqin Baohua Road No. 6, room 105 -478 Patentee after: GUANGDONG ZHONGXING ELECTRONICS CO., LTD. Address before: 100083 Haidian District, Xueyuan Road, No. 35, the world building, the second floor of the building on the ground floor, No. 16 Patentee before: Beijing Vimicro Corporation |