CN103150319B - The rearmounted video frequency abstract searching system of a kind of characteristic key based on CS framework - Google Patents
The rearmounted video frequency abstract searching system of a kind of characteristic key based on CS framework Download PDFInfo
- Publication number
- CN103150319B CN103150319B CN201210488338.7A CN201210488338A CN103150319B CN 103150319 B CN103150319 B CN 103150319B CN 201210488338 A CN201210488338 A CN 201210488338A CN 103150319 B CN103150319 B CN 103150319B
- Authority
- CN
- China
- Prior art keywords
- background
- moving target
- information
- task
- file
- 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 claims abstract description 18
- 238000000605 extraction Methods 0.000 claims description 11
- 238000001914 filtration Methods 0.000 claims description 9
- 230000033001 locomotion Effects 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 4
- 230000006837 decompression Effects 0.000 claims description 2
- 230000008520 organization Effects 0.000 claims description 2
- 230000002123 temporal effect Effects 0.000 claims 2
- 238000007635 classification algorithm Methods 0.000 claims 1
- 238000013144 data compression Methods 0.000 claims 1
- 238000001514 detection method Methods 0.000 claims 1
- 230000032683 aging Effects 0.000 abstract 1
- 230000015572 biosynthetic process Effects 0.000 abstract 1
- 238000004891 communication Methods 0.000 description 7
- 230000008859 change Effects 0.000 description 6
- 238000007630 basic procedure Methods 0.000 description 4
- 239000000284 extract Substances 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000004040 coloring Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000012806 monitoring device Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000012856 packing Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Television Signal Processing For Recording (AREA)
Abstract
The invention provides the rearmounted video frequency abstract searching system solution of a kind of characteristic key based on C/S framework and the application in intelligent security guard thereof, the method effectively solves the inefficiencies that current video summarization system filtercondition is preposition, the availability making abstract system strengthens, ageing good, rearmounted filtercondition makes summary previous information formation speed promote substantially simultaneously.
Description
Technical field
The invention belongs to computer vision field, particularly to the rearmounted video frequency abstract searching system solution of a kind of characteristic key based on C/S framework, and the application that the method is in intelligent security guard.
Technical background
Along with computer network, digitized video technology develop rapidly, video monitoring based on digital network obtains a wide range of application, but the thing followed is, along with the scope expanding day that monitoring part is anti-, the quantity of monitoring device also increases with surprising quantity, in the face of ten hundreds of monitoring devices, the video main contents how quickly understanding magnanimity control point become the maximum restriction that the projects such as safe city develop further.
At present in the face of the general solution of this restriction is video summarization system, the video difference quick Play System of alienation etc..But due to traditional video frequency abstract solution or be there is no filtercondition, namely it is a simple moving target summary, this plays a role in practice owing to the interference of local frequently motion is very difficult in practicality, occurring in that afterwards can to certain area, moving target under certain logical condition carries out the video frequency abstract solution extracted, but these systems will set filtercondition often before generating summary, once summary starts to generate, these conditions will be unable to amendment, and it is known that the process generating summary is a relatively time-consuming process of ratio, if it is desired to the change cost that must regenerate of filtercondition is also very poor efficiency in actual applications.The present invention is be extracted the rearmounted video frequency abstract solution of the filtercondition based on C/S model on the basis of above existing summary solution, the present invention mails to client after the information of moving target is carried out extraction in advance by video frequency abstract service end, and client loads summary info dynamically according to the search condition of user.So can save the substantial amounts of time for the summary of many condition retrieval, thus being largely the practicality that improve video summarization system.
Summary of the invention
It is an object of the invention to for existing video summarization system, treatment conditions be processed the feature of poor efficiency, propose the rearmounted video frequency abstract solution of a kind of characteristic key based on C/S framework, thus greatly enhancing the treatment effeciency of video frequency abstract, add the practicality of video frequency abstract.
In order to realize goal of the invention, the solution of employing is as follows:
Based on C/S framework the rearmounted video frequency abstract searching system of characteristic key basic procedure as shown in Figure 1.First, the summarization generation that abstract system summary client is submitted to controls acquisition of information and specifies data stream, data stream is carried out moving Object Segmentation process by service end, extract the moving target information in video flowing, and moving target is quickly followed the tracks of, and to its classification information of each Objective extraction after following the tracks of, colouring information and size information；Afterwards, it is taken based on the task split-run that Background divides, moving target is carried out file organization, and mails to client after packing；After client receives generated video intermediate file, file being resolved, user passes through client input inquiry condition, gets summary info further according to resolving Summary file, is finally opened up aobvious.
Hereinafter the present invention is introduced respectively by service end and client:
1, video frequency abstract service end:
In video frequency abstract service end, by big functional module, service end can be divided into communication module, task module, data acquisition module and management module.
Communication module is mainly used in work such as controlling the reception parsing of stream information, summary data transmits.
The mission bit stream that data acquisition module is submitted to according to client connects designated equipment and obtains data message.
Task module is the main functional modules of summary service end, and it realizes the basic function that video generates.According to the basic procedure of algorithm, this module can be divided into background model maintenance module, foreground extracting module, moving target recognition module, motion target tracking module, moving target type-collection module, moving target color extraction module, moving target size extraction module, Summary file molded tissue block.
In the present invention, employing the typical mixed Gaussian module extraction module as video background, video background updates and carries out according to following strategy: (1) carries out background model renewal according to user configured fixed interval；(2) feedback information according to foreground module, carries out background module forcing to update.
Foreground extracting module have employed the statistical method based on pixel Gauss distribution and carries out foreground extraction.The method is assumed for pixel any in image, and it is necessarily two kinds of roles on a timeline: prospect, background.For security protection scene, no matter this pixel changes to background from prospect, or from background to prospect, the gray value of pixel all has a bigger change, present invention assumes that for any pixel in video image, during as its role for prospect or background (when character does not change), this pixel is in the time range of same nature, the change service Gauss distribution of pixel value, when the change of pixel value does not service from Gauss distribution, it is believed that role's character of this pixel there occurs change.The present invention, according to assumed above, utilizes slip window sampling to extract display foreground.Simultaneously, the prospect extracted is carried out largest connected regional determination, if finding, the area of a certain prospect connected region has exceeded a certain fixed threshold T, then it is assumed that video scene receives interference, then this information is fed back in background model module, carry out forcing context update to process.
Moving target recognition module adopts the method for traditional largest connected extracted region of foreground picture to realize.
Fast-moving target tracing is quickly followed the tracks of according to the maximum overlapping area method of consecutive frame motion, the method target that namely supposes to coexist can have more than partially overlapping of threshold value T in the prospect of front and back two frame, the moving target sequence currently extracted is mated one by one with historical movement target, if sport foreground area coincidence degree exceedes threshold value T, then think the current goal that this target is this history target, update this history target, otherwise continue, if current goal fails and history object matching, then it is assumed that current goal is a fresh target.
Moving target type-collection module is characterized as feature pool with HUG, adopts random forest grader, it is achieved the classification of type of sports people's car.
In moving target color extraction module, system initially sets up a set of Standard Colors scale, and this scale only has specific N kind color, as red, yellow, blue, green, orange etc..The each color of this scale is set up the HS color cartogram module in corresponding hsv color space, extract moving target HS color statistic histogram in hsv color space simultaneously, the each color rectangular histogram of the rectangular histogram of moving target Yu Standard Colors scale carries out the contrast based on the distribution of card side, and the colouring information of definition current goal is the immediate color of color in scale.
In moving target size extraction module, determine moving target size with the minimum edge of moving target and maximal side.
In the present invention, Summary file molded tissue block is comparatively crucial module, in system, adopts three grades of file structure organizing video Summary files.The first order is task level, and the video processing duties that namely video frequency abstract is submitted to user is for elementary cell.The second level is background summary level, the method adopting self adaptation to divide at the corresponding levels carries out Summary file tissue, the method follows following rule: (1) often sets up a task, and acquiescence sets up first Summary file module, the background image being this Summary file with current background image；(2) in this task, if finding, the background image after updating is when relatively current background image has larger difference, encapsulate current Summary file module, set up the new video file module based on renewal rear backdrop, system adopts the method based on hsv color space HS Color Statistical rectangular histogram relevant comparative's method realize the contrast of Background differentiation.The third level is target level, and this grade of file structure preserves image information and moving target type, color, the size information of the sequence of moving target.The mode of every grade of file structure XML concordance list carries out the management of subordinate's file structure in systems.When user shift to an earlier date task summary processed time, service end by task Summary file module compress after mail to client by communication module.Video frequency abstract file structure is as shown in Figure 2.
2, video frequency abstract client:
Video frequency abstract client, according to big functional module, it is possible to be divided into communication module, management module, summary parsing and information filtering module, summary playing module, the corresponding original document playback of summary and broadcasting to control module.
Communication module realizes controlling stream information transmission, service end status information receives, video frequency abstract file module receives and decompression.
Summary resolves and information filtering module can specifically be divided into video frequency abstract and filter information interpolation module, video frequency abstract document analysis module, summarized radio molded tissue block.
Video frequency abstract filters information and arranges module and mainly realize client and arrange current filtration information by this module, namely extract specified type, color, size, region summary play out.
The video frequency abstract file module that service end is sent by video frequency abstract document analysis module resolves, first the task ID information submitted to according to user, navigate to task Summary file, video frequency abstract management index file in simulation task Summary file is to traveling through each video frequency abstract file based on background, for any one video frequency abstract file, according to the moving target index file in this document, traversal target information, combined filtering information retrieval go out target travel target and are committed to summarized radio molded tissue block.
Summarized radio molded tissue block, according to the video frequency abstract received, carries out the summarized radio in units of Background and generates, and be committed to video frequency abstract playing module.
Video frequency abstract playing module carries out the broadcasting of Summary file according to the video frequency abstract received and plays control accordingly.And respond user's query event to target, target information is corresponded to according to the target location that user clicks, finally give the original video files information that this target is corresponding, original video files information it is submitted to the corresponding original document playback of summary and plays control module, it is achieved making a summary back and look into playing function.
The corresponding original document of summary plays back and broadcasting control module realizes original video and plays and corresponding broadcasting control function.
Above method proposed by the invention, more traditional video abstraction generating method, it is possible to more quickly generate summary, and the quick-replaceable of search condition can be realized, thus the practicality of the video summarization system largely improved.
Accompanying drawing explanation
Fig. 1 is the basic flow sheet of the present invention；
Fig. 2 is the video frequency abstract file structure figure of the present invention.
Detailed description of the invention
1, between server module, each module realizes function and basic procedure and describes as follows:
Communication module, it is achieved sending based on the control stream of TCP, state stream information, the compression that video frequency abstract file throttles based on binary word, based on the Summary file sending function of File Transfer Protocol.
Task module mainly realizes algorithm task class, the unique information task object of each task creation one, this algorithm task class mainly realizes the maintenance of video background module, video foreground information retrieval, moving target recognition, motion target tracking, moving target type information extracts, the extraction of moving target colouring information, the extraction of moving target size information, Summary file tissue and management.
Data acquisition module, the main data source information realizing providing according to task, load video stream data, and be ultimately converted to RGB frame image data stream.
Management mould, mainly realizes the organizing and coordinating of function of each module, it is achieved task is distributed, and task resource reclaims.
2, between client modules, each module realizes function and basic procedure and describes as follows:
Communication module mainly realizes the control stream information based on Transmission Control Protocol and state stream information receives and sends, based on the video frequency abstract file reception of File Transfer Protocol, based on the Summary file solution compression functions of binary file stream.
Management module mainly realizes the coordination between each functions of modules and management.
Summary resolves and information filtering module realizes video frequency abstract file module and resolves, and the information that filters is arranged, and moving target in summary is filtered, and generates the function of summary after filtering.
Summary playing module mainly realizes the broadcasting of video frequency abstract, and corresponding broadcasting controls function, video frequency abstract target playback acquisition of information function.
Summary corresponding original document playback and broadcasting control module mainly realize video file remote playing corresponding to summary and corresponding broadcasting control function.
Claims (3)
1. the rearmounted video frequency abstract searching system of the characteristic key based on C/S framework, it is characterized in that system scheme takes C/S framework to carry out system organization, system includes server and client side, system realizes video data acquiring at server end, moving target recognition and moving target information retrieval, data compression, packet decompression is realized in client, goal filtering based on moving target information, moving target broadcasting represents and corresponding playing back videos represents, the characteristic of system is described as: first, the moving target extracted is carried out based on kind by system server, the moving target information retrieval of color and size, it shows as and first target is divided into " vehicle " and " people " by target classification algorithm moving target carried out based on improving HOG feature, secondly, system utilizes the algorithm based on color model to carry out color target information retrieval, finally, foundation is retrieved carrying out the size as target of the record object minimum length of side in whole scene and maximal side during target traversal is analyzed；Second, utilize the task split-run divided based on Background, it shows as and is undertaken based on the rational division of background by a longer task, and in this, as the foundation of local Summary file tissue, so-called background is rationalized division and is referred to, if one section of original video contains the some sections of video segments that background difference is bigger, then this section of video is clustered with the difference of background image for classification foundation, and in units of each class, carry out the local file tissue of task video, these local files have a feature in that in file to comprise unique background image, file comprises the some moving target foreground images being associated in time with this background, the background summary segmentation that this file structure called after task is made a summary, and the background summary segmentation for task summary sets up with the index file of task name；3rd, the background summary segmentation index file of task summary has herein below: first, this index file includes appointed task relevant information, including temporal information, fileinfo；Secondly, index file is with moving target for unit, and the essential information of record moving target, including temporal information, classification information, size information, positional information.
2. system according to claim 1, it is characterized in that system server terminal moving target recognition adopts background modeling and mobilism more new solution, it shows, background dynamics updates the context update being primarily based on fixed frequency, next is in that context update carries out feedback with foreground detection to be associated, if prospect maximum area is more than the certain proportion of whole scene, carrying out forcing context update, background model uses mixed Gauss model to carry out background modeling；Foreground extraction adopts the statistics side based on pixel Gauss distribution to extract；Motion target tracking method adopts area coincidence method quickly to follow the tracks of.
3. system according to claim 1, it is characterised in that system client carries out summary based on below scheme and represents and play back: first, client loads the background summary segmentation index of the task summary that service end is set up；Secondly, background summary segmentation directory entry is filtered by the search condition that client inputs according to user, and the entry after filtration is final retrieval result；Finally, the directory entry after filtering according to retrieval navigates to moving target, and plays out and represent.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210488338.7A CN103150319B (en) | 2012-11-16 | 2012-11-16 | The rearmounted video frequency abstract searching system of a kind of characteristic key based on CS framework |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210488338.7A CN103150319B (en) | 2012-11-16 | 2012-11-16 | The rearmounted video frequency abstract searching system of a kind of characteristic key based on CS framework |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103150319A CN103150319A (en) | 2013-06-12 |
| CN103150319B true CN103150319B (en) | 2016-06-29 |
Family
ID=48548400
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210488338.7A Active CN103150319B (en) | 2012-11-16 | 2012-11-16 | The rearmounted video frequency abstract searching system of a kind of characteristic key based on CS framework |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103150319B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104427369B (en) * | 2013-09-09 | 2018-08-10 | 联想(北京)有限公司 | Remote control end equipment is remotely-controlled end equipment and the method for it |
| CN104754248B (en) * | 2013-12-30 | 2018-05-01 | 浙江大华技术股份有限公司 | A kind of method and device for obtaining target snapshot |
| CN104581437B (en) * | 2014-12-26 | 2018-11-06 | 中通服公众信息产业股份有限公司 | A kind of video frequency abstract generates and the method and system of video backtracking |
| CN109165317B (en) * | 2018-10-31 | 2019-08-06 | 杭州恒生数字设备科技有限公司 | A kind of real time monitoring aspect indexing inquiry system |
| CN110674347B (en) * | 2019-09-02 | 2022-04-01 | 南京邮电大学 | Visual shielding double-layer AP video abstract generation method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101753853A (en) * | 2009-05-13 | 2010-06-23 | 中国科学院自动化研究所 | Fusion method for video scene segmentation |
| CN102156707A (en) * | 2011-02-01 | 2011-08-17 | 刘中华 | Video abstract forming and searching method and system |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003230127A (en) * | 2002-01-31 | 2003-08-15 | Matsushita Electric Ind Co Ltd | Digest video specifying system, digest video providing system, digest video specifying method, digest video providing method, medium, and program |
| KR20070001240A (en) * | 2004-04-23 | 2007-01-03 | 코닌클리케 필립스 일렉트로닉스 엔.브이. | Method and apparatus for catch-up of running broadcast or stored content |
| US8949235B2 (en) * | 2005-11-15 | 2015-02-03 | Yissum Research Development Company Of The Hebrew University Of Jerusalem Ltd. | Methods and systems for producing a video synopsis using clustering |
-
2012
- 2012-11-16 CN CN201210488338.7A patent/CN103150319B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101753853A (en) * | 2009-05-13 | 2010-06-23 | 中国科学院自动化研究所 | Fusion method for video scene segmentation |
| CN102156707A (en) * | 2011-02-01 | 2011-08-17 | 刘中华 | Video abstract forming and searching method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103150319A (en) | 2013-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110166851B (en) | Video abstract generation method and device and storage medium | |
| CN103150319B (en) | The rearmounted video frequency abstract searching system of a kind of characteristic key based on CS framework | |
| EP3291558B1 (en) | Video coding and decoding methods and apparatus | |
| Chatzichristofis et al. | CEDD: Color and edge directivity descriptor: A compact descriptor for image indexing and retrieval | |
| CN104200671B (en) | A kind of virtual bayonet socket management method based on large data platform and system | |
| CN101873414B (en) | A Hierarchical Event Video Detection System | |
| US20230412769A1 (en) | Scalable Visual Computing System | |
| CN118075427A (en) | Intelligent monitoring management method and system | |
| CN102156707A (en) | Video abstract forming and searching method and system | |
| Gan et al. | Video object forgery detection algorithm based on VGG-11 convolutional neural network | |
| CN110197135A (en) | A kind of video structural method based on multidimensional segmentation | |
| CN110659333A (en) | A Multi-level Visual Feature Description Method and Visual Retrieval System | |
| CN104320625A (en) | Intelligent video monitoring method and system for safe village | |
| CN106682628A (en) | Face attribute classification method based on multilayer depth feature information | |
| Chato et al. | Image processing and artificial neural network for counting people inside public transport | |
| CN104333732A (en) | Distributed video analysis method and system thereof | |
| CN114372167B (en) | Image processing method and device | |
| CN109657098B (en) | Video fingerprint extraction method and device | |
| CN105657372A (en) | Method and system for realizing intelligent detection and early warning of on-duty guard posts by videos | |
| CN117197713A (en) | An extraction method based on digital video surveillance system | |
| CN106097483B (en) | A kind of video record processing method and system based on intelligent back vision mirror | |
| CN111881794A (en) | Video behavior identification method and system | |
| CN113744528A (en) | Wisdom urban traffic video monitor system | |
| CN107169004A (en) | A kind of property data base update method and device | |
| Ouyang et al. | The comparison and analysis of extracting video key frame |
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 | ||
| DD01 | Delivery of document by public notice | ||
| DD01 | Delivery of document by public notice |
Addressee: PCI-Suntektech Co., Ltd. Document name: Notification of Passing Examination on Formalities |
|
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: Room 306, area 2, building 1, Fanshan Venture Center, Panyu energy saving science and Technology Park, 832 Yingbin Road, Donghuan street, Panyu District, Guangzhou, Guangdong 510000 Patentee after: Jiadu Technology Group Co.,Ltd. Address before: No.4, Jiangong Road, Tianhe Software Park, Guangzhou, Guangdong 510665 Patentee before: PCI-SUNTEKTECH Co.,Ltd. |