CN104581437A - Video abstract generation and video backtracking method and system - Google Patents
Video abstract generation and video backtracking method and system Download PDFInfo
- Publication number
- CN104581437A CN104581437A CN201410830140.1A CN201410830140A CN104581437A CN 104581437 A CN104581437 A CN 104581437A CN 201410830140 A CN201410830140 A CN 201410830140A CN 104581437 A CN104581437 A CN 104581437A
- Authority
- CN
- China
- Prior art keywords
- target
- video
- frequency abstract
- video frequency
- information
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
- H04N21/85—Assembly of content; Generation of multimedia applications
- H04N21/854—Content authoring
- H04N21/8549—Creating video summaries, e.g. movie trailer
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Computer Security & Cryptography (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a video abstract generation and video backtracking method and system. According to the method, moving objects in an original video are separated, the separated moving objects are identified and classified, and a highly concentrated video abstract file with compact content is generated with a scalable image fusion technology and a multi-mode information fusion method. During video backtracking, firstly, the number and types of the objects appearing in a video image are elastically controlled according to the number of the objects appearing in the image selected by a user as well as an object which the user is interested in. When the user clicks one object in the image, the system can quickly backtrack the original video on a server and present the whole appearing process of the object to the user.
Description
Technical field
The present invention relates to field of video monitoring, particularly relate to the method and system that a kind of video frequency abstract generates and video is recalled.
Background technology
In field of storage, along with developing rapidly of society, people are more and more stronger to safety precaution, and demand is also more and more higher simultaneously, and video monitoring is as the effective means in safety precaution field, and range of application is more and more wider, and application demand is also in continuous improve.Surveillance video is as the carrier in safety precaution field, it can allow people more effectively look back reduction time in the past institute occurrence, but the storage data volume of video record is large, storage time is long, by video recording search people that a certain time in the past occurs, car or some event time just seem very consuming time, considerably increase human and material resources.In order to browse these surveillance videos quickly and efficiently, video summarization technique just seems particularly important.
Video summarization technique is by the analysis to video content and structure, extracts main information in video, and they is merged in some way the brief video or video frame image that can give full expression to video semanteme content.The object of video frequency abstract is the redundant information removed in original video, thus improves the utilization factor of video.
Video frequency abstract plays key player in video analysis and content based video retrieval system, the brief video frequency abstract generated by video summarization technique, contains all important activities in former video.Multiple events that different time in original video occurs are compressed into a brief video frequency abstract by multinomial technological means such as image recognition, redundancy removal, video merging by video summarization technique.The thing of the personage that can be occurred in fast browsing special time period by video frequency abstract, vehicle and generation, the cost short period understand grasp process in several hours even this monitoring range of several days people, car and generation behavior, can quick lock in destination object when there is case, accelerate for public security speed of solving a case, the efficiency of solving a case of raising major case, important case has great importance.
The object of video frequency abstract is that conveniently user checks video fast, and the quality of the video frequency abstract of generation directly affects the experience effect of user.The problem such as imperfect, ghost of ubiquity target in current video frequency abstract.In addition, because video frequency abstract has upset the sequential logic of target in original video, if user will understand the truth of some targets, also need to check by original video, for the use of user is made troubles, the efficiency of video frequency searching is not high.
Summary of the invention
In view of this, the present invention proposes a kind of video frequency abstract and generates and the method and system of video backtracking, for solving the technical matters how tracing back to original video the original case of checking this target from video frequency abstract file fast.
For achieving the above object, the technical scheme of the embodiment of the present invention is achieved in that
The method that video frequency abstract generates and video is recalled, the method comprises:
Video background image is extracted from original video files;
Utilize the video background image extracted to isolate moving target from original video files, and carry out foreground target tracking, the moving target separated is identified and classifies;
Store the target information of background image, the moving target separated and correspondence, described target information comprises: the temporal information that target occurs in original video files, spatial information, target type;
Be fused to by the moving target separated in the background image extracted, generating video is made a summary, and described video frequency abstract comprises the target information of moving target and correspondence;
The graphical user interface provided by video frequency abstract directly recalls according to target information the video pictures that in original video files, target occurs.
Further, the described moving target to separating identifies and classification is specially:
Histograms of oriented gradients HOG feature is extracted from the positive negative sample of all kinds of target;
The HOG feature of the positive negative sample obtained is put in support vector machines and trains, obtain the feature templates of each target type;
Input the frame of video of moving target to be detected, extract the HOG feature of moving target to be detected;
Use the described characteristic module obtained to mate with the HOG feature of the moving target to be detected of extraction, determine moving target type.
Further, store described corresponding with moving target target information by the mode of index file or database table, between described video frequency abstract, index file or database table, original video files, there is man-to-man incidence relation; Temporal information in described target information comprises the time span of moment that target occurs in original video, appearance; Described temporal information is used for the position of locating the appearance of described moving target in described original video files.
Further, the described method be fused to by the moving target separated in the background image extracted is specially:
Video frequency abstract playout software is realized by OO program implementation, when showing described video frequency abstract, read the video file of described moving target, background image file and target information, for each target sets up destination object, each destination object comprises the target information of this target;
Described video frequency abstract playout software provides display-object number configuration interface one, and this interface is for realizing the number controlling the target occurred in video frequency abstract picture simultaneously;
Described video frequency abstract playout software provides target number configuration interface two, this interface for realizing temporal information in the target information associated by destination object, spatial information judges that target accounts for the overlapping degree between the ratio of video pictures and target, and determines the number of the target in present video frequency abstract picture according to the parameter preset;
Described video frequency abstract playout software provides type selecting configuration interface, and this interface determines the destination object shown by video frequency abstract picture for the target type realized in the target information associated by destination object;
Described video frequency abstract playout software provides scalable configuration interface, and this interface determines whether to zoom in or out target for the spatial information realized in the target information associated by destination object.
Further, the video pictures that the described graphical user interface provided by video frequency abstract directly recalls target appearance in original video files according to target information is specially:
Response user is in the window events of video frequency abstract picture, and the target information that based target object is corresponding, directly transfers original video files and trace back to the position that moving target corresponding to destination object occur in original video files and carry out video playback.
Based on the embodiment of the present invention, the present invention also provides the system that a kind of video frequency abstract generates and video is recalled, and this system comprises:
Background extracting device, for extracting video background image from original video files;
Target tripping device, for utilizing the video background image of extraction to isolate moving target from original video files, and carries out foreground target tracking;
Target classification device, for identifying the moving target separated and generating target information after classifying;
Memory storage, for storing the target information of the correspondence after the background image extracted, the moving target separated and discriminator, target information comprises: the temporal information that target occurs in original video files, spatial information, target type;
Video frequency abstract generating apparatus, for being fused to by the moving target separated in the background image that extracts, generating video Summary file also stores, and described video frequency abstract comprises the target information of moving target and correspondence;
Video frequency abstract playing device, for display and the video backtracking of video frequency abstract, directly recalls according to target information the video pictures that in original video files, target occurs by graphical user interface.
Further, described sorter comprises:
Target's feature-extraction unit, for extracting histograms of oriented gradients HOG feature from the positive negative sample of all kinds of target; And in the frame of video from the moving target to be detected inputted, extract the HOG feature of moving target to be detected;
Training unit, training for the HOG feature of the positive negative sample obtained being put in support vector machines, obtaining the feature templates of each target type;
Taxon, the HOG feature for the moving target to be detected using described characteristic module and the extraction obtained is mated, and determines moving target type.
Further, described memory storage stores described corresponding with moving target target information by the mode of index file or database table, has man-to-man incidence relation between described video frequency abstract, index file or database table, original video files; Temporal information in described target information comprises the time span of moment that target occurs in original video, appearance; Described temporal information is used for the position of locating the appearance of described moving target in described original video files.
Further, described video frequency abstract generating apparatus and video frequency abstract playing device realize based on OO program implementation;
Described video frequency abstract playing device is when showing described video frequency abstract, and read the video file of described moving target, background image file and target information, for each target sets up destination object, each destination object comprises the target information of this target;
Described video frequency abstract playing device provides display-object number configuration interface one, and this interface is for realizing the number controlling the target occurred in video frequency abstract picture simultaneously;
Described video frequency abstract playing device provides target number configuration interface two, this interface for realizing temporal information in the target information associated by destination object, spatial information judges that target accounts for the overlapping degree between the ratio of video pictures and target, and determines the number of the target in present video frequency abstract picture according to the parameter preset;
Described video frequency abstract playing device provides type selecting configuration interface, and this interface is for realizing the destination object shown by the target type determination video frequency abstract picture in the target information associated by destination object;
Described video frequency abstract playing device provides scalable configuration interface, and this interface determines whether to zoom in or out target for the spatial information realized in the target information associated by destination object.
Further, described video frequency abstract playing device response user is in the window events of video frequency abstract picture, the target information that based target object is corresponding, directly transfers original video files and traces back to the position that moving target corresponding to destination object occur in original video files and carry out video playback.
Moving target in original video is separated and identifies the moving target be separated and classify by the present invention program, use the method for telescopic image integration technology and multi-mode information fusion generate content compact, the video frequency abstract file of high enrichment.In video trace-back process, there is number and the interested object of user of target in the picture first selected according to user, flexibly controls the number of target appearance and the type of target in video pictures.When user clicks object wherein, system will trace back to rapidly the original video on server, and the whole process occurred by this object presents to user.
Accompanying drawing explanation
Fig. 1 is a kind of video frequency abstract based on scalable integration technology of providing of the embodiment of the present invention and retrogressive method schematic flow sheet;
Fig. 2 is the schematic diagram that background image that the embodiment of the present invention provides is separated with sport foreground;
Fig. 3 is the schematic diagram that the embodiment of the present invention is provided as foreground object classification process;
Fig. 4 be the embodiment of the present invention provide a kind of based on the video frequency abstract of scalable integration technology and the structural representation of backtracking system.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearly understand, below by way of specific embodiment and see accompanying drawing, the present invention is described in detail.
Existing video frequency abstract generation technique can realize the concentrated of original video content, can different target under same video background be spliced and combined in common background image at the video content of different time sections, but the various types of targets in video frequency abstract, the destination object of different time sections simply mixes and summary image may be caused chaotic, be unfavorable for browsing fast and lock onto target object, in addition, owing to there is no direct incidence relation between the destination object in video frequency abstract and original video, need user's localizing objects object in original video by hand, also cause location difficulty and recall precision not high.
The object of the invention is to propose a kind of video frequency abstract generate and video retrogressive method and system, from original video, the Hot Contents of visualization is extracted by the method for multi-mode information fusion, as destination object, event etc., and be concentrated in specific spatial domain by time-domain information and pass through generated video frequency abstract compactly and express magnanimity vision content, finally video frequency abstract, video low-level image feature and high-level semantics features are combined to facilitate the interested video content of user search.After the HI SA highly saturated concentrated video frequency abstract of acquisition, user can browse concentrated video frequency abstract picture, and clicks wherein interested focus, directly traces back in original video frames, achieves the high speed retrieval of picture interested.
The steps flow chart schematic diagram of a kind of method that video frequency abstract generates and video is recalled that Fig. 1 provides for the embodiment of the present invention, comprises the following steps:
Step 100, from original video files, extract video background image;
This step is used for carrying out video background reconstruction to original video files, and after reading in an original video files, use Background Reconstruction algorithm, such as Gaussian Background reconstruction algorithm/mixture Gaussian background model extraction algorithm, extracts video background image from original video;
The video background image that step 102, utilization are extracted isolates moving target from original video files, and carries out foreground target tracking, identifies and classify to the moving target separated;
Shown in figure 2, in an embodiment of the present invention, the method separated from original video by moving target is: the frame sequence of input original video files, use Gaussian Background difference algorithm, carry out calculus of differences by the frame of video of extracted background image and current original video, thus the moving target in original video and foreground target are separated.
In an embodiment of the present invention, the algorithm isolated moving target being carried out to foreground target tracking can be Kalman filtering algorithm, and the present invention is not specifically limited.
In an embodiment of the present invention, the moving target separated is identified and classifies, thus determine the type of moving target, described target type includes but not limited to: people, motor vehicle, bicycle, animal etc., pedestrian can also be further divided into man, woman, motor vehicle also can divide into truck, car, motorcycle etc. further, and bicycle also can divide into bicycle, rickshaw etc. further.
Shown in figure 3, in an embodiment of the present invention, in the following way moving target identified and classify: first, histograms of oriented gradients (Histogramof Oriented Gradient is extracted from the positive negative sample of all kinds of target, HOG) feature, then the HOG feature of the positive negative sample obtained is put into support vector machine (Support Vector Machine, SVM, a kind of trainable machine learning method) in train, obtain the feature templates of each target type; When carrying out discriminator, first the frame of video of moving target to be detected is inputted, extract the HOG feature of moving target to be detected, then the HOG feature of the moving target to be detected of characteristic module and the extraction obtained is used to mate, the type of moving target is determined according to the matching degree thresholding preset, such as when the matching degree of the feature templates of isolated moving target and pedestrian meets or exceeds 90%, then the type of this moving target is defined as pedestrian.
The target information of step 104, storage background image, the moving target separated and correspondence, described target information comprises: the temporal information that target occurs in original video files, spatial information, target type;
In one embodiment of the invention, the video of the moving target separated is carried out classification by the target type determined and stores, and store the target information of each target accordingly.Described target information includes but not limited to: the temporal information that target occurs in original video files, the spatial information occurred in picture, target type etc.Described temporal information comprises the moment (frame position that corresponding target occurs) that target occurs in original video, the time span (target survival time span in video) etc. occurred.Described spatial information comprises the information such as coordinates regional, target sizes that target occurs in background image.
In one embodiment of the invention, the target information corresponding with moving target is stored by index file, such as, set up an index file for each original video and store the corresponding relation of original video files and index file, index file is for storing the temporal information, spatial information, target type etc. of each moving target occurred in original video, by target information corresponding with moving target in this index file, the position of original video files and the appearance of moving target in original video files can be navigated to fast.In addition, the information such as original video totalframes, total number of targets, total target information number can also be stored in this index file, for initialization and the display of video frequency abstract.
In an alternative embodiment of the invention, mode store video also by relational database is made a summary, the corresponding relation of original video files and target information, such as set up video frequency abstract table, a list item in the corresponding video frequency abstract table of each video frequency abstract, comprises the reference address information of video frequency abstract title and video frequency abstract file, the title of original video and reference address information, target information etc. from the isolated moving target of original video files in list item.
Step 106, to be fused to by the moving target separated in the background image that extracts, generating video is made a summary, and described video frequency abstract comprises the target information of moving target and correspondence;
Each foreground target is fused in the video background picture set up above by this step, generates video frequency abstract file that is compact, concentrated, that express massive video content.
In one embodiment of the invention, the moving target separated and background image merge and adopt the distinctive telescopic image integration technology of the present invention, namely video frequency abstract playout software is realized by OO program implementation, when showing described video frequency abstract, read the video file of described moving target, background image file and target information, for each target sets up destination object, each destination object contains the target information of this target.Described destination object carries out instantiation in a program by reading corresponding target information from the index file or video frequency abstract table of preceding step foundation, by the data member of destination object, event functions and Interface realization to the control of destination object and mutual, thus realize following function:
(1) described video frequency abstract playout software provides display-object number configuration interface one, and this interface is for realizing the number controlling the target occurred in video frequency abstract picture simultaneously;
(2) described video frequency abstract playout software provides target number configuration interface two, this interface for realizing temporal information in the target information associated by destination object, spatial information judges that target accounts for the overlapping degree between the ratio of video pictures and target, and determines the number of the target in present video frequency abstract picture according to the parameter preset;
(3) described video frequency abstract playout software provides type selecting configuration interface, this interface determines the destination object shown by video frequency abstract picture for the target type realized in the target information associated by destination object, and the configuration interface namely by software human-computer interaction interface selects the destination object showing which target type;
Such as, that require only concern is pedestrian, then only show pedestrian when playing summarized radio, and also can select to pay close attention to the target type occurred, namely all types of target is all shown in video pictures.
(4) described video frequency abstract playout software provides scalable configuration interface, this interface determines whether to zoom in or out target for the spatial information (comprising the position of target appearance, the size of target) realized in the target information associated by destination object, better visual effect is obtained when target is appeared in video frequency abstract picture, the complete of video content can be remained like this, and obtain information saturation degree high in summary frame.
Step 108, the graphical user interface provided by video frequency abstract directly recall according to target information the video pictures that in original video files, target occurs.
In one embodiment of the invention, after obtaining video frequency abstract that is HI SA highly saturated, that concentrate, based on generated video frequency abstract, user can browsing video summary picture, and responded the interface operation of user by the graphical user interface (Graphic User Interface, GUI) that video frequency abstract playout software provides, such as, when after user's click wherein interested destination object, can directly trace back in original video frame, achieve the high speed retrieval of picture interested.
Based on foregoing OO program implementation, the window events of user at video frequency abstract picture is responded by the event functions of destination object, such as double mouse click event, based on the calling interface that video frequency abstract playout software provides in the response function of event, directly transfer original video files and trace back to the position that moving target corresponding to destination object occur in original video files and carry out video playback, wherein the input parameter of calling interface comprises the access location of original video files, destination object identifies, the parameters such as the target information that destination object is corresponding.
In one embodiment of the invention, in order to improve the speed that summarized radio generates and accesses, the video data of the video frequency abstract file separated, moving target and index file are stored in local storage, original video files is stored in the data center with mass storage capacity, by streaming media on demand technology, directly can trace back to the original video on data center server rapidly, preserve original video without the need to user in this locality, greatly reduce carrying cost and the storage space of user like this.
The system that a kind of video frequency abstract generates and video is recalled that Fig. 4 provides for the embodiment of the present invention, this system 400 comprises: background extracting device 410, target tripping device 420, target classification device 430, memory storage 440, video frequency abstract generating apparatus 450, video frequency abstract playing device 460.
Background extracting device 410 extracts video background image from original video files, and target tripping device 420 utilizes the video background image extracted to isolate moving target from original video files, and carries out foreground target tracking.Target classification device 430 identifies the moving target separated and generates target information after classifying.Memory storage 440 stores the target information of the correspondence after the background image extracted, the moving target separated and discriminator, and target information comprises: the temporal information that target occurs in original video files, spatial information, target type;
The target information of video frequency abstract generating apparatus 450 background extraction image, the moving target separated and correspondence from memory storage 440, the moving target separated is fused in the background image extracted, generating video Summary file is also stored in memory storage 440, and video frequency abstract comprises the target information of moving target and correspondence.Wherein, memory storage 440 answers being interpreted as of broad sense to comprise internal memory, magnetic storage medium, storage networking or cloud memory device of being provided by network etc. all can provide memory storage that is interim and permanent storage function.
Video frequency abstract playing device 450 is for the display of video frequency abstract and video backtracking, its graphical user interface allows user by window operations such as clicks, and the target information based on moving target object and association directly recalls the video pictures that in original video files, target occurs.
Further, sorter 430 comprises: target's feature-extraction unit 431, training unit 432, taxon 433.Target's feature-extraction unit 431 extracts histograms of oriented gradients HOG feature from the positive negative sample of all kinds of target, from the frame of video of the moving target to be detected of input, extracts the HOG feature of moving target to be detected.The HOG feature of the positive negative sample obtained is put in support vector machines and trains by training unit 432, obtains the feature templates of each target type.Taxon 433 uses the described characteristic module obtained to mate with the HOG feature of the moving target to be detected of extraction, determines moving target type.
Further, memory storage 440 stores the target information corresponding with moving target by the mode of index file or database table, has man-to-man incidence relation between described video frequency abstract, index file or database table, original video files;
Further, the temporal information in described target information comprises the time span of moment that target occurs in original video, appearance; Described temporal information is used for the position of locating the appearance of described moving target in described original video files.
Further, video frequency abstract generating apparatus 450 and video frequency abstract playing device 460 realize based on OO program implementation, video frequency abstract playing device 460 is when showing described video frequency abstract, read the video file of described moving target, background image file and target information, for each target sets up destination object, each destination object comprises the target information of this target; Video frequency abstract playing device 460 provides display-object number configuration interface one, and this interface is for realizing the number controlling the target occurred in video frequency abstract picture simultaneously.Video frequency abstract playing device 460 also provides target number configuration interface two, this interface for realizing temporal information in the target information associated by destination object, spatial information judges that target accounts for the overlapping degree between the ratio of video pictures and target, and determines the number of the target in present video frequency abstract picture according to the parameter preset.Video frequency abstract playing device 460 provides type selecting configuration interface, and this interface determines the destination object shown by video frequency abstract picture for the target type realized in the target information associated by destination object.Video frequency abstract playing device 460 provides scalable configuration interface, and this interface determines whether to zoom in or out target for the spatial information realized in the target information associated by destination object.
Further, based on above-mentioned OO implementation, video frequency abstract playing device 460 can respond the window events of user at video frequency abstract picture, the target information that based target object is corresponding, directly transfers original video files and traces back to the position that moving target corresponding to destination object occur in original video files and carry out video playback.
The embodiment of the present invention extracts the Hot Contents of visualization by the method for multi-mode information fusion, propose and set up the multi-modal video frequency abstract model of a kind of MMAR (Multiple Mode Abstract Retrival), video has been merged in concentrated video frequency abstract picture, audio frequency, text, the media contents such as case index, the video summary information Layering manifestation that formation can be recalled at a high speed, the subject object of classification and case index, original video frames can be got back to by click to object of interest any in frame or case index, thus realize retrieving the high speed of interested target picture.
The video frequency abstract based on scalable integration technology that the present invention proposes and retrogressive method and system, ensure that the video frequency abstract file of generation can comprise the object of interest and event that all original videos occur completely.When video playback, have employed telescopic image integration technology, can according to the needs of user, the number that in flexible control video pictures, target occurs and the type of target, both target can have been made intactly to appear in video pictures, good visual effect can have been obtained again.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment made, equivalent replacement, improvement etc., all should be included within the scope of protection of the invention.
Claims (10)
1. the method that video frequency abstract generates and video is recalled, it is characterized in that, the method comprises:
Video background image is extracted from original video files;
Utilize the video background image extracted to isolate moving target from original video files, and carry out foreground target tracking, the moving target separated is identified and classifies;
Store the target information of background image, the moving target separated and correspondence, described target information comprises: the temporal information that target occurs in original video files, spatial information, target type;
Be fused to by the moving target separated in the background image extracted, generating video is made a summary, and described video frequency abstract comprises the target information of moving target and correspondence;
The graphical user interface provided by video frequency abstract directly recalls according to target information the video pictures that in original video files, target occurs.
2. the method for claim 1, is characterized in that, the described moving target to separating identifies and classification is specially:
Histograms of oriented gradients HOG feature is extracted from the positive negative sample of all kinds of target;
The HOG feature of the positive negative sample obtained is put in support vector machines and trains, obtain the feature templates of each target type;
Input the frame of video of moving target to be detected, extract the HOG feature of moving target to be detected;
Use the described characteristic module obtained to mate with the HOG feature of the moving target to be detected of extraction, determine moving target type.
3. the method for claim 1, is characterized in that,
Store described corresponding with moving target target information by the mode of index file or database table, between described video frequency abstract, index file or database table, original video files, there is man-to-man incidence relation;
Temporal information in described target information comprises the time span of moment that target occurs in original video, appearance; Described temporal information is used for the position of locating the appearance of described moving target in described original video files.
4. the method for claim 1, is characterized in that, the described method be fused to by the moving target separated in the background image extracted is specially:
Video frequency abstract playout software is realized by OO program implementation, when showing described video frequency abstract, read the video file of described moving target, background image file and target information, for each target sets up destination object, each destination object comprises the target information of this target;
Described video frequency abstract playout software provides display-object number configuration interface one, and this interface is for realizing the number controlling the target occurred in video frequency abstract picture simultaneously;
Described video frequency abstract playout software provides target number configuration interface two, this interface for realizing temporal information in the target information associated by destination object, spatial information judges that target accounts for the overlapping degree between the ratio of video pictures and target, and determines the number of the target in present video frequency abstract picture according to the parameter preset;
Described video frequency abstract playout software provides type selecting configuration interface, and this interface determines the destination object shown by video frequency abstract picture for the target type realized in the target information associated by destination object;
Described video frequency abstract playout software provides scalable configuration interface, and this interface determines whether to zoom in or out target for the spatial information realized in the target information associated by destination object.
5. method as claimed in claim 4, is characterized in that, the described graphical user interface provided by video frequency abstract is directly recalled according to target information the video pictures that in original video files, target occurs and is specially:
Response user is in the window events of video frequency abstract picture, and the target information that based target object is corresponding, directly transfers original video files and trace back to the position that moving target corresponding to destination object occur in original video files and carry out video playback.
6. the system that video frequency abstract generates and video is recalled, it is characterized in that, this system comprises:
Background extracting device, for extracting video background image from original video files;
Target tripping device, for utilizing the video background image of extraction to isolate moving target from original video files, and carries out foreground target tracking;
Target classification device, for identifying the moving target separated and generating target information after classifying;
Memory storage, for storing the target information of the correspondence after the background image extracted, the moving target separated and discriminator, target information comprises: the temporal information that target occurs in original video files, spatial information, target type;
Video frequency abstract generating apparatus, for being fused to by the moving target separated in the background image that extracts, generating video Summary file also stores, and described video frequency abstract comprises the target information of moving target and correspondence;
Video frequency abstract playing device, for display and the video backtracking of video frequency abstract, directly recalls according to target information the video pictures that in original video files, target occurs by graphical user interface.
7. system according to claim 6, is characterized in that, described sorter comprises:
Target's feature-extraction unit, for extracting histograms of oriented gradients HOG feature from the positive negative sample of all kinds of target; And in the frame of video from the moving target to be detected inputted, extract the HOG feature of moving target to be detected;
Training unit, training for the HOG feature of the positive negative sample obtained being put in support vector machines, obtaining the feature templates of each target type;
Taxon, the HOG feature for the moving target to be detected using described characteristic module and the extraction obtained is mated, and determines moving target type.
8. system according to claim 6, is characterized in that,
Described memory storage stores described corresponding with moving target target information by the mode of index file or database table, has man-to-man incidence relation between described video frequency abstract, index file or database table, original video files;
Temporal information in described target information comprises the time span of moment that target occurs in original video, appearance; Described temporal information is used for the position of locating the appearance of described moving target in described original video files.
9. system according to claim 6, is characterized in that,
Described video frequency abstract generating apparatus and video frequency abstract playing device realize based on OO program implementation,
Described video frequency abstract playing device is when showing described video frequency abstract, and read the video file of described moving target, background image file and target information, for each target sets up destination object, each destination object comprises the target information of this target;
Described video frequency abstract playing device provides display-object number configuration interface one, and this interface is for realizing the number controlling the target occurred in video frequency abstract picture simultaneously;
Described video frequency abstract playing device provides target number configuration interface two, this interface for realizing temporal information in the target information associated by destination object, spatial information judges that target accounts for the overlapping degree between the ratio of video pictures and target, and determines the number of the target in present video frequency abstract picture according to the parameter preset;
Described video frequency abstract playing device provides type selecting configuration interface, and this interface is for realizing the destination object shown by the target type determination video frequency abstract picture in the target information associated by destination object;
Described video frequency abstract playing device provides scalable configuration interface, and this interface determines whether to zoom in or out target for the spatial information realized in the target information associated by destination object.
10. system according to claim 9, is characterized in that,
Described video frequency abstract playing device response user is in the window events of video frequency abstract picture, the target information that based target object is corresponding, directly transfers original video files and traces back to the position that moving target corresponding to destination object occur in original video files and carry out video playback.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410830140.1A CN104581437B (en) | 2014-12-26 | 2014-12-26 | A kind of video frequency abstract generates and the method and system of video backtracking |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410830140.1A CN104581437B (en) | 2014-12-26 | 2014-12-26 | A kind of video frequency abstract generates and the method and system of video backtracking |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104581437A true CN104581437A (en) | 2015-04-29 |
| CN104581437B CN104581437B (en) | 2018-11-06 |
Family
ID=53096473
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410830140.1A Active CN104581437B (en) | 2014-12-26 | 2014-12-26 | A kind of video frequency abstract generates and the method and system of video backtracking |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104581437B (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105007464A (en) * | 2015-07-20 | 2015-10-28 | 江西洪都航空工业集团有限责任公司 | Method for concentrating video |
| CN105469425A (en) * | 2015-11-24 | 2016-04-06 | 上海君是信息科技有限公司 | Video condensation method |
| CN105872859A (en) * | 2016-06-01 | 2016-08-17 | 深圳市唯特视科技有限公司 | Video compression method based on moving target trajectory extraction of object |
| WO2017049577A1 (en) * | 2015-09-25 | 2017-03-30 | Qualcomm Incorporated | Systems and methods for video processing |
| CN106557534A (en) * | 2015-09-25 | 2017-04-05 | 财团法人工业技术研究院 | Video index establishing method and device applying same |
| CN106714007A (en) * | 2016-12-15 | 2017-05-24 | 重庆凯泽科技股份有限公司 | Video abstract method and apparatus |
| CN106708890A (en) * | 2015-11-17 | 2017-05-24 | 创意引晴股份有限公司 | Intelligent high fault-tolerant video identification system based on multimoding fusion and identification method thereof |
| CN106780664A (en) * | 2016-11-17 | 2017-05-31 | 温州医科大学 | A kind of technical journal figure summary editing system based on vector graphics element |
| CN107396165A (en) * | 2016-05-16 | 2017-11-24 | 杭州海康威视数字技术股份有限公司 | A kind of video broadcasting method and device |
| CN107426631A (en) * | 2016-05-23 | 2017-12-01 | 安讯士有限公司 | Summarized radio sequence is generated from source video sequence |
| CN107493441A (en) * | 2016-06-12 | 2017-12-19 | 杭州海康威视数字技术股份有限公司 | A kind of summarized radio generation method and device |
| CN107637089A (en) * | 2015-05-18 | 2018-01-26 | Lg电子株式会社 | Display device and its control method |
| WO2018205991A1 (en) * | 2017-05-12 | 2018-11-15 | 华为技术有限公司 | Method, apparatus and system for video condensation |
| CN109219844A (en) * | 2016-05-27 | 2019-01-15 | 杜比实验室特许公司 | It is converted between Video priority and graphics priority |
| CN109309868A (en) * | 2018-08-19 | 2019-02-05 | 朱丽萍 | Video file Command Line Parsing system |
| CN109661808A (en) * | 2016-07-08 | 2019-04-19 | 汉阳大学校产学协力团 | Simplify the recording medium of video-generating device, method and logger computer program |
| CN109783688A (en) * | 2018-12-28 | 2019-05-21 | 广州烽火众智数字技术有限公司 | A kind of distributed video abstract processing system |
| CN110166851A (en) * | 2018-08-21 | 2019-08-23 | 腾讯科技（深圳）有限公司 | A kind of video abstraction generating method, device and storage medium |
| CN111526413A (en) * | 2020-04-29 | 2020-08-11 | 江苏加信智慧大数据研究院有限公司 | Course video playback system and playback method |
| CN112004117A (en) * | 2020-09-02 | 2020-11-27 | 维沃移动通信有限公司 | Video playing method and device |
| CN112711966A (en) * | 2019-10-24 | 2021-04-27 | 阿里巴巴集团控股有限公司 | Video file processing method and device and electronic equipment |
| WO2021248432A1 (en) * | 2020-06-12 | 2021-12-16 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for performing motion transfer using a learning model |
| CN115455275A (en) * | 2022-11-08 | 2022-12-09 | 广东卓维网络有限公司 | Video processing system fusing inspection equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102156707A (en) * | 2011-02-01 | 2011-08-17 | 刘中华 | Video abstract forming and searching method and system |
| CN102930061A (en) * | 2012-11-28 | 2013-02-13 | 安徽水天信息科技有限公司 | Video abstraction method and system based on moving target detection |
| CN103150319A (en) * | 2012-11-16 | 2013-06-12 | 佳都新太科技股份有限公司 | CS (Client Server) framework-based feature retrieval rear video abstraction retrieval system |
| CN103617234A (en) * | 2013-11-26 | 2014-03-05 | 公安部第三研究所 | Device and method for active video concentration |
-
2014
- 2014-12-26 CN CN201410830140.1A patent/CN104581437B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102156707A (en) * | 2011-02-01 | 2011-08-17 | 刘中华 | Video abstract forming and searching method and system |
| CN103150319A (en) * | 2012-11-16 | 2013-06-12 | 佳都新太科技股份有限公司 | CS (Client Server) framework-based feature retrieval rear video abstraction retrieval system |
| CN102930061A (en) * | 2012-11-28 | 2013-02-13 | 安徽水天信息科技有限公司 | Video abstraction method and system based on moving target detection |
| CN103617234A (en) * | 2013-11-26 | 2014-03-05 | 公安部第三研究所 | Device and method for active video concentration |
Cited By (42)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11962934B2 (en) | 2015-05-18 | 2024-04-16 | Lg Electronics Inc. | Display device and control method therefor |
| US10986302B2 (en) | 2015-05-18 | 2021-04-20 | Lg Electronics Inc. | Display device and control method therefor |
| CN107637089A (en) * | 2015-05-18 | 2018-01-26 | Lg电子株式会社 | Display device and its control method |
| US11323651B2 (en) | 2015-05-18 | 2022-05-03 | Lg Electronics Inc. | Display device and control method therefor |
| CN105007464A (en) * | 2015-07-20 | 2015-10-28 | 江西洪都航空工业集团有限责任公司 | Method for concentrating video |
| CN108028969B (en) * | 2015-09-25 | 2021-07-06 | 高通股份有限公司 | System and method for video processing |
| WO2017049577A1 (en) * | 2015-09-25 | 2017-03-30 | Qualcomm Incorporated | Systems and methods for video processing |
| CN106557534A (en) * | 2015-09-25 | 2017-04-05 | 财团法人工业技术研究院 | Video index establishing method and device applying same |
| CN108028969A (en) * | 2015-09-25 | 2018-05-11 | 高通股份有限公司 | system and method for video processing |
| US10708673B2 (en) | 2015-09-25 | 2020-07-07 | Qualcomm Incorporated | Systems and methods for video processing |
| CN106708890A (en) * | 2015-11-17 | 2017-05-24 | 创意引晴股份有限公司 | Intelligent high fault-tolerant video identification system based on multimoding fusion and identification method thereof |
| CN105469425A (en) * | 2015-11-24 | 2016-04-06 | 上海君是信息科技有限公司 | Video condensation method |
| CN107396165A (en) * | 2016-05-16 | 2017-11-24 | 杭州海康威视数字技术股份有限公司 | A kind of video broadcasting method and device |
| CN107396165B (en) * | 2016-05-16 | 2019-11-22 | 杭州海康威视数字技术股份有限公司 | A kind of video broadcasting method and device |
| US10701301B2 (en) | 2016-05-16 | 2020-06-30 | Hangzhou Hikvision Digital Technology Co., Ltd. | Video playing method and device |
| CN107426631A (en) * | 2016-05-23 | 2017-12-01 | 安讯士有限公司 | Summarized radio sequence is generated from source video sequence |
| US10192119B2 (en) | 2016-05-23 | 2019-01-29 | Axis Ab | Generating a summary video sequence from a source video sequence |
| CN107426631B (en) * | 2016-05-23 | 2019-05-28 | 安讯士有限公司 | The method and video process apparatus of summarized radio sequence are generated from source video sequence |
| CN109219844A (en) * | 2016-05-27 | 2019-01-15 | 杜比实验室特许公司 | It is converted between Video priority and graphics priority |
| CN109219844B (en) * | 2016-05-27 | 2021-08-20 | 杜比实验室特许公司 | Transitioning between video priority and graphics priority |
| US11183143B2 (en) | 2016-05-27 | 2021-11-23 | Dolby Laboratories Licensing Corporation | Transitioning between video priority and graphics priority |
| CN105872859A (en) * | 2016-06-01 | 2016-08-17 | 深圳市唯特视科技有限公司 | Video compression method based on moving target trajectory extraction of object |
| CN107493441A (en) * | 2016-06-12 | 2017-12-19 | 杭州海康威视数字技术股份有限公司 | A kind of summarized radio generation method and device |
| CN107493441B (en) * | 2016-06-12 | 2020-03-06 | 杭州海康威视数字技术股份有限公司 | Abstract video generation method and device |
| CN109661808A (en) * | 2016-07-08 | 2019-04-19 | 汉阳大学校产学协力团 | Simplify the recording medium of video-generating device, method and logger computer program |
| CN109661808B (en) * | 2016-07-08 | 2021-10-26 | 汉阳大学校产学协力团 | Simplified video generation device, method, and recording medium for recording computer program |
| CN106780664A (en) * | 2016-11-17 | 2017-05-31 | 温州医科大学 | A kind of technical journal figure summary editing system based on vector graphics element |
| CN106714007A (en) * | 2016-12-15 | 2017-05-24 | 重庆凯泽科技股份有限公司 | Video abstract method and apparatus |
| CN108881119A (en) * | 2017-05-12 | 2018-11-23 | 华为技术有限公司 | A kind of methods, devices and systems of video concentration |
| CN108881119B (en) * | 2017-05-12 | 2021-02-12 | 华为技术有限公司 | Method, device and system for video concentration |
| WO2018205991A1 (en) * | 2017-05-12 | 2018-11-15 | 华为技术有限公司 | Method, apparatus and system for video condensation |
| CN109309868A (en) * | 2018-08-19 | 2019-02-05 | 朱丽萍 | Video file Command Line Parsing system |
| CN110166851A (en) * | 2018-08-21 | 2019-08-23 | 腾讯科技（深圳）有限公司 | A kind of video abstraction generating method, device and storage medium |
| CN109783688A (en) * | 2018-12-28 | 2019-05-21 | 广州烽火众智数字技术有限公司 | A kind of distributed video abstract processing system |
| CN112711966A (en) * | 2019-10-24 | 2021-04-27 | 阿里巴巴集团控股有限公司 | Video file processing method and device and electronic equipment |
| CN112711966B (en) * | 2019-10-24 | 2024-03-01 | 阿里巴巴集团控股有限公司 | Video file processing method and device and electronic equipment |
| CN111526413A (en) * | 2020-04-29 | 2020-08-11 | 江苏加信智慧大数据研究院有限公司 | Course video playback system and playback method |
| WO2021248432A1 (en) * | 2020-06-12 | 2021-12-16 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for performing motion transfer using a learning model |
| US11830204B2 (en) * | 2020-06-12 | 2023-11-28 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for performing motion transfer using a learning model |
| US20210390713A1 (en) * | 2020-06-12 | 2021-12-16 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for performing motion transfer using a learning model |
| CN112004117A (en) * | 2020-09-02 | 2020-11-27 | 维沃移动通信有限公司 | Video playing method and device |
| CN115455275A (en) * | 2022-11-08 | 2022-12-09 | 广东卓维网络有限公司 | Video processing system fusing inspection equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104581437B (en) | 2018-11-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104581437B (en) | A kind of video frequency abstract generates and the method and system of video backtracking | |
| Xu et al. | Video structured description technology based intelligence analysis of surveillance videos for public security applications | |
| US10970334B2 (en) | Navigating video scenes using cognitive insights | |
| KR102290419B1 (en) | Method and Appratus For Creating Photo Story based on Visual Context Analysis of Digital Contents | |
| CN101689394B (en) | Method and system for video indexing and video synopsis | |
| Tiwari et al. | A survey of recent work on video summarization: approaches and techniques | |
| CN103347167A (en) | Surveillance video content description method based on fragments | |
| CN106663196A (en) | Computerized prominent person recognition in videos | |
| Alam et al. | Video big data analytics in the cloud: A reference architecture, survey, opportunities, and open research issues | |
| Malon et al. | Toulouse campus surveillance dataset: scenarios, soundtracks, synchronized videos with overlapping and disjoint views | |
| CN102231820A (en) | Monitoring image processing method, device and system | |
| CN103198110A (en) | Method and system for rapid video data characteristic retrieval | |
| Ul Haq et al. | An effective video summarization framework based on the object of interest using deep learning | |
| Negi et al. | Object detection based approach for an efficient video summarization with system statistics over cloud | |
| Kim et al. | TVDP: Translational visual data platform for smart cities | |
| Atrey et al. | Intelligent multimedia surveillance: current trends and research | |
| Montalvo-Lezama et al. | Improving Transfer Learning for Movie Trailer Genre Classification using a Dual Image and Video Transformer | |
| Yang et al. | Semi-automatic image and video annotation system for generating ground truth information | |
| Nagaraja et al. | Content based video retrieval using support vector machine classification | |
| Mishra et al. | Parameter free clustering approach for event summarization in videos | |
| CN115687692A (en) | Video processing method and device, computer storage medium and intelligent interactive panel | |
| Sharma et al. | Analyzing the Need for Video Summarization for Online Classes Conducted During Covid-19 Lockdown | |
| Xue et al. | ISD-SSD: image splicing detection by using modified single shot MultiBox detector | |
| CN105528458A (en) | Method for realizing extrasensory experience by retrieving video content | |
| GM | Insights to Video Analytic Modelling Approach with Future Line of Research |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |