CN103617234A - Device and method for active video concentration - Google Patents
Device and method for active video concentration Download PDFInfo
- Publication number
- CN103617234A CN103617234A CN201310611321.0A CN201310611321A CN103617234A CN 103617234 A CN103617234 A CN 103617234A CN 201310611321 A CN201310611321 A CN 201310611321A CN 103617234 A CN103617234 A CN 103617234A
- Authority
- CN
- China
- Prior art keywords
- video
- target
- concentrated
- description document
- target trajectory
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/73—Querying
- G06F16/738—Presentation of query results
- G06F16/739—Presentation of query results in form of a video summary, e.g. the video summary being a video sequence, a composite still image or having synthesized frames
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/783—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/7847—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using low-level visual features of the video content
- G06F16/786—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using low-level visual features of the video content using motion, e.g. object motion or camera motion
Abstract
The invention discloses a device and method for active video concentration. According to the technical scheme, an input video is analyzed, a video structured description file is generated, and a relevant database is built; searching is carried out, and searching results are sequenced according to relevancy; an abstract video is generated according to the searching results. The video concentration process is rearrangement (time displacement) and compact piling up of target tracks, moving objects which appear in different time periods are translated to the same time period, and a transparent mapping mode is used for ensuring that overlapped areas of a plurality of targets are visible respectively. The original timestamp and the foreground rectangle of each target in the abstract video is labeled, a video indexing function is achieved, and corresponding original video clips are skipped to by clicking moving targets. The active video concentration scheme can be used for removing redundant target tracks and showing needed information of users, and the investigation efficiency based on videos is improved.
Description
Technical field
The present invention relates to a kind of video processing technique, be specifically related to the technology of a kind of video concentrated (being generating video summary).
Background technology
In social public security field, video monitoring system becomes and maintains public order, and strengthens an important component part of social management.Yet there is the features such as storage data volume is large, and storage time is long in video record, by video recording, finds clue, and obtaining the traditional way of evidence is to expend a large amount of human and material resources and time, and efficiency is extremely low, to such an extent as to miss the best, solves a case opportunity.
Video frequency abstract is to video content simplified summary, in automatic or automanual mode, first by moving target, analyzes, extract moving target, then the movement locus of each target is analyzed, different targets is recombinated and be spliced in common background scene, form video.Therefore in video monitoring system, people can pass through video frequency abstract fast browsing, and locking searching object, accelerates for public security the speed of solving a case, and the efficiency of solving a case that improves major case, important case has great importance.
The amorphousness of video media is to hinder the bottleneck problem of Video Applications of new generation, and in order to solve the amorphousness problem of video, researcher has proposed the technological approaches of video structural description.Video structural description technology comprises two layers of meaning: the identification of target signature in video and extraction are formed to descriptor text, based on descriptor, realize video content retrieval and data mining.
The technology of traditional generating video summary mainly contains two kinds: first method is the video frequency abstract based on key frame of video, first the method obtains the key frame of video, then by the method for hierarchical clustering, detect the similarity of video segment, the fragment that similarity is surpassed to threshold value is removed, and the part remaining is spliced into video frequency abstract; Another kind method is the video frequency abstract based on video features, and the method is according to the feature of the input video extracting, and calculates the excellent degree of video, and by the higher Highlight of excellent degree by splicing simply, form video frequency abstract.
But these two kinds of formed video frequency abstracts of method all cannot meet the requirement that user obtains the information of specific objective.And these two kinds of methods are all stitched together by video segment, still there is duration length, the not compact problem of motion.
Summary of the invention
For above-mentioned prior art, the object of the present invention is to provide a kind of active video to concentrate scheme, for solving existing video summarization method, cannot meet user to specifying the demand of specific objective generation summarized radio, and the summarized radio duration that conventional art produces is long, the not compact problem of motion.
In order to achieve the above object, the present invention adopts following technical scheme:
An active video enrichment facility, described device comprises:
Video structural description module, described video structural description module is used for input video to analyze, generating video structural description file;
Memory management module, described memory management module is used for storing and managing video structural description file, and according to the database of video structural description file set up;
Retrieval module, described retrieval module is retrieved the video structural description file in memory management module, and by result for retrieval according to its relevancy ranking;
Video concentrates summarization module, the concentrated summarization module of described video is for generating the summarized radio that meets search condition, do not losing moving target and not overlapping or do not have under overlapping prerequisite, the moving target that different time sections is occurred moves to the same time period, simmer down to summarized radio.
In the preferred embodiment of enrichment facility, the video structural description file that described video structural description module generates comprises target signature description document and target trajectory description document, the descriptor that described target signature description document comprises each target; The essential information that described target trajectory description document comprises each target trajectory.
Further, described memory management module is set up relevant database by the target signature description document in video structural description file, and the while is by all information of data base administration target signature description document.
Further, the concentrated summarization module of described video, when generating summarized radio, provides locus and the temporal information of the prospect rectangle sequence that track is corresponding according to target trajectory description document, guarantees not overlapping after moving target translation or there is no an overlapping.
Further, the concentrated summarization module of described video, when generating summarized radio, refers to the translation to the time of moving target to the translation of moving target.
Further, the concentrated summarization module of described video, when generating summarized radio, guarantees that with the form of transparent pinup picture each target is visible for the region that a plurality of target trajectorys are overlapped; Each target in summarized radio is marked to its original timestamp and prospect rectangle simultaneously, and realized video index function, click moving target and can jump to corresponding original video fragment.
The method that active video is concentrated, the method comprises the steps:
(1) input video is analyzed, generating video structural description file, and according to the structural description file set up Relational database generating;
(2) database of setting up is retrieved, and by result for retrieval according to relevancy ranking;
(3), according to step (2) result for retrieval, to target trajectory screening and reset, not overlapping or do not have under the prerequisite of overlapping after guaranteeing moving target translation, the moving target that different time sections is occurred moves to the same time period, simmer down to summarized radio.
In the preferred version of the method, the video structural description file generating in described step (1) specifically comprises: target signature description document and target trajectory description document, the descriptor that described target signature description document comprises each target, the essential information that described target trajectory description document comprises each target.
Further, in described step (1), utilize the target signature description document in video structural description file to set up relevant database, and by all information of Relational database management objectives feature description document.
Further, while concentrating summarized radio in described step (3), target trajectory description document in video structural description file provides locus and the temporal information of the prospect rectangle sequence that track is corresponding, guarantees not overlapping after moving target translation or there is no an overlapping.
Further, in described step (3), for the overlapped region of a plurality of target trajectorys, with the form of transparent pinup picture, guarantee that each target is visible.
Further, in described step (3), when concentrated summarized radio, each target in summarized radio is marked to its original timestamp and prospect rectangle, realize thus video index function.
Further, in described step (3), to the translation of moving target, be the translation to the time of moving target, do not comprise the translation of locus.
In further described step (3), to moving target, translation is adopted with the following method:
(31) exact position of each foreground picture of each target trajectory recording according to target trajectory description document in original video picture, obtains the prospect rectangle sequence that target trajectory is corresponding;
(32) queue container of the newly-built prospect rectangle sequence for storage track, i.e. concentrated queue;
(33), for prospect rectangle sequence corresponding to each target trajectory, search suitable and translation and be placed into optimal placement interval:
(34) current goal trajectory-offset is placed into this optimal placement interval.
Further, described step (33) realizes as follows:
A) the overlapping area sum between the prospect rectangle sequence of existing other tracks in the prospect rectangle sequence of calculating current goal track and concentrated queue;
B) set the max-thresholds of overlapping area, lower than this threshold value, represent that this interval can hold current target trajectory, otherwise can not;
C) from the step-length traversal to fix to ending that starts of concentrated queue, record can hold all intervals of current goal track;
D) computing time cost: at c) in step, the current goal track mean value of the time difference between interval existing several tracks is therewith calculated respectively in each possible interval;
E) search the smallest positive integral value of above several time costs, the interval of recording its correspondence, interval as optimal placement.
Further, the summarized radio in described step (3) is the target trajectory of listing in result for retrieval by selecting, and foreground picture corresponding to a plurality of moving targets is spliced to same background and forms.
Further, foreground picture corresponding to a plurality of moving targets is spliced to the detailed process of same background as follows:
By mixed Gauss model method, calculate static background picture;
Exact position and the precise time of each foreground picture of each target trajectory recording according to target trajectory description document in original video picture, extracts the foreground picture sequence that target trajectory is corresponding;
From first to last traversal concentrates prospect rectangle corresponding to all target trajectorys in queue, realizes transparent pinup picture:
A), for not overlapped with other targets motion target area, the original pixels of direct copying moving target is to position corresponding to background picture;
B), for the overlapped region of a plurality of moving targets, each pixel value is the average of a plurality of moving target respective pixel values.
Active video method for concentration provided by the invention, target trajectory that can eliminate redundancy, represents the information of user's request, improves the investigation efficiency based on video.
Meanwhile, this programme can meet user to specifying specific objective to generate the demand of summarized radio, and generates summarized radio and take time and greatly shorten, and length, motion compactness during the summarized radio producing.
Accompanying drawing explanation
Below in conjunction with the drawings and specific embodiments, further illustrate the present invention.
Fig. 1 is the active video enrichment facility schematic diagram based on video structural description in the present invention;
Fig. 2 stores the schematic diagram of content in memory management module in active video enrichment facility;
Fig. 3 is the process flow diagram of the active video method for concentration based on video structural description in the present invention.
Embodiment
For technological means, creation characteristic that the present invention is realized, reach object and effect is easy to understand, below in conjunction with concrete diagram, further set forth the present invention.
Referring to Fig. 2, it is depicted as the structural representation of the active video enrichment facility based on video structural description.This active video enrichment facility can be inputted the condition of retrieval according to user's actual demand, and according to the result of its retrieval, generates the function of concentrated summarized radio.
As seen from the figure, this video enrichment facility 100 mainly comprises that video structural description module 101, memory management module 102, retrieval module 103 and video concentrate summarization module 104.
Wherein, video structural description module 101 is for input video is analyzed, and generating video structural description file.This video structural description file specifically comprises: target signature description document and target trajectory description document.The descriptor that wherein target signature description document comprises each target, for setting up relevant database; The essential information that target trajectory description document comprises each target, provides locus and the temporal information of prospect rectangle sequence for the concentrated summarization module 104 of video, to guarantee not overlapping after moving target translation or there is no an overlapping.
Referring to Fig. 2, it is depicted as the schematic diagram of the storage content of memory management module 102.This module stores management objectives feature description document 102a, target trajectory description document 102b and database 102c; Database 102c is set up according to target signature description document 102a by memory management module 102, and by this database 102c, is carried out all information of management objectives feature description document 102a.
Retrieval module 103 joins with memory management module 102 data, it provides a graphic user interface, user can be retrieved the video structural description information in memory management module 102 according to the demand of its actual conditions, and by result for retrieval according to its relevancy ranking.Wherein video structural description information comprises: the information recording in the target signature description document in video structural description file and target trajectory description document.
The concentrated summarization module 104 of video is joined with retrieval module 103 and memory management module 102 data respectively, for generating the summarized radio that meets search condition.Concrete, the concentrated summarization module 104 of this video result for retrieval that 103 retrievals are returned according to retrieval module, to target trajectory screening and rearrangement, locus and the temporal information of the prospect rectangle sequence providing according to target trajectory description document, guarantee not lose after moving target translation moving target and not overlapping or do not have under the prerequisite of overlapping, the moving target that different time sections is occurred moves to the same time period, thus simmer down to summarized radio.
This module, when concentrated summarized radio, guarantees that with the form of transparent pinup picture each target is visible for the region that a plurality of target trajectorys are overlapped; Each target in summarized radio is marked to its original timestamp and prospect rectangle simultaneously, realize thus video index function, click moving target and can jump to corresponding original video fragment.
Referring to Fig. 3, it is depicted as active video enrichment facility and concentrates the process flow diagram that forms summarized radio.As seen from the figure, whole flow process is divided into three large steps:
Step 1, input video is analyzed, by the structural description generating video structural description file to input video, can carry out the prerequisite that structural description is this method to video, existing video structural description technology normally realizes on industrial computer or server, certainly also has based on embedded platforms such as DSP, ARM and realizes video structural description.
The video structural description file that structural description generates mainly comprises two parts: target signature description document and target trajectory description document.
Wherein, target signature description document is for setting up relevant database, the descriptor that it comprises each moving target.This descriptor can include but not limited to: the color of target time of occurrence, place, license plate number, vehicle, vehicle, vehicle heading, whether in violation of rules and regulations, the information such as people's dress ornament feature, hair style, video origin url.
Target trajectory description document is for concentrated by summarized radio, and it comprises the prospect rectangle sequence information that each movement objective orbit is corresponding.This prospect rectangle sequence information can include but not limited to: the path of recording of video in disk or the URL of video are, the information such as sequence (sequence that rectangle position and size information form) that the number of each moving target prospect rectangle, the frame number in original video, original time stamp, prospect rectangle form.
After obtaining video structural description file, utilize target signature description document wherein to set up Relational database, and by all information of this data base administration target signature description document, comprising: the color of target time of occurrence, place, license plate number, vehicle, vehicle, vehicle heading, whether in violation of rules and regulations, the information such as people's dress ornament feature, hair style, video origin url.
Step 2, after Database, according to corresponding search condition, (specifically can by graphic user interface, input corresponding search condition by user, this search condition can be specifically the combination of the information that records in target signature description document 102a), generate SQL query statement, subsequently access searching database.Resolve after this result for retrieval, according to the correlativity of itself and search condition, sort, before the high result of correlativity is come, the result that concrete retrieval is returned can be according to the sequence number of the target trajectory of relevance ranking.
Step 3, the concentrated summarization module of video, according to the target trajectory sequence number providing in result for retrieval, filters out corresponding target trajectory from target trajectory description document; And by the concentrated summarized radio that forms of the target trajectory screening.
Particularly, this summarized radio is the target trajectory of listing in result for retrieval by selecting, and foreground picture corresponding to a plurality of moving targets is spliced to same background and forms.
This summarized radio is not to be formed by motion fragment simple concatenation, but highly concentrated video: the moving target that different time sections is occurred moves to the same time period, and simmer down to summarized radio, realizes the object that a plurality of targets are play simultaneously; Here to the translation of moving target, be the translation to the time of moving target, do not comprise the translation of locus.
It is specific as follows that the movement objective orbit that different time sections is occurred moves to the process of same time period:
1. the exact position of each foreground picture of each target trajectory recording according to target trajectory description document in original video picture, obtain the prospect rectangle sequence (rectangular area at moving target place that target trajectory is corresponding, in several video pictures, occurring continuously, forming prospect rectangle sequence corresponding to movement objective orbit);
2. the queue container of the newly-built prospect rectangle sequence for storage track, referred to as concentrated queue;
3. for prospect rectangle sequence corresponding to each target trajectory, search suitable and translation and be placed into optimal placement interval:
A) the overlapping area sum between the prospect rectangle sequence of existing other tracks in the prospect rectangle sequence of calculating current goal track and concentrated queue;
B) set the max-thresholds of overlapping area, lower than this threshold value, represent that this interval can hold current target trajectory, otherwise can not (can be understood as " punishment ");
C) from the step-length traversal to fix to ending that starts of concentrated queue, record can hold all intervals of current goal track;
D) computing time cost: at c) in step, the current goal track mean value of the time difference between interval existing several tracks (residual quantity of start time in original video with the contiguous track start time of current track in original video) is therewith calculated respectively in each possible interval;
E) search the smallest positive integral value of above several time costs, the interval of recording its correspondence, as optimal placement interval (can be understood as " award ");
4. current goal trajectory-offset is placed into this optimal placement interval.
All target trajectorys are placed into concentrated queue according to above method, can guarantee between track overlapping not serious, and original time order and function relation between basic maintenance and other tracks.
Meanwhile, foreground picture corresponding to a plurality of moving targets is spliced to the process of same background as follows:
1. by mixed Gauss model method, calculate static background picture;
2. exact position and the precise time of each foreground picture of each target trajectory recording according to target trajectory description document in original video picture, extract the foreground picture sequence that target trajectory is corresponding (the rectangle subgraph at moving target place, in several video pictures, occurring continuously, forming foreground picture sequence corresponding to movement objective orbit);
3. from first to last traversal concentrates prospect rectangle corresponding to all target trajectorys in queue, realizes transparent pinup picture:
A), for not overlapped with other targets motion target area, the original pixels of direct copying moving target is to position corresponding to background picture;
B), for the overlapped region of a plurality of moving targets, each pixel value is the average of a plurality of moving target respective pixel values.
Transparent pinup picture mode can be guaranteed above: the region that a plurality of target trajectorys are overlapped, each moving target is visible.
As from the foregoing, in generating the process of concentrated summarized radio, according to the target trajectory sequence number of result for retrieval record, filter out the moving target in now concentrated video.In order to guarantee higher activity density and enrichment factor, between time sequencing confusion degree is lower, track, collide overlapping still less such principle and be converted into " award " and " punishment ", adopt greedy algorithm, solve optimum sequence combination.Subsequently according to optimal alignment mode by originally not at one time the target trajectory of section move to same time section.Tight stacking inlaying between track, and adopt the form of transparent pinup picture to guarantee that the overlapping region of a plurality of targets is all visible separately.Brief and the movable intensive video that formation comprises all target trajectorys.The summarized radio forming like this makes full use of image spacing, represents all events of former video (target trajectory) with the shortest time.
Moreover, in the process that generates summarized radio, also according to target trajectory description document, in concentrated summary, to each target mark additional information, be specially each target mark original time stamp and prospect rectangle, realized thus video index function.The summarized radio that synchronous broadcasting has generated in the process that generates summarized radio, the prospect rectangle in the summarized radio having generated provides video index function simultaneously, clicks the former video segment that prospect rectangle can jump to its correspondence.
After last summarized radio has generated, the video can playback having generated, equally also provides video index function in playing process.
More than show and described ultimate principle of the present invention, principal character and advantage of the present invention.The technician of the industry should understand; the present invention is not restricted to the described embodiments; that in above-described embodiment and instructions, describes just illustrates principle of the present invention; without departing from the spirit and scope of the present invention; the present invention also has various changes and modifications, and these changes and improvements all fall in the claimed scope of the invention.The claimed scope of the present invention is defined by appending claims and equivalent thereof.
Claims (14)
1. an active video enrichment facility, is characterized in that, described device comprises:
Video structural description module, described video structural description module is used for input video to analyze, generating video structural description file;
Memory management module, described memory management module is used for storing and managing video structural description file, and according to the database of video structural description file set up;
Retrieval module, described retrieval module is retrieved the video structural description file in memory management module, and by result for retrieval according to its relevancy ranking;
Video concentrates summarization module, the concentrated summarization module of described video is for generating the summarized radio that meets search condition, do not losing moving target and not overlapping or do not have under overlapping prerequisite, the moving target that different time sections is occurred moves to the same time period, simmer down to summarized radio.
2. a kind of active video enrichment facility according to claim 1, it is characterized in that, the video structural description file that described video structural description module generates comprises target signature description document and target trajectory description document, the descriptor that described target signature description document comprises each target; The essential information that described target trajectory description document comprises each target trajectory.
3. a kind of active video enrichment facility according to claim 2, it is characterized in that, described memory management module is set up relevant database by the target signature description document in video structural description file, and the while is by all information of data base administration target signature description document.
4. a kind of active video enrichment facility according to claim 2, it is characterized in that, the concentrated summarization module of described video is when generating summarized radio, locus and the temporal information of the prospect rectangle sequence that track is corresponding are provided according to target trajectory description document, guarantee not overlapping after moving target translation or there is no an overlapping.
5. a kind of active video enrichment facility according to claim 1, is characterized in that, the concentrated summarization module of described video, when generating summarized radio, refers to the translation to the time of moving target to the translation of moving target.
6. a kind of active video enrichment facility according to claim 1, is characterized in that, the concentrated summarization module of described video, when generating summarized radio, guarantees that with the form of transparent pinup picture each target is visible for the region that a plurality of target trajectorys are overlapped; Each target in summarized radio is marked to its original timestamp and prospect rectangle simultaneously, and realized video index function, click moving target and can jump to corresponding original video fragment.
7. the concentrated method of active video, is characterized in that, described method comprises the steps:
(1) input video is analyzed, generating video structural description file, and according to the structural description file set up Relational database generating;
(2) database of setting up is retrieved, and by result for retrieval according to relevancy ranking;
(3), according to step (2) result for retrieval, to target trajectory screening and reset, not overlapping or do not have under the prerequisite of overlapping after guaranteeing moving target translation, the moving target that different time sections is occurred moves to the same time period, simmer down to summarized radio.
8. the concentrated method of a kind of active video according to claim 7, it is characterized in that, the video structural description file generating in described step (1) specifically comprises: target signature description document and target trajectory description document, the descriptor that described target signature description document comprises each target, the essential information that described target trajectory description document comprises each target.
9. the concentrated method of a kind of active video according to claim 8, it is characterized in that, in described step (1), utilize the target signature description document in video structural description file to set up relevant database, and by all information of Relational database management objectives feature description document; In described step (3), during concentrated summarized radio, the target trajectory description document in video structural description file provides locus and the temporal information of the prospect rectangle sequence that track is corresponding, guarantees not overlapping after moving target translation or there is no an overlapping.
10. the concentrated method of a kind of active video according to claim 7, is characterized in that, in described step (3), when concentrated summarized radio, for the region that a plurality of target trajectorys are overlapped, with the form of transparent pinup picture, guarantees that each target is visible; Each target in summarized radio is marked to its original timestamp and prospect rectangle, realize thus video index function; To the translation of moving target, it is the translation to the time of moving target.
11. according to the concentrated method of a kind of active video described in claim 7 or 10, it is characterized in that, in described step (3), to moving target, translation is adopted with the following method:
(31) exact position of each foreground picture of each target trajectory recording according to target trajectory description document in original video picture, obtains the prospect rectangle sequence that target trajectory is corresponding;
(32) queue container of the newly-built prospect rectangle sequence for storage track, i.e. concentrated queue;
(33), for prospect rectangle sequence corresponding to each target trajectory, search suitable and translation and be placed into optimal placement interval:
(34) current goal trajectory-offset is placed into this optimal placement interval.
The concentrated method of 12. a kind of active video according to claim 11, is characterized in that, described step (33) realizes as follows:
A) the overlapping area sum between the prospect rectangle sequence of existing other tracks in the prospect rectangle sequence of calculating current goal track and concentrated queue;
B) set the max-thresholds of overlapping area, lower than this threshold value, represent that this interval can hold current target trajectory, otherwise can not;
C) from the step-length traversal to fix to ending that starts of concentrated queue, record can hold all intervals of current goal track;
D) computing time cost: at c) in step, the current goal track mean value of the time difference between interval existing several tracks is therewith calculated respectively in each possible interval;
E) search the smallest positive integral value of above several time costs, the interval of recording its correspondence, interval as optimal placement.
13. according to the concentrated method of a kind of active video described in claim 7 or 10, it is characterized in that, summarized radio in described step (3) is the target trajectory of listing in result for retrieval by selecting, and foreground picture corresponding to a plurality of moving targets is spliced to same background and forms.
The concentrated method of 14. a kind of active video according to claim 13, is characterized in that, the detailed process that foreground picture corresponding to a plurality of moving targets is spliced to same background is as follows:
By mixed Gauss model method, calculate static background picture;
Exact position and the precise time of each foreground picture of each target trajectory recording according to target trajectory description document in original video picture, extracts the foreground picture sequence that target trajectory is corresponding;
From first to last traversal concentrates prospect rectangle corresponding to all target trajectorys in queue, realizes transparent pinup picture:
A), for not overlapped with other targets motion target area, the original pixels of direct copying moving target is to position corresponding to background picture;
B), for the overlapped region of a plurality of moving targets, each pixel value is the average of a plurality of moving target respective pixel values.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310611321.0A CN103617234B (en) | 2013-11-26 | 2013-11-26 | Active video enrichment facility and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310611321.0A CN103617234B (en) | 2013-11-26 | 2013-11-26 | Active video enrichment facility and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103617234A true CN103617234A (en) | 2014-03-05 |
| CN103617234B CN103617234B (en) | 2017-10-24 |
Family
ID=50167937
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310611321.0A Active CN103617234B (en) | 2013-11-26 | 2013-11-26 | Active video enrichment facility and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103617234B (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104093001A (en) * | 2014-07-23 | 2014-10-08 | 山东建筑大学 | Online dynamic video compression method |
| CN104252539A (en) * | 2014-09-24 | 2014-12-31 | 深圳市傲天智能系统有限公司 | Intelligent video analysis and management method and system based on search engine |
| CN104284158A (en) * | 2014-10-23 | 2015-01-14 | 南京信必达智能技术有限公司 | Event-oriented intelligent camera monitoring method |
| CN104394379A (en) * | 2014-12-05 | 2015-03-04 | 北京厚吉科技有限公司 | Fast previewing system and fast viewing method of surveillance video |
| CN104484457A (en) * | 2014-12-29 | 2015-04-01 | 广州中国科学院软件应用技术研究所 | Method and system for extracting and searching moving object in parallel video |
| CN104581437A (en) * | 2014-12-26 | 2015-04-29 | 中通服公众信息产业股份有限公司 | Video abstract generation and video backtracking method and system |
| WO2015117572A1 (en) * | 2014-07-28 | 2015-08-13 | 中兴通讯股份有限公司 | Labelling method for moving objects of concentrated video, and playing method and device |
| CN106803937A (en) * | 2017-02-28 | 2017-06-06 | 兰州理工大学 | A kind of double-camera video frequency monitoring method and system with text log |
| US20170337429A1 (en) | 2016-05-23 | 2017-11-23 | Axis Ab | Generating a summary video sequence from a source video sequence |
| WO2018149376A1 (en) * | 2017-02-17 | 2018-08-23 | 杭州海康威视数字技术股份有限公司 | Video abstract generation method and device |
| CN108446620A (en) * | 2018-03-13 | 2018-08-24 | 成都蓉兴伟业科技有限公司 | A kind of Intelligent target method for tracing based on video |
| CN110519532A (en) * | 2019-09-02 | 2019-11-29 | 中移物联网有限公司 | A kind of information acquisition method and electronic equipment |
| WO2020038243A1 (en) * | 2018-08-21 | 2020-02-27 | 腾讯科技（深圳）有限公司 | Video abstract generating method and apparatus, computing device, and storage medium |
| CN111464882A (en) * | 2019-01-18 | 2020-07-28 | 杭州海康威视数字技术股份有限公司 | Video abstract generation method, device, equipment and medium |
| US11076132B2 (en) | 2017-12-15 | 2021-07-27 | Zhejiang Dahua Technology Co., Ltd. | Methods and systems for generating video synopsis |
| CN113326399A (en) * | 2021-06-16 | 2021-08-31 | 中国人民解放军91388部队 | Video concentration method and device, electronic equipment and storage medium |
| CN113949823A (en) * | 2021-09-30 | 2022-01-18 | 广西中科曙光云计算有限公司 | Video concentration method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007126780A2 (en) * | 2006-03-28 | 2007-11-08 | Object Video, Inc. | Automatic extraction of secondary video streams |
| CN102256265A (en) * | 2011-08-12 | 2011-11-23 | 电子科技大学 | Channel initialization method for distributed cognitive radio network |
| CN102708182A (en) * | 2012-05-08 | 2012-10-03 | 浙江捷尚视觉科技有限公司 | Rapid video concentration abstracting method |
| US20120278827A1 (en) * | 2011-04-26 | 2012-11-01 | International Business Machines Corporation | Method for Personalized Video Selection |
| CN103150375A (en) * | 2013-03-11 | 2013-06-12 | 浙江捷尚视觉科技有限公司 | Quick video retrieval system and quick video retrieval method for video detection |
-
2013
- 2013-11-26 CN CN201310611321.0A patent/CN103617234B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007126780A2 (en) * | 2006-03-28 | 2007-11-08 | Object Video, Inc. | Automatic extraction of secondary video streams |
| US20120278827A1 (en) * | 2011-04-26 | 2012-11-01 | International Business Machines Corporation | Method for Personalized Video Selection |
| CN102256265A (en) * | 2011-08-12 | 2011-11-23 | 电子科技大学 | Channel initialization method for distributed cognitive radio network |
| CN102708182A (en) * | 2012-05-08 | 2012-10-03 | 浙江捷尚视觉科技有限公司 | Rapid video concentration abstracting method |
| CN103150375A (en) * | 2013-03-11 | 2013-06-12 | 浙江捷尚视觉科技有限公司 | Quick video retrieval system and quick video retrieval method for video detection |
Non-Patent Citations (1)
| Title |
|---|
| 庞志恒 等: ""视频浓缩与基于对象检索技术研究"", 《科技信息》, 15 January 2013 (2013-01-15), pages 62 * |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104093001B (en) * | 2014-07-23 | 2017-05-03 | 山东建筑大学 | Online dynamic video compression method |
| CN104093001A (en) * | 2014-07-23 | 2014-10-08 | 山东建筑大学 | Online dynamic video compression method |
| WO2015117572A1 (en) * | 2014-07-28 | 2015-08-13 | 中兴通讯股份有限公司 | Labelling method for moving objects of concentrated video, and playing method and device |
| CN104252539A (en) * | 2014-09-24 | 2014-12-31 | 深圳市傲天智能系统有限公司 | Intelligent video analysis and management method and system based on search engine |
| CN104284158A (en) * | 2014-10-23 | 2015-01-14 | 南京信必达智能技术有限公司 | Event-oriented intelligent camera monitoring method |
| CN104394379B (en) * | 2014-12-05 | 2017-10-13 | 北京厚吉科技有限公司 | Monitoring video rapid preview system and rapid preview method |
| CN104394379A (en) * | 2014-12-05 | 2015-03-04 | 北京厚吉科技有限公司 | Fast previewing system and fast viewing method of surveillance video |
| 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 |
| CN104581437A (en) * | 2014-12-26 | 2015-04-29 | 中通服公众信息产业股份有限公司 | Video abstract generation and video backtracking method and system |
| CN104484457A (en) * | 2014-12-29 | 2015-04-01 | 广州中国科学院软件应用技术研究所 | Method and system for extracting and searching moving object in parallel video |
| US20170337429A1 (en) | 2016-05-23 | 2017-11-23 | Axis Ab | Generating a summary video sequence from a source video sequence |
| US10192119B2 (en) | 2016-05-23 | 2019-01-29 | Axis Ab | Generating a summary video sequence from a source video sequence |
| WO2018149376A1 (en) * | 2017-02-17 | 2018-08-23 | 杭州海康威视数字技术股份有限公司 | Video abstract generation method and device |
| CN108460032A (en) * | 2017-02-17 | 2018-08-28 | 杭州海康威视数字技术股份有限公司 | A kind of generation method and device of video frequency abstract |
| CN106803937A (en) * | 2017-02-28 | 2017-06-06 | 兰州理工大学 | A kind of double-camera video frequency monitoring method and system with text log |
| US11076132B2 (en) | 2017-12-15 | 2021-07-27 | Zhejiang Dahua Technology Co., Ltd. | Methods and systems for generating video synopsis |
| CN108446620A (en) * | 2018-03-13 | 2018-08-24 | 成都蓉兴伟业科技有限公司 | A kind of Intelligent target method for tracing based on video |
| WO2020038243A1 (en) * | 2018-08-21 | 2020-02-27 | 腾讯科技（深圳）有限公司 | Video abstract generating method and apparatus, computing device, and storage medium |
| US11347792B2 (en) | 2018-08-21 | 2022-05-31 | Tencent Technology (Shenzhen) Company Limited | Video abstract generating method, apparatus, and storage medium |
| CN111464882A (en) * | 2019-01-18 | 2020-07-28 | 杭州海康威视数字技术股份有限公司 | Video abstract generation method, device, equipment and medium |
| CN111464882B (en) * | 2019-01-18 | 2022-03-25 | 杭州海康威视数字技术股份有限公司 | Video abstract generation method, device, equipment and medium |
| CN110519532A (en) * | 2019-09-02 | 2019-11-29 | 中移物联网有限公司 | A kind of information acquisition method and electronic equipment |
| CN113326399A (en) * | 2021-06-16 | 2021-08-31 | 中国人民解放军91388部队 | Video concentration method and device, electronic equipment and storage medium |
| CN113949823A (en) * | 2021-09-30 | 2022-01-18 | 广西中科曙光云计算有限公司 | Video concentration method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103617234B (en) | 2017-10-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103617234A (en) | Device and method for active video concentration | |
| CN101821734B (en) | Detection and classification of matches between time-based media | |
| Aigrain et al. | Content-based representation and retrieval of visual media: A state-of-the-art review | |
| US10277946B2 (en) | Methods and systems for aggregation and organization of multimedia data acquired from a plurality of sources | |
| Adcock et al. | Talkminer: a lecture webcast search engine | |
| Liu et al. | Near-duplicate video retrieval: Current research and future trends | |
| US7181757B1 (en) | Video summary description scheme and method and system of video summary description data generation for efficient overview and browsing | |
| Saba et al. | Analysis of vision based systems to detect real time goal events in soccer videos | |
| KR100547370B1 (en) | Apparatus and method for abstracting summarization video using shape information of object, and video summarization and indexing system and method using the same | |
| Jiang et al. | Spatial and temporal content-based access to hypervideo databases | |
| Li et al. | Bridging the semantic gap in sports video retrieval and summarization | |
| CN101587483B (en) | Information storage/retrival method of video files based on MPEG-7 in monitoring system | |
| CN104021140A (en) | Network video processing method and device | |
| Jensen et al. | Valid Time. | |
| Ang et al. | LifeConcept: an interactive approach for multimodal lifelog retrieval through concept recommendation | |
| CN104750853A (en) | Method and device for searching heterogeneous data | |
| KR101640317B1 (en) | Apparatus and method for storing and searching image including audio and video data | |
| CN110933520B (en) | Monitoring video display method based on spiral abstract and storage medium | |
| US10534812B2 (en) | Systems and methods for digital asset organization | |
| Kim et al. | Geosearch: georeferenced video retrieval system | |
| Karray et al. | Indexing video summaries for quick video browsing | |
| Geisler et al. | Crowdsourcing the indexing of film and television media | |
| Dönderler | Data modeling and querying for video databases | |
| Dao et al. | Sports event detection using temporal patterns mining and web-casting text | |
| Li et al. | Bridging the semantic gap in sports |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |