CN103686453A - Method for improving video abstract accuracy by dividing areas and setting different particle sizes - Google Patents
Method for improving video abstract accuracy by dividing areas and setting different particle sizes Download PDFInfo
- Publication number
- CN103686453A CN103686453A CN201310713685.XA CN201310713685A CN103686453A CN 103686453 A CN103686453 A CN 103686453A CN 201310713685 A CN201310713685 A CN 201310713685A CN 103686453 A CN103686453 A CN 103686453A
- Authority
- CN
- China
- Prior art keywords
- video
- different
- need
- target
- time
- 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.)
- Pending
Links
Images
Abstract
The invention discloses a method for improving video abstract accuracy by dividing areas and setting different particle sizes. Before a video abstract is processed, according to practical situations, the method judges whether different areas need to be set for an original video or not, whether different analysis particle sizes need to be set for the different areas or not and whether occurrence time of a target in the original video needs to be specially concerned or not, so that the processing application range of the video abstract is expanded. By the method for improving video abstract accuracy by dividing the areas and setting the different particle sizes, video abstract technology is applicable to more scenes, and some special requirements are met when the video is viewed.
Description
Technical field
The invention belongs to technical field of video processing, relate to a kind of video summarization method, in particular to by zoning and the method that different grain size improves video frequency abstract precision is set.
Background technology
Current social is to unprecedented attention that keep the peace, and Video Surveillance Industry is fast-developing, and the camera number that domestic every thousand people have is over 25 and in quick increase, and the monitor video of magnanimity produces.When maintaining public order, also for checking of magnanimity video, brought very large difficulty.Fast finding useful information from magnanimity video how?
At present, public security has started to improve video by video summarization technique and has checked efficiency, with a setting analysis granularity, the moving target in picture is taken passages, and neglects the picture that there is no moving target.But, in picture, often exist a lot of disturb as leaf waves, zones of different target sizes differs, while checking video to target time of occurrence degree of concern difference, make existing video frequency abstract application scenarios limited.If zones of different can adopt, different to analyze granularities be different sensitivity analysis, and the mode of video frequency abstract can select according to actual conditions, will greatly expand the range of application of video summarization technique, improves the efficiency of video frequency abstract.
Summary of the invention
For overcoming shortcoming of the prior art, the object of the present invention is to provide by zoning and the method that different grain size improves video frequency abstract precision is set, this video summarization method solves the deficiency of current video summarization technique in actual application, the range of application of expansion video summarization technique, improves video and checks efficiency.
For realizing above-mentioned technical purpose, reach above-mentioned technique effect, the present invention is achieved through the following technical solutions:
By zoning and the method that different grain size improves video frequency abstract precision is set, before carrying out video frequency abstract processing, according to actual conditions, judge whether different regions to be set for original video, whether need, for zones of different, is set the different time of analyzing granularities and whether needing special concern target to occur in original video, with this, expand the range of application that video frequency abstract is processed, specifically comprise the following steps:
Step 1) original video imports
Import the required original video of checking; Preferably, described original video adopts the method importing in batches
.
Step 2) whether need region division
According to actual needs, confirm whether described original video needs to arrange interested and uninterested region, if do not need to carry out region division, directly judge whether the time that needs special concern target to occur in described original video.
Step 3) region division
If need setting area, interested and uninterested region is set on picture; Preferably, described region division is polygonal region.
Whether step 4) needs to analyze granularity setting
According to the situation of video, determine whether zones of different needs to arrange the different granularities of analyzing, with the different sensitivity of correspondence.If do not need to arrange the different granularities of analyzing, directly judge whether the time that needs special concern target to occur in described original video.
Step 5) is analyzed granularity setting
If analysis granularity need to be set, it is the different different analysis granularities of region division.
Step 6) judges whether the time that needs special concern target to occur in original video
When confirmation is checked at video, be only to need to verify whether certain target passes through certain picture, or need definite the understand time point of target appearance and the environmental aspect of front and back.
Step 7) is selected different video frequency abstract modes
If need time that special concern moving target occurs in described original video, adopt the summarize by without moving target F.F.; If the time that moving target is occurred does not need to pay close attention to, adopt the summarize by of space-time superimposed type.
Step 8) video frequency abstract is processed
According to selection and setting above, the described original video importing is carried out to video frequency abstract processing, and show result.
Further, described driftlessness F.F. summarize by refers in described original video, without moving target picture, adopt fast and play, the summarize by that has moving target picture to adopt constant speed to play.
Further, the summarize by of described space-time superimposed type refers to that the target array that different time can be occurred is presented in same picture.
The invention has the beneficial effects as follows:
1, zones of different adopts different analysis granularities to carry out video frequency abstract, gets rid of unnecessary interference, accurately detected activity target.
2, according to the difference to moving target time of occurrence attention rate, and different user habits, provide two kinds of video frequency abstract modes that respectively have superiority to select for user.
Above-mentioned explanation is only the general introduction of technical solution of the present invention, in order to better understand technological means of the present invention, and can be implemented according to the content of specification, below with preferred embodiment of the present invention and coordinate accompanying drawing to be described in detail as follows.The specific embodiment of the present invention is provided in detail by following examples and accompanying drawing thereof.
Accompanying drawing explanation
Accompanying drawing described herein is used to provide a further understanding of the present invention, forms the application's a part, and schematic description and description of the present invention is used for explaining the present invention, does not form inappropriate limitation of the present invention.In the accompanying drawings:
Fig. 1 is of the present invention by zoning and the schematic flow sheet that different grain size improves the method for video frequency abstract precision is set.
Embodiment
Below with reference to the accompanying drawings and in conjunction with the embodiments, describe the present invention in detail.
Shown in Figure 1, by zoning and the method that different grain size improves video frequency abstract precision is set, it comprises the following steps:
Step 1) original video imports
Import the required original video of checking.Video can import in batches.
Step 2) whether need region division
According to actual needs, confirm whether described original video needs to arrange interested and uninterested region, if do not need to carry out region division, directly judge whether the time that needs special concern target to occur in described original video.The picture that for example imports video is positioned at a T-shaped road junction, has traffic lights not stop flicker.At this moment can select be concerned about crossing as area-of-interest, also traffic lights region can be fallen as the regions shield of loseing interest in.
Step 3) region division
If need setting area, interested and uninterested region is set on picture; Preferably, described region division is polygonal region;
Whether step 4) needs to analyze granularity setting
According to the situation of video, determine that whether zones of different needs to arrange the different granularities of analyzing, and that is to say and adopts different sensitivity to analyze.If do not need to arrange the different granularities of analyzing, directly judge whether the time that needs special concern target to occur in described original video.Above-mentioned T-shaped road junction picture for example, near-end, because object is comparatively obvious, can be selected lower analysis granularity, masks some and does not need that wind leaves of paying close attention to are moving waits slight change.Far-end, because target is less, needs to adopt higher analysis granularity, does not ignore any details.
Step 5) is analyzed granularity setting
If analysis granularity need to be set, it is the different different analysis granularities of region division.
Step 6) judges whether the time that needs special concern target to occur in original video
When confirmation is checked at video, be only to need certain target of checking whether to pass through certain picture, still need to know the time point of target appearance and the environmental aspect of front and back.When video is checked, often in order to verify whether certain target passes through certain picture, at this time as long as confirm whether target occurs at picture; But also sometimes need hard objectives to occur the environment of front and back, require the target in summarized radio successively to arrange by time of occurrence.
Step 7) is selected different video frequency abstract modes
If need time that special concern moving target occurs in described original video, adopt without moving target F.F. summarize by; If the time that moving target is occurred does not need special concern, adopt the summarize by of space-time superimposed type.
Step 8) video frequency abstract is processed
According to selection and setting above, the magnanimity monitor video importing is carried out to video frequency abstract processing, and show result.
embodiment 1:
After case occurs, tend to transfer a large amount of videos and carry out naked eyes and check or adopt video frequency abstract mode to process.Existing video summarization technique, when processing, often in the concentrated mode of space-time, is processed the video of a whole probe with a certain analysis granularity, some scene application is limited.If passing through zoning and arranging the method that different grain size improves video frequency abstract precision can be how to process by the present embodiment?
Step 1) imports the required original video of checking in batches.
Need to step 2), according to the actual conditions of video, be confirmed whether arrange the interested or region of loseing interest in? the picture that for example camera is taken is positioned at a T-shaped road junction, has a tree because blowing of wind do not stop waving.At this moment can only select the crossing of being concerned about as area-of-interest, also the tree shaking can be fallen as the regions shield of loseing interest in.Behind setting area, can reduce the operand of system, improve system processing speed, reduce the number of targets of disturbing, improve the concentrated summary ratio of video.
Step 3), according to above-mentioned selection, arranges the interested/region of loseing interest in.
Step 4), according to the situation of video, determines whether zones of different needs to arrange the different granularities of analyzing.Above-mentioned T-shaped road junction picture for example, near-end, because object is comparatively obvious, can be selected lower analysis granularity, and namely lower sensitivity masks some and does not need the variation of paying close attention to as leaf swing etc., gets rid of and disturbs; And far-end is because target is less, need to adopt higher analysis granularity, adopt higher sensitivity for analysis, can not miss any details.
Step 5), according to above-mentioned selection, is the different analysis granularity of different region divisions;
Step 6) is confirmed when video is checked, the concern situation to moving target time of occurrence.Check the object of video, whether checking has interested target through picture often, at this moment as long as confirm whether target occurs at picture; But also sometimes, need hard objectives to occur the environment of front and back, require the target in summarized radio successively to arrange by time of occurrence.
Step 7), according to the above-mentioned situation that the time is paid close attention to, is selected different summarize bies.If the time that moving target is occurred is not special concern, adopt the summarize by of space-time superimposed type, the target that different time occurs may be arranged in same picture, and can regulate target density in picture, reduce in this way the length of summarized radio, obtain larger cocnentration factor; If the time that special concern moving target occurs, adopts driftlessness picture fast-forward play, the mode that has target picture constant speed to play is carried out video frequency abstract, and target is occurred sequentially.Along with the carrying out of video investigation, selected summarize by likely changes.The investigation initial stage looks for a needle in a haystack often, searches a certain target in magnanimity video, should adopt the summarize by of space-time stack, shortens and checks video time; If there has been certain clue in the later stage, in order to verify that whether target passes through a certain region at a certain special time, can adopt the summarize by of driftlessness F.F..
Selection and the setting above of step 8) basis, carries out video frequency abstract processing to the magnanimity monitor video importing, and shows result.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for a person skilled in the art, the present invention can have various modifications and variations.Within the spirit and principles in the present invention all, any modification of doing, be equal to replacement, improvement etc., within all should being included in protection scope of the present invention.
Claims (5)
1. by zoning and the method that different grain size improves video frequency abstract precision is set, it is characterized in that, before carrying out video frequency abstract processing, according to actual conditions, judge whether different regions to be set for original video, whether need different analysis granularities are set and whether need to pay close attention to the time that target occurs in original video for zones of different, with this, expand the range of application that video frequency abstract is processed, specifically comprise the following steps:
Step 1) original video imports
Import the required original video of checking;
Step 2) whether need region division
According to actual needs, confirm whether described original video needs to arrange interested and uninterested region, if do not need to carry out region division, directly judge whether the time that needs special concern target to occur in described original video;
Step 3) region division
If need setting area, interested and uninterested region is set on picture;
Whether step 4) needs to analyze granularity setting
According to the situation of video, determine whether zones of different needs to arrange the different granularities of analyzing, and the different granularities of analyzing are by the different sensitivity for analysis of correspondence;
If do not need to arrange the different granularities of analyzing, directly judge whether the time that needs special concern target to occur in described original video;
Step 5) is analyzed granularity setting
If analysis granularity need to be set, it is the different different analysis granularities of region division;
Step 6) judges whether the time that needs special concern target to occur in original video
When confirmation is checked at video, be only to need certain target of checking whether to pass through certain picture, still need the time point of special concern target appearance and the environmental aspect of front and back;
Step 7) is selected different video frequency abstract modes
If the time that needs special concern target to occur in described original video, adopted without moving target picture F.F. summarize by; If the time that moving target occurs does not need special concern, adopt the summarize by of space-time superimposed type;
Step 8) video frequency abstract is processed
According to selection and setting above, the described original video importing is carried out to video frequency abstract processing, and show result.
2. according to claim 1 by zoning and the method that different grain size improves video frequency abstract precision is set, it is characterized in that: original video described in step 1 adopts the method importing in batches
.
3. according to claim 1 by zoning and the method that different grain size improves video frequency abstract precision is set, it is characterized in that: in step 3, described region division is polygonal region.
4. according to passing through zoning and the method that different grain size improves video frequency abstract precision be set described in claim 1 or 2 or 3, it is characterized in that: described driftlessness F.F. summarize by refers to that driftlessness picture adopts broadcasting fast, the summarize by that has target picture to adopt constant speed to play in described original video.
5. according to passing through zoning and the method that different grain size improves video frequency abstract precision is set described in claim 1 or 2 or 3, it is characterized in that: the summarize by of described space-time superimposed type refers to that the target array that different time can be occurred is presented in same picture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310713685.XA CN103686453A (en) | 2013-12-23 | 2013-12-23 | Method for improving video abstract accuracy by dividing areas and setting different particle sizes |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310713685.XA CN103686453A (en) | 2013-12-23 | 2013-12-23 | Method for improving video abstract accuracy by dividing areas and setting different particle sizes |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103686453A true CN103686453A (en) | 2014-03-26 |
Family
ID=50322512
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310713685.XA Pending CN103686453A (en) | 2013-12-23 | 2013-12-23 | Method for improving video abstract accuracy by dividing areas and setting different particle sizes |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103686453A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104038705A (en) * | 2014-05-30 | 2014-09-10 | 无锡天脉聚源传媒科技有限公司 | Video producing method and device |
| CN104270608A (en) * | 2014-09-28 | 2015-01-07 | 武汉烽火众智数字技术有限责任公司 | Intelligent video player and playing method thereof |
| CN105530554A (en) * | 2014-10-23 | 2016-04-27 | 中兴通讯股份有限公司 | Video abstraction generation method and device |
| CN107729809A (en) * | 2017-09-11 | 2018-02-23 | 汕头大学 | A kind of method, apparatus and its readable storage medium storing program for executing of adaptive generation video frequency abstract |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005017899A1 (en) * | 2003-08-18 | 2005-02-24 | Koninklijke Philips Electronics N.V. | Video abstracting |
| CN102750383A (en) * | 2012-06-28 | 2012-10-24 | 中国科学院软件研究所 | Spiral abstract generation method oriented to video content |
| CN103096185A (en) * | 2012-12-30 | 2013-05-08 | 信帧电子技术（北京）有限公司 | Method and device of video abstraction generation |
-
2013
- 2013-12-23 CN CN201310713685.XA patent/CN103686453A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005017899A1 (en) * | 2003-08-18 | 2005-02-24 | Koninklijke Philips Electronics N.V. | Video abstracting |
| CN102750383A (en) * | 2012-06-28 | 2012-10-24 | 中国科学院软件研究所 | Spiral abstract generation method oriented to video content |
| CN103096185A (en) * | 2012-12-30 | 2013-05-08 | 信帧电子技术（北京）有限公司 | Method and device of video abstraction generation |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104038705A (en) * | 2014-05-30 | 2014-09-10 | 无锡天脉聚源传媒科技有限公司 | Video producing method and device |
| CN104270608A (en) * | 2014-09-28 | 2015-01-07 | 武汉烽火众智数字技术有限责任公司 | Intelligent video player and playing method thereof |
| CN105530554A (en) * | 2014-10-23 | 2016-04-27 | 中兴通讯股份有限公司 | Video abstraction generation method and device |
| CN105530554B (en) * | 2014-10-23 | 2020-08-07 | 南京中兴新软件有限责任公司 | Video abstract generation method and device |
| CN107729809A (en) * | 2017-09-11 | 2018-02-23 | 汕头大学 | A kind of method, apparatus and its readable storage medium storing program for executing of adaptive generation video frequency abstract |
| CN107729809B (en) * | 2017-09-11 | 2021-06-08 | 汕头大学 | Method and device for adaptively generating video abstract and readable storage medium thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104813357B (en) | For the matched system and method for live media content | |
| CN102708182B (en) | Rapid video concentration abstracting method | |
| CN103686453A (en) | Method for improving video abstract accuracy by dividing areas and setting different particle sizes | |
| Muhammad et al. | Intelligent and energy-efficient data prioritization in green smart cities: Current challenges and future directions | |
| KR101606100B1 (en) | Video synopsis method and apparatus | |
| CN103347167A (en) | Surveillance video content description method based on fragments | |
| CN103210651A (en) | Method and system for video summarization | |
| CN103942811A (en) | Method and system for determining motion trajectory of characteristic object in distributed and parallel mode | |
| TW202036375A (en) | Method and device for real-time statistical analysis of pedestrian flow in open space | |
| WO2020259449A1 (en) | Method and device for generating short video | |
| CN102890866B (en) | Traffic flow speed estimation method based on multi-core support vector regression machine | |
| EP2735984A1 (en) | Video query method, device and system | |
| Taylor et al. | CCTV: Reflections on its use, abuse and effectiveness | |
| CN109872541A (en) | A kind of information of vehicles analysis method and device | |
| EP3565258A1 (en) | Computerized system and method for in-video modification | |
| CN105138525A (en) | Traffic video processing device and method, and retrieval device and method | |
| CN106570147A (en) | GIS road network analysis-based jump type video tracking method and system | |
| CN108734022A (en) | The secret protection track data dissemination method divided based on three-dimensional grid | |
| CN103605956A (en) | Emergency intelligence identification system based on video abstraction | |
| Zhu et al. | On development of security monitoring system via wireless sensing network | |
| CN107133574A (en) | A kind of vehicle feature recognition method and device | |
| CN109003122A (en) | A kind of user classification method and server launched based on advertisement | |
| CN102034243B (en) | Method and device for acquiring crowd density map from video image | |
| CN105205819A (en) | Bus detection method based on hierarchical structure | |
| Li | Dilemma and solution of copyright justice system for health short video under network big data environment monitoring |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 215000 No. 209 Chuk Yuen Road, hi tech Industrial Development Zone, Jiangsu, Suzhou Applicant after: Suzhou vision Polytron Technologies Inc Address before: 215000 No. 209 Chuk Yuen Road, hi tech Industrial Development Zone, Jiangsu, Suzhou Applicant before: SUZHOU QIANSHI COMMUNICATION TECHNOLOGY CO., LTD. |
|
| COR | Change of bibliographic data | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140326 |
|
| WD01 | Invention patent application deemed withdrawn after publication |