CN105898313A - Novel video synopsis-based monitoring video scalable video coding technology - Google Patents
Novel video synopsis-based monitoring video scalable video coding technology Download PDFInfo
- Publication number
- CN105898313A CN105898313A CN201410782676.0A CN201410782676A CN105898313A CN 105898313 A CN105898313 A CN 105898313A CN 201410782676 A CN201410782676 A CN 201410782676A CN 105898313 A CN105898313 A CN 105898313A
- Authority
- CN
- China
- Prior art keywords
- video
- technology
- outline
- scalable
- cost
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/30—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using hierarchical techniques, e.g. scalability
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention relates to a novel video synopsis-based monitoring video scalable video coding technology. According to the technology, a large amount of redundant information in a video is removed through feature recognition, so that video compression can be achieved. Scalable video coding is a scheme which can well adapt to the diversity of terminals and dynamic heterogeneity of a network. According to the scalable video coding, only a high-quality video source is required to be subjected to encoding for once, so that top-level video streams can be obtained, and specific portions are extracted as basic quality code streams, when a terminal does not support the top-level video streams or the network is congested, only the basic quality code streams can be transmitted to the terminal. According to the novel video synopsis-based monitoring video scalable video coding technology of the invention, an area-of-interest is extracted from an image acquired by a vision sensor through an image segmentation algorithm, and non-useful information is removed, so that the calculation amount of subsequent work can be decreased; in an image segmentation process, the edge information of an object is identified mainly, an ROI coding strategy is adopted, and therefore, the problem of a complex calculation process caused by the redundancy of feature information can be solved; and a coding monitoring video is analyzed through adopting a detail retaining-based video synopsis technology, so that the regional information and mapping information of a moving object can be obtained, and the ROI and strip types in video coding can be determined.
Description
Technical field
The present invention relates to video information compression technology.
Background technology
Occur in that the control terminal of vehicle-mounted band multi-channel video monitor the most both at home and abroad, but most simply supervision, actively Deviant Behavior situation active analysis inside and outside car can not be reported to the police, such as driver dozes off, child old man anemia of pregnant woman falls down, by the photographic head Quick Acquisition image/video information being erected on bus, key-frame extraction can be carried out, moving target is analyzed, video information compression, image-fused-reconstruction, may finally be in car, the key positions such as luggage compartment position and follow the tracks of, utilize artificial intelligence technology, the number of getting on the bus can be identified, passenger's sex and substantially age, and then preferably manage and control all situations in car.
The mass data caused that is significantly increased of widely available along with Video Applications, monitor video quantity and definition makes it browse and storage becomes very stubborn problem.On the one hand, complete view takes considerable time and manpower up to the video council of several days, and F.F. is easily caused again the omission of important information；On the other hand, not Moving Objects in most of frame of monitor video, a lot of browsing times can be wasted, even if occurring in that Moving Objects, also tends to be distributed the most sparse, and the image spacing of every frame exists the biggest redundancy.Therefore, the time-space domain redundancy of video is the most rationally eliminated, and the target of Video coding is also by the process display capabilities pursuing merely high compression rate and having turned to the network bandwidth and the different terminals adapting to more diversification.Scalable coding is a kind of scheme that can adapt to terminal multiformity and the dynamic heterogeneous sex chromosome mosaicism of network very well, have only to high-quality video source is carried out first encoding, obtain top video flowing, specific part is extracted as gross code stream from this code stream, when terminal does not support high layer bit stream or network congestion, it is possible to only transmission gross code stream is to terminal.Existing video outline technology cannot carry out scalable browsing to the outline information of monitor video, and the protection to key component fidelity is the most unsatisfactory.Therefore, explore the scalable coding technology of new video information based on video outline technology, and apply it to, in vehicle-mounted intelligent monitoring system, there is important theory significance and using value.
Summary of the invention
Present invention is primarily intended to be solved detection identification and the warning of bus accident by image/video technology.Mission critical is intended to propose the video outline technology of a kind of practicality, by the analysis to complete video, extracts main Moving Objects and background picture, map and rebuild, generate an outline video and as much as possible retain raw information.On the one hand, owing to outline video eliminates the time-space domain redundancy of original video, length is greatly shortened, and therefore alleviates coding, transmits and store burden.On the other hand, outline video remains the main movement information of original video, easily video can be carried out fast browsing and retrieval in the face of growing massive video data, how video outline technology improves the fidelity of Key detail in limited memory space as far as possible, how to meet user and inquire and browse demand, the efficiency how improving encoding and decoding the most flexibly, reduce the pressure of memory space and transmission network the most further, these problems have become the hot issue of MultiMedia Field, have far-reaching Research Significance and are widely applied prospect.
On the basis of the research and analysis to conventional video outline technology, tradition scalable video technology and scalable video technology based on ROI, a kind of new monitor video scalable coding technology based on video outline is proposed.It is a kind of video outline technology retained based on details, the key component of original video is used different concentration strength with non-key part, and correspondingly update the computational methods of cost function in video outline, so would be beneficial for showing most possibly key message under identical storage cost, being expected to be greatly improved picture compression efficiency, integral monitoring video compress structure is shown in Fig. 1.
How showing key message under identical storage cost most possibly, picture compression efficiency is greatly improved, and this is the difficult problem of main solution of project.At present video-frequency compression method is computationally intensive, operation time length, stability and feature extraction poor performance, it is impossible to reliability when meeting on-board running, quick requirement of real-time.Change time-space domain mapping relations and can obtain the concentration video of more high compression degree.But one section concentrate video length once it is determined that, inevitably there is fracture and the loss in detail of a lot of movement locus in outline video, wherein it is no lack of some and can show entrance, walk out, the time detail of key message such as meet, and this situation have impact on the ability of outline video reduction original video information to a great extent.How details being taken into account, improving compression efficiency is also a major issue.
The coding flow process of whole scheme is as follows: first the monitor video to be encoded video outline technology retained based on details is carried out outline analysis, obtains Moving Objects area information and Moving Objects map information；The ROI region in Video coding is determined based on subject area information, the type of strip in Video coding is determined based on object map information, the present invention will be mapped to the Moving Objects in initial outline video and background is encoded to Pslice, is not mapped to the Moving Objects in outline video by initializing and background is encoded to B slice；It is then based on this encoding scheme and original monitor video is carried out scalable video based on ROI, carry out encoding and merge into object flag position by Moving Objects flag bit and moving region flag bit simultaneously；Ultimately generating the scalable bitstream containing object flag position, thus realize the scalable transmission of monitor video code stream and scalable browse, structure is shown in Fig. 2.
The maximum advantage of video outline based on Moving Objects is exactly Moving Objects itself and movement locus is monitored by the unit complete as, analyzes and shows, therefore the real-time of moving object extract, integrity and accuracy are the keys of system realization.Optical flow method, frame difference method, background subtraction method were the most once applied to video outline system as object extraction method, but considering algorithm complex, extract degree of accuracy and integrity, the present invention compromises and selects mixed Gauss model background modeling method based on pixel to extract.
Moving object tracking determines that track, behavior analysis and the basis determining mapping function.The most common method has the particle filter that the average drift proposed by Fukunag and Gordon et al propose, and wherein Mean Shift belongs to deterministic algorithm, and computation complexity is low；Particle filter belongs to statistically algorithm, and accuracy is high.It is contemplated that monitor video outline analyzes the requirement to real-time, the present invention uses the Mean Shift algorithm that algorithm complex is relatively low.
Based on video compress, the present invention proposes a kind of video outline technology retained based on details, and structure is shown in Fig. 3.Its core concept is to find object-based Key detail to carry out processing and labelling, then introduces details cost and non-details cost replaces original lost motion, be updated conventional cost function, obtain the minima of new cost function through simulated annealing.In the method, in motion cost, the division of details cost and non-details cost can have the reservation of emphasis to Key detail, and the sliding window treatment technology of key frame then be can solve the mutability problem such as movement locus fracture in outline video.
Accompanying drawing explanation
Fig. 1 is telescopic video outline frame diagram
Fig. 2 is quality scalable coding framework figure
Fig. 3 is the monitor video scalable coding structure chart retained based on details.
Claims (3)
1. feature of present invention is intended to propose the video outline technology of a kind of practicality, by the analysis to complete video,
Main Moving Objects and background picture are extracted, map and rebuild, generates an outline video most maximum
Raw information may be retained.
2. to conventional video outline technology, tradition scalable video technology with based on ROI scalable regard
Frequently, on the basis of the research and analysis of coding techniques, proposing a kind of new monitor video based on video outline can
Scalable coding technique.It is a kind of video outline technology retained based on details, the crucial portion to original video
Divide and use different concentration strength with non-key part, and correspondingly update cost function in video outline
Computational methods, so would be beneficial for showing most possibly key message under identical storage cost, are expected to big
Width improves picture compression efficiency.
3. its core concept of the present invention is to find object-based Key detail to carry out processing and labelling, then draws
Enter details cost and non-details cost replace original lost motion, conventional cost function is updated,
The minima of new cost function is obtained through simulated annealing.In the method, details cost in motion cost
Division with non-details cost can have a reservation of emphasis to Key detail, and the sliding window to key frame
Mouthful treatment technology then can solve in outline video the mutability problems such as movement locus fracture.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410782676.0A CN105898313A (en) | 2014-12-15 | 2014-12-15 | Novel video synopsis-based monitoring video scalable video coding technology |
| PCT/CN2015/095913 WO2016095696A1 (en) | 2014-12-15 | 2015-11-30 | Video-outline-based method for monitoring scalable coding of video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410782676.0A CN105898313A (en) | 2014-12-15 | 2014-12-15 | Novel video synopsis-based monitoring video scalable video coding technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105898313A true CN105898313A (en) | 2016-08-24 |
Family
ID=56125883
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410782676.0A Pending CN105898313A (en) | 2014-12-15 | 2014-12-15 | Novel video synopsis-based monitoring video scalable video coding technology |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN105898313A (en) |
| WO (1) | WO2016095696A1 (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018228130A1 (en) * | 2017-06-15 | 2018-12-20 | 腾讯科技（深圳）有限公司 | Video encoding method, apparatus, device, and storage medium |
| CN109168032A (en) * | 2018-11-12 | 2019-01-08 | 广州酷狗计算机科技有限公司 | Processing method, terminal, server and the storage medium of video data |
| CN110427865A (en) * | 2019-07-29 | 2019-11-08 | 三峡大学 | High voltage prohibited area human behavior video features picture extracts and reconstructing method |
| CN110717422A (en) * | 2019-09-25 | 2020-01-21 | 北京影谱科技股份有限公司 | Method and system for identifying interactive action based on convolutional neural network |
| CN111933156A (en) * | 2020-09-25 | 2020-11-13 | 广州佰锐网络科技有限公司 | High-fidelity audio processing method and device based on multiple feature recognition |
| CN112804188A (en) * | 2020-12-08 | 2021-05-14 | 鹏城实验室 | Scalable vision computing system |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108873762A (en) * | 2017-05-11 | 2018-11-23 | 上海航天电子有限公司 | Space Experiments multi-channel video capturing test controller |
| CN110719438A (en) * | 2019-08-28 | 2020-01-21 | 北京大学 | Synchronous transmission control method for digital retina video stream and characteristic stream |
| CN116744006B (en) * | 2023-08-14 | 2023-10-27 | 光谷技术有限公司 | Video monitoring data storage method based on block chain |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103108160A (en) * | 2013-01-24 | 2013-05-15 | 中国联合网络通信集团有限公司 | Surveillance video data acquisition method, server and terminal |
| US20140267714A1 (en) * | 2013-03-12 | 2014-09-18 | Futurewei Technologies, Inc. | Context Based Video Distribution and Storage |
| CN105306945A (en) * | 2014-07-10 | 2016-02-03 | 北京创鑫汇智科技发展有限责任公司 | Scalable synopsis coding method and device for monitor video |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100594713C (en) * | 2008-04-21 | 2010-03-17 | 中国科学院计算技术研究所 | A method and system for generating video outline |
| CN102375816B (en) * | 2010-08-10 | 2016-04-20 | 中国科学院自动化研究所 | A kind of Online Video enrichment facility, system and method |
| CN102395029B (en) * | 2011-11-05 | 2013-09-18 | 江苏物联网研究发展中心 | Video encoding and decoding method and device supporting retractable video browse |
| CN102905200B (en) * | 2012-08-07 | 2015-01-14 | 上海交通大学 | Video interesting region double-stream encoding and transmitting method and system |
| CN103678299B (en) * | 2012-08-30 | 2018-03-23 | 中兴通讯股份有限公司 | A kind of method and device of monitor video summary |
| CN103596011B (en) * | 2013-11-20 | 2017-07-04 | 北京中星微电子有限公司 | The storage processing method and device of view data |
-
2014
- 2014-12-15 CN CN201410782676.0A patent/CN105898313A/en active Pending
-
2015
- 2015-11-30 WO PCT/CN2015/095913 patent/WO2016095696A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103108160A (en) * | 2013-01-24 | 2013-05-15 | 中国联合网络通信集团有限公司 | Surveillance video data acquisition method, server and terminal |
| US20140267714A1 (en) * | 2013-03-12 | 2014-09-18 | Futurewei Technologies, Inc. | Context Based Video Distribution and Storage |
| CN105306945A (en) * | 2014-07-10 | 2016-02-03 | 北京创鑫汇智科技发展有限责任公司 | Scalable synopsis coding method and device for monitor video |
Non-Patent Citations (3)
| Title |
|---|
| 朱志辉: "基于视频摘要生成技术的研究", 《微电子学与计算机》 * |
| 王婷婷: "H.264运动估计算法的研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 黄志伟,陈元枝,王师峥,蔡续: "一种支持监控视频可伸缩快速浏览的区域信息编码方法", 《小型微型计算机系统》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018228130A1 (en) * | 2017-06-15 | 2018-12-20 | 腾讯科技（深圳）有限公司 | Video encoding method, apparatus, device, and storage medium |
| US10893275B2 (en) | 2017-06-15 | 2021-01-12 | Tencent Technology (Shenzhen) Company Ltd | Video coding method, device, device and storage medium |
| US11297328B2 (en) | 2017-06-15 | 2022-04-05 | Tencent Technology (Shenzhen) Company Ltd | Video coding method, device, device and storage medium |
| CN109168032A (en) * | 2018-11-12 | 2019-01-08 | 广州酷狗计算机科技有限公司 | Processing method, terminal, server and the storage medium of video data |
| CN110427865A (en) * | 2019-07-29 | 2019-11-08 | 三峡大学 | High voltage prohibited area human behavior video features picture extracts and reconstructing method |
| CN110427865B (en) * | 2019-07-29 | 2023-08-25 | 三峡大学 | Human behavior video feature picture extraction and reconstruction method for high-voltage forbidden region |
| CN110717422A (en) * | 2019-09-25 | 2020-01-21 | 北京影谱科技股份有限公司 | Method and system for identifying interactive action based on convolutional neural network |
| CN111933156A (en) * | 2020-09-25 | 2020-11-13 | 广州佰锐网络科技有限公司 | High-fidelity audio processing method and device based on multiple feature recognition |
| CN112804188A (en) * | 2020-12-08 | 2021-05-14 | 鹏城实验室 | Scalable vision computing system |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2016095696A1 (en) | 2016-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105898313A (en) | Novel video synopsis-based monitoring video scalable video coding technology | |
| Mahjourian et al. | Occupancy flow fields for motion forecasting in autonomous driving | |
| Yao et al. | Unsupervised traffic accident detection in first-person videos | |
| EP3291558B1 (en) | Video coding and decoding methods and apparatus | |
| CN109005409B (en) | Intelligent video coding method based on target detection and tracking | |
| US10567771B2 (en) | System and method for compressing video data | |
| CN109697424A (en) | A kind of high-speed railway impurity intrusion detection device and method based on FPGA and deep learning | |
| CN102395029B (en) | Video encoding and decoding method and device supporting retractable video browse | |
| CN101266132B (en) | Running disorder detection method based on MPFG movement vector | |
| CN105306945B (en) | A kind of scalable concentration coding method of monitor video and device | |
| CN103402087A (en) | Video encoding and decoding method based on gradable bit streams | |
| CN103235944A (en) | Crowd flow division and crowd flow abnormal behavior identification method | |
| CN114363623A (en) | Image processing method, image processing apparatus, image processing medium, and electronic device | |
| CN110688905A (en) | Three-dimensional object detection and tracking method based on key frame | |
| CN104301735B (en) | The overall situation coding method of urban transportation monitor video and system | |
| CN101493944A (en) | Moving target detecting and tracking method and system | |
| CN110598540B (en) | Method and system for extracting gait contour map in monitoring video | |
| CN108898842A (en) | A kind of high efficiency encoding method and its system of multi-source monitor video | |
| CN103826125A (en) | Concentrated analysis method of compressed surveillance video and device | |
| Hu et al. | Learning a spatio-temporal embedding for video instance segmentation | |
| Chen et al. | Womd-lidar: Raw sensor dataset benchmark for motion forecasting | |
| Czech et al. | On-board pedestrian trajectory prediction using behavioral features | |
| CN105163093B (en) | A kind of compressed domain video abstract extraction method and device towards AVS monitoring shelves | |
| CN108965873A (en) | A kind of adaptive division methods of pulse array coding | |
| Šarić et al. | Dense semantic forecasting in video by joint regression of features and feature motion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160824 |
|
| RJ01 | Rejection of invention patent application after publication |