Design And Implementation Of An AVS-Based Surveillance Video Content Storage And Quick Retrieval System

As we know, the security problems get more concerned these years. More and more video surveillance systems are used in security field. For example, access control system, check on work attendance system, and ID recognition system, etc.Video compression is the kernel technology of a video surveillance system. Video surveillance systems should choose a suitable video compression in dependence on theirs demands. At present, no video compress standard has been chosen to be a specialized standard in video surveillance systems. AVS is a fully intellectual property of second generation Source Coding and Decoding Standard owned by China. And AVS is a set of standard system which includes system, video, audio, and media copyright management. It will be potentially used in digital video and audio industry of China.The traditional video surveillance system costs a lot of human resource, because the detection, recognition and comprehension of surveillance video content absolutely depend on humans. This approach reduces the working efficiency of video surveillance system, as well as lacks of security and veracity. So, the research on automatic video surveillance system is quite necessary.Firstly, the thesis surveys the video surveillance systems which are currently used, research and analyzes their characteristics and shortcomings, and acquaints ourselves with their major technologies. Secondly, we are completely familiar with AVS coding standard, and then compare it with MPEG4/H.264 coding standard, specify the advantages of AVS using in video surveillance system in many aspects. Finally, the thesis design and implement an integrated AVS-based video surveillance system. In the AVS-based video surveillance system, AVS video streams are used as input, and the AVS coding and decoding standards are combined with other processing techniques, such as background/non-background Classification, face detection, and recognition and two-level indexing structure to automatically identify the people who appeared in the surveillance videos and quick retrieval, and can be used in many areas like digital surveillance, access control, etc.