Network Video Surveillance System Based On H.264

With the development of the society, people are paying more and more attention to the safety issue, thus network video surveillance system are becoming more and more widely used; at the same time, video compression technology,especially H.264,provides a good surrort for network application of video; in the application platform,DSP video development platform,especially TI’s DM6437 provides a nice environment for the development,debug and application of video codec.This paper designs a network video surveillance system based on H.264,which developed under the platform of DM6437 EVM. The paper’s thesis is as follows.1)Firstly, the paper studies the development of video compression technology and several important video compression standards in depth, including H.120,H.261,MPEG-1,H.262/MPEG-2 H.263 MPEG-4 H.264,H.265 and etc.2)Secondly, the paper studies the design princeple of DM6437 EVM, including memory map,EMIF interface, video interface, ethernet interface as well as video preprocessing module TVP5146’s structure and function.Besides, a overall video procession flow is also designed.3)Thirdly,the paper studies H.264,including basic codec flow, structural frame,profile,level, supported video format, stream format, frame structure and some key parts for optimization, including Intra Predictive Coding, P slice macro blocks’ s prediction,transform and quantization. These theory can be applied to the transport of H.264 stream and optimization of coding. Based on these theory, the paper designs software part of H.264 network video surveillance system.4)On the base of X264 Open source encoder, this paper transplant it for DM6437 to realize coding function. Coding task associated with video acquisition task and network transmission task to complete the function of the encoding end. Queue and semaphore is used for communication between tasks. To compare result of coding, the paper chose delay, compression ratio and frame rate as index and design two methods to compute them, respectively.5)Finally, optimization methods of project level and instruction level is designed to optimizate X264 coder and the result after optimization is better than before.