Research And Implementation Based On Frequency Domain Of Distributed Video Coding

With the rapid development of internet network and video technology, more and more mobilevideo devices are invented, such as tablet personal computers, mobile cameras, video cameras,wireless monitors, sensors, and so on. One thing in common is that these devices are always limitedin processing capability and resources consumption capacity, at the same time, the coding algorithmof low complexity is required. However, traditional video coding technology is unable to solvethese problems. Therefore, Distributed Video Coding which is a new video coding framework cameinto being. Distributed video coding is able to shift computational complexity from the encoder tothe decoder. It not only has the characteristic of low encoding complexity and high decodingcomplexity, but also higher compression efficiency than the traditional intra-coded, as well as betterrobustness of communication network packet loss rate.The main content of this thesis is the research and implementation of a distributed videocoding system. It includes the analysis of video coding scheme, the study of key technology andside information prediction technology. A new distributed video coding algorithm is proposed onthe basis of existing algorithms. Firstly, it introduces the basic principles of distributed video codingsystem, and has compared and analyzed on several typical coding algorithms. Secondly, the thesisanalyzes some key technology in DVC, and has made out based on some mature technology. Ratecontrol and side-information generation method are researched on emphases. A new rate controlmethod based on the low-frequency emphasis protection and improved EM side informationgeneration algorithm are put forward. Thirdly, these key technologies are adopted to a distributedvideo coding system, and bring a complete video coding system based on DCT.The experimental results shows more higher efficiency than the pixel domain and frequencydomain, and has a distinct advantage in the PSNR ,which proves the feasibility and effectiveness ofthe improvement algorithm.