Research On Error Concealment In Video Decoder Of H.264

With the development of information technology and the advancement, the multimedia application is getting more and more widespread. It is hoped that we can communicate conveniently, fast and flexibly through media such as voice, data, images and video, etc. In multimedia communications, the video transmission has received more widespread attention since images can bring more information. However, it is impossible to avoid transmission error as interruption and delay always happen in channel. Furthermore, the use of video compression technology in order to reduce the transmission of date makes it more sensitive to video transmission of error, it is necessary to adopt effective measures to enhance the Robust of video transmission.H.264 video encoding standard makes use of motion estimation / motion compensation to eliminate time redundancy of video, discrete cosine transformation to the margin image to eliminate spatial redundancy and Variable Length Encoding to encode quantification coefficient to elimination statistical redundancy, and obtain very compression efficiency. The compressed bitstream is more sensitive to the mistake, although the application of a variety of error recovery tools (such as flexible macroblock order, parameter sets, data partition, and redundancy slice, etc.) enhances the robustness of video stream, the transmission error will occur, especially in the wireless communications field where the application is more and more extensive. After a detailed introduction of video coding standard H.264, this paper mainly focuses on the realization of post-processing techniques as error concealment in the decoder-video. The major works finished is as follows:Base on a detailed analysis of error concealment algorithm used by JM (reference software of H.264) in I frames and P frames, an algorithm aimed at improving the perceptual quality of reconstructed video is designed and this paper has given a detailed introduction of both theoretical basis and principle. Then the algorithm is implemented on reference software JM version 8.6 and verified by using a standard test sequence. According to the tests, the scheme proposed by this paper can obtain better reconstructed video quality.