Research And Implementation Error Elimination Mechanism Based Distributed Video Coding

With the rapid development of information age in human society, the application of wireless multi-media communication becomes a certain trend. But the large amount of video data is difficult to transmit and store, so the traditional video coding standards apply DCT, motion estimation and compensation and predictive coding to compression coding. In this case, the video data obtain the high compression ratio, but also reduce the sensitivity of mistakes. The exiting wireless network or the Internet network can not provide reliable transmission, the video data are prone to emerge error or lost package in the process of transmission. The mistakes affect the decoding effect of the current frame, even appear error drift. In this case, how to eliminate error drift becomes a new research topic.The mainly research work of this paper includes the following respects:1. On the basis of research the existing Wyner-ziv video coding structures, the thesis analyzes its advantages and problems, which include how to tectonic accurate side information, balancing complexity and compression efficiency at the encoder, probability statistics model between current decoded frame and side information and how to generate accurate side information based on motion estimation and reconstruct frames jointly.2. The thesis proposes the interframe correlation judgment algorithm in the view of temporal correlation. The thesis illustrates the coding process of error elimination mechanism based Wyner-Ziv coding. In the view of lack of flexibility about choice of Peg frame. The thesis proposes and implements the codec based temporal correlation Wyner-Ziv coding. Though this structure, the choice of Peg fame is relevant to the test sequence. Experimental results show that the proposed coding system reduces the transmission bits, which does not reduce the decoding quality. For sports severe test sequence, it improves the decoding quality effectively. In summary, the PSNR of test sequence is higher 0.2dB than the traditional method. And the subjective effects have been improved obviously.3. This thesis implements DCT domain-based Wyner-Ziv video coding based the second part of theoretical basis. And then, the performances of this Wyner-Ziv video coding system is made and analyzes in details.