Study Of Decoding Algorithms Based On Distributed Video Coding

With the high development of wireless network, more and more mobile video terminals are used in multimedia communications. Since these wireless devices are always limited in computation capability and energy resources, the encoder should be low-complexity and error resilient with high compression efficiency. The traditional video compression standards such as MPEG and H.26x could not used in the above-mentioned application environment because of the complicated encoder. Therefore, the scholars around the world proposed a novel video compression method, distributed video coding technology, especially Wyner-Ziv video coding has recently become a research focus of distributed video coding.The influence of the key technologies such as correlation noise model and reconstruction on Wyner-Ziv video coding efficiency is analyzed in the thesis. At first, an adaptive correlation noise model algorithm is proposed, which adjusts the distribution of the correlation noises based on different bands and the degree of the video motion. It makes a more precise model. Then, the robust reconstruction algorithm of the Wyner-Ziv video is proposed which takes importance on the upper bands and the wrong quantization. The proposed method can greatly improve the quality of reconstructed frame. Finally, a practical Wyner-Ziv video coding system without feedback based on transform domain is constructed using the above-mentioned algorithms.Experimental results show that compared with H.263+ intra-frame coding ,H.264/AVC intra-frame coding method and traditional Wyner-Ziv video coding, the proposed Wyner-Ziv video coding efficiency is significantly improved and the reconstruction video quality is obviously enhanced, besides, the proposed system whose encoder is simple and easy to realize, can be applied to the video communication...