Research On Error Concealment Algorithm Based On Motion Vector Reconstruction For H.264

With the fast development of multimedia communications and network technologies, information is transforming gradually from the traditional text, voice and image to digitized video. For the high compressed video bit-stream easily damaged due to the variety of channel noises and fluctuations which will lead to the quality decline of the reconstructed video sequence. So the present video coding standards have employed particular technologies to ensure the quality of reconstructed video sequence. The error concealment(EC) technology is widely used to recover the lost information of the transmitted video signal on terminals through received correct information. The error concealme nt technology is very effective since it performed at the decoder side with no calculation pressure to the encoder and will not bring additional transmission load to the channels. It has become the hotspot in the research fields of video coding technologies.The thesis conducts research on the H.264 which is the well-known and the most widely used video coding standard. Based on the study of the state of art error concealment algorithms, the thesis presents a novel error concealment algorithm for video communication systems with H.264/AVC standard which exploits both the spatial and temporal information in the reference and current frames. The main works of thesis are listed below:1. The thesis briefly discusses the present situation and development of video coding standards. The coding framework and core technologies of H.264 have been introduced in detail. In addition, the causes and influence of error have been analyzed. The detection, control and recover of error have also been introduced systematically.2. Several state of art error concealment algorithms have been studied. These algorithms are divided into three categories. The thesis describes the theoretical basis, application scena rios, merits and faults of algorithms in each category.3. In this thesis, a novel error concealment algorithm which exploits both the spatial and temporal information in the reference and current frames is proposed. The motion vectors of the lost sub-blocks are recovered by those of the neighboring sub-blocks using K-Means mechanism, motion vectors extrapolation and tendency factor. According to the experimental results, the proposed method can recover the distorted video signal more efficiently compared with the existing methods in different QP and packet loss rates. The proposed algorithm can improve the quality of the decoded video sequence and can be applied to practical use.