Video Scalable Coding And Error Concealment For MPEG-4 Codec System

While developing of video compression and encoding technology, a series of video compression international standards come on one after the other, and the video multimedia data are widely used in many fields. Especially after MPEG-4 international standard is issued, its features of efficiency, scalability, interaction and robustness extend the application area of digital video from image storage and digital TV to networks and wireless channels. The developing of new transmission techniques speed the innovation of video encoding techniques, scalable encoding and error concealment become more important. Wavelet transformation has inherent features of layered encoding and robust encoding, it will be a important encoding method in network video. In this dissertation, we are mainly focusing on the following three fields: wavelet transformation, scalable encoding, and error Concealment.The performance of wavelet transformation is better than DCT transformation in static 2D picture, so wavelet transformation is adopt in JPEG2000 and MPEG-4 international standards. There are spatial and temporal scalability simultaneously in three-dimensional wavelet transformation for sequence pictures, it is one of best video encoding methods in network video. The big memory requirement is the important obstacle in 3D wavelet transformation, the algorithm of phased temporal wavelet transformation in this dissertation can reduce the system memory requirement observably. The efficient organizational format of scalable 3D wavelet coefficients improves the scalability of video data stream for transcoding and decoding. Without exploiting motion information, normal 3D wavelet transformation can decline its coding efficiency in complex and movable scene.Video scalable encoding can be divided into temporal, spatial, quality, and content scalability. Fine granularity scalable encoding can be used in DCT-based progressive scalable encoding, it is also under development. The motion-based layered compression scheme in this dissertation is different from the traditional full picture-process method, using different encoding strategy by classify marcoblock in temporal motion, a high encoding efficiency is achieved in experiment.The congestion, packet lose and transmission error is an inherent problem with internet and wireless mobile video communication system, the techniques of error control and concealment is necessary to this video encoding and decoding applications. The postprocessing techniques can be applied in any circumstances, it don't depand on overhead of encoding information, so the postprocessing error concealment is diffcult and important to any applications. A inter-frame error concealment algorithm based on motion estimation is proposed in this dissertation, the motion vectors and coding mode can be rebuilt, it can be applied in any frame modes and macroblock modes. In intra-frame postprocessing error concealment techniques in this dissertation, interpolating pixel values within a damaged macroblock from its four 1-pixel-wide boundaries and introducing the texture information from its four adjacent blocks can improve the quality of reconstructed image. It was found that the visual texture in damaged macroblocks can be reconverted with his algorithm in experiment. When several error concealment techniques are adopted, the quality of reconstructed image can be improved greatly.