误码信道下基于率失真优化的视频压缩编码参数选择研究

项目名称： 误码信道下基于率失真优化的视频压缩编码参数选择研究

项目编号： No.60802027

项目类型： 青年科学基金项目

立项/批准年度： 2009

项目学科： 金属学与金属工艺

项目作者： 杨静

作者单位： 上海海事大学

项目金额： 12万元

中文摘要： 本项目研究围绕信源信道的一体化联合编码展开，研究核心主要集中于误码信道下视频压缩编码参数的率失真优化选择。研究成果主要包括：提出了新颖的端到端预测误差估计模型，算法有利于率失真优化运动矢量选择的快速实现，并且与标准编解码器完全兼容，大大提高了实用性；对宏块编码模式选择代价公式和运动矢量选择代价公式进行了联合简化，进一步降低了算法的复杂度；设计完成了基于误码信道的率失真优化选择宏块帧内/帧间编码模式，块大小，参考帧，运动矢量等编码参数的完整算法；通过大量实验研究总结了不同编解码选项对率失真优化鲁棒运动补偿算法的性能影响。同时，研究过程中涉及了误码掩盖及信源编码等相关基础领域，所取得的成果对项目核心研究以有力的支持。本项目的研究成果对视频压缩编码的抗误码理论研究以及推广率失真优化编码理论在误码环境下的实际应用分别具有重要的理论和实际意义。

中文关键词： 率失真优化；运动估计；抗误码；

英文摘要： This project is concerned with the joint source-channel integration coding. The research is mainly focused on the rate-distortion optimization decision on video coding parameters under error channels. The research results include: a novel end-to-end prediction error estimation model is proposed. It is conductive for the fast realization of rate-distortion optimized motion estimation. In addition, it is fully compatible with standard coder and decoder hence greatly improves practicability. The macroblock coding mode selection cost function and motion vector selection cost function are jointly simplified and the computation complexity is further decreased. Rate-distortion optimized selection algorithms on coding parameters such as macroblock intra/inter mode, multiple block size, reference frame and motion vector are integrated. Extensive experiments have been done to investigate the performance impact of different coding or decoding options on robust rate-distortion optimization motion-compensation algorithm. Useful conclusions are drawn. Meanwhile, during the course of study of the project, some basic research areas such as error concealment and source coding were touched upon. The related results strongly support the core research of the project. The results of this project have important theoretical and practical significance on theoretical research of error resilient video compression coding and promotion of practical application of the rate-distortion optimized coding theory under error environment, respectively.

英文关键词： Rate distortion optimization; Motion-estimation;Error resilient coding;