一种新型的无损视频压缩算法  

Lossless Video Compression System Designation

作　　者：夏杰[1] 侯朝焕[1] 

机构地区：[1]中国科学院声学研究所,北京100080

出　　处：《电子与信息学报》2006年第3期385-389,共5页Journal of Electronics & Information Technology

基　　金：国家973计划(G1999032900)国家自然科学基金(60028102)资助课题

摘　　要：多次使用有损压缩技术压缩数字视频,会导致视频的质量大幅下降,从而降低了数字视频的可再利用率, 为此设计了一种新型的无损视频压缩算法。该算法的特点在于:(1)提出了一种改进的基于上下文树的算术编码来压缩运动补偿后的误差帧。(2)针对改进的算术编码,优化基于宏块的运动估计与补偿算法,以提高无损视频压缩算法的压缩率。对于压缩运动补偿后误差帧的算法,与静态图像无损压缩算法JPEG-LS和CALIC相比表明,该文设计的无损视频压缩算法的压缩率超过JPEG-LS算法最高为23．3％,超过CALIC算法最高为19．3％。It will create the problem of great degradation in video quality, when the digital video is compressed by using lossy video compression algorithm more than once. That will reduce the rate of reusing the digital video. Thus a new lossless video compression algorithm has been proposed. This algorithm features in： （1） an improved context-tree based arithmetic coding is proposed to compress the residual frame after motion compensation, and（2） macro-block based motion estimation and compensation method is optimized to improve the compression ratio of lossless video compression algorithm, with consideration to the improved arithmetic coding. When using JPEG-LS and CALIC, the existing state-of-the-art in lossless compression of still images, to encode those motion-compensated error-frames or residue frames to formulate benchmarks, the experiments illustrate that the proposed algorithm outperforms JPEG-LS by up to 23.3%, and CALIC by up to 19.3%.

关 键 词：无损视频压缩 算术编码 运动估计与补偿 

分 类 号：TN919.8[电子电信—通信与信息系统]