一种基于类haar小波的MCTF视频编码方案

An Adaptive Haar-Like MCTF Based Wavelet Video Coding Scheme

在传统的基于运动补偿的时间滤波小波视频编码方案中，帧组尺寸和低通帧位置都是固定的，没有考虑到视频序列中运动的性质．本文提出了一种类似haar小波的基于运动补偿的时间滤波编码方案，利用互信息技术，根据视频的运动特征自适应的选择帧组尺寸和低通帧位置．同时，根据选定的帧组尺寸和低通帧位置决定时间上的编码层次．大量的实验比较了提出的方法和传统方法的压缩性能．当视频序列运动较为剧烈时，本文提出的自适应选择低通帧方法较固定低通帧方法的PSNR提高约0．3～0．5dB；当视频序列中运动缓急不一，如频繁的画面切换，本文提出的自适应的选择帧组尺寸的方法能取得更好的效果．

In conventional motion compensated temporal filtering based wavelet coding scheme, where the size of group of pieture and low-pass frames＇ positions are fixed, and the motion activities in video sequences are not considered.In this paper,we propose a haar-like motion compensated temporal filtering based wavelet video coding scheme, which can adaptively select the size of group of picture and the low-pass frames based on mutual informafon technology.Then,the temporal decomposition level is determined using the selected size of group of picture and the low-pass frames＇ position. We compaer the PSNR performance of the scheme with the conventional scheme by carrying out a large amount of experimental work. When the video sequence has high motion activity,the proposed low-pass frame selection can improve the compression quality by about 0.3～0.5 dB compared with the conventional scheme;And,the proposed adaptive approach for selecting the size of group of picture is more suitable for situations with un-even variation of motion activities,e, g. with frequent shot cuts.