一种基于小波和MP变换的细粒度视频编码算法  

Fine granular video coding algorithm based on wavelet and MP transformation

作　　者：蒲菊华[1] 熊璋[1] 方义[1] 

机构地区：[1]北京航空航天大学计算机学院,北京100083

出　　处：《计算机工程与设计》2004年第1期13-16,共4页Computer Engineering and Design

基　　金：航天基金;航空基础科学基金资助项目(00F51057)。

摘　　要：针对目前的细粒度算法的计算复杂度大(如Matching Pursuit编码,简称MP编码)或者视频质量有各种效应(如离散小波变换编码DWT)的缺点,提出了一种基于小波变换和MP变换联合的细粒度视频编码算法。对连续的8帧视频采用一维小波变换,然后对变换后第1帧低频图像用二维小波变换,其它7帧高频图像信息采用MP变换编码,并采用基于能量的原子搜索与基于人眼视觉特性的分配策略。实验表明,该细粒度算法对帧间运动较小的视频应用,有较高的恢复质量。An algorithm based on wavelet and matching pursuit is presented for fine granular video coding. It eliminates the in-terframe redundancy in a sequence of eight video frames by 3-level wavelet transformation. The transformed low-frequency frame is further processed by 2-dimensional wavelet transformation to eliminate the intraframe redundancy with the other 7 high-frequency frames coded by matching pursuit, and the MP atoms are allocated according to the energy and human eyes characteristics. This fine granular algorithm results in high recovery quality for video application with slow motion just as shown in our experiment.

关 键 词：细粒度算法 小波变换 MP变换 视频编码 原子搜索 

分 类 号：TN919.81[电子电信—通信与信息系统]