WT域图象分类的矢量量化编码

基于边缘方向分类的CVQ能够减少在较低位码率进行图象信号矢量量化时的边缘失真和计算的复杂程 度.根据静止图象小波变换后小波系数的特征提出了一种在WT域进行分类的分类算法.此方法可以容易地扩展 到WT变换编码.实验结果表明,与同类方法相比,在相同或稍高信噪比的情况下,此方法的位码率大大减小.

Image Vector Quantizer based on a Classificatin in the WT Domain

A classified VQ (CVQ) tehchnique.based on the edge-oriented classifier.can reduce the edge degradation as well as the encoding complexity. In this paper according to the WT coefficients feature of a still image transformed,we propose a classification algorithm in the wavelet transform(WT)domain for the CVQ. This approach can be easily extended to the wavelet transform coding technique. Coded images based on this classification are shown to enjoy considerable bpp reduction over three existing methods in the same or a little higher PSNR conditions.