采用方向树结构矢量组合小波图像分类矢量量化.pdfVIP

第27 卷第 12 期 电 子 与 信 息 学 报 Vol.27No.12 2005 年 12 月 Journal of Electronics Information Technology Dec. 2005 采用方向树结构矢量组合的小波图像分类矢量量化 郑 勇 李德明 朱维乐* (世宏科技(苏州)有限公司 苏州 215021) *( 电子科技大学电子工程学院 成都 610054 ) 摘 要: 该文提出了采用方向树结构矢量组合对小波图像进行分类矢量量化的新方法。该方法的矢量构成结合了 子带系数的方向性，充分利用了子带系数的带间和带内的相关性，按能量和活跃度进行两级分类，降低了类中矢量 的内部离散度，分类信息占用比特数少，并采用了基于人眼视觉特性的加权均方误差准则进行矢量量化，提高了量 化增益。仿真结果表明，该方法实现简单，可以达到很好的压缩效果。 关键词: 小波变换, 方向树结构矢量, 矢量分类, 矢量量化 中图分类号：TN919.8 文献标识码：A 文章编号：1009-5896(2005)12-1959-05 Classified Vector Quantiztion on Wavelet Image with Directional Tree Structure Vector Combination * Zheng Yong Li De-Ming Zhu Wei-le ( Innosis Technology , Suzhou 215021, China) *(Dept. of Electronic Eng., UEST of China, Chengdu 610054, China) Abstract A new method to perform classified vector quantization on wavelet images using directional tree structure vector combination is proposed. It makes vector combination by employing the direction character of subband coefficients and the correlation between and within subbands. It has two steps of classification with the vector’s energy and activity and thus reduces the inner dispersion of the classified vectors. The classified information needs few bits to carry. The weighted mean square error rule using human visual characteristics is used to improve quantization gain. Simulation results show its coding efficiency for wavelet images. Key words Wavelet transform, Directional tree structure, Vector classification, Vector quantization 1