| 一种基于小波变换的矢量量化算法 |
| |
| 引用本文： | 陈善学.一种基于小波变换的矢量量化算法[J].重庆邮电学院学报(自然科学版),2002,14(1):87-90. |
| |
| 作者姓名： | 陈善学 |
| |
| 作者单位： | 重庆邮电学院 重庆400065 |
| |
| 摘 要： | 提出了一种基于波变换的矢量量化逄法，算法利用小波变换后各子带间的相关性，合理地构造矢量。采用最优矢量量化器设计原则，给出了小波域的误差竞争学习算法（DCL）。实验表明，这种算法获得的码本优于其它几种算法。
|
| 关 键 词： | 小波变换 矢量量化 神经网络 数据压缩 误差竞争学习算法 |
A Vector Quantization Algorithm Based on Wavelet Transforms |
| |
| CHEN Shan,xue.A Vector Quantization Algorithm Based on Wavelet Transforms[J].Journal of Chongqing University of Posts and Telecommunications(Natural Sciences Edition),2002,14(1):87-90. |
| |
| Authors: | CHEN Shan xue |
| |
| Abstract: | A vector quantization algorithm based on wavelet transforms is introduced. The correlation of the wavelet coefficients in different frequency bands is effectively exploited through constructing the crossband vector. By adopting the optimal design principle of the vector quintile, the author in this paper presents a distortion competitive learning (DCL) algorithm. The experiment results show that this algorithm is superior to other methods. |
| |
| Keywords: | wavelet transform vector quantization neural network |
| 本文献已被 CNKI 维普 等数据库收录！ |
