基于小波变换的图像信号分解研究

本文通过对小波变换多分辨分析特性的介绍，讨论了基于小波基分解的滤波器组设计原理，给出了静止图像数据的树形分解的方法和分解结果，结果表明，树形分解后，可有效压缩数据动态范围，便于采用适当的量化方式，结合其它非线性编码，达到较高的压缩比。

Study of Image Decomposition Based on Wavelet Transform

By introducing the multiresolution characteristics of wavelet transform, the requirements for designing filter banks, which are based on wavelet decomposition, have been discussed in the paper. This kind of filter bands have been applied to decompose several still pictures into a kind of tree structure. The result of the decomposition demonstrates that the dynamic range of the raw data has been decreased greatly, which indicates that once a certain quantilization method along with nonlinear code is adopted, a great compression rate can be obtained.