基于小波分析的条纹图滤波方法  

A Non linear Filtering Method Based on Wavelet Analysis and its Application to Speckle Fringe Pattern

作　　者：缪泓[1] 束方军[1] 冯传玉[1] 

机构地区：[1]中国科学技术大学力学和机械工程系,合肥230027

出　　处：《实验力学》1999年第3期354-358,共5页Journal of Experimental Mechanics

摘　　要：本文提出了一种新颖的干涉条纹图非线性滤波方法，可以选择性地在频域图像的不同区域选用不同的滤波方式，在滤掉图像中大部分散斑噪声的同时，能够减少信息的损失，使图像的内部边界仍然保持清晰。文中首先简介了正交小波变换的原理，然后介绍了去除散斑噪声的具体算法。A new image processing method, which is based on the technique of wavelet analysis, is presented. When applied to the processing of speckles fringe pattern, where the speckle noise is to be removed from the fringe pattern, the method is more effective in reducing the loss of information owing to that it allows different filtering methods used for the different areas of the fringe pattern. As a result, the method can give a clearer board for the fringe pattern. In this paper, the principle of wavelet transform is briefly introduced, then a practical algorithm for removing the speckle noise from the fringe pattern is given. As the last part, a computer simulation result by using this method is shown.

关 键 词：小波分析 非线性滤波 图像处理 去噪声 

分 类 号：TN919.8[电子电信—通信与信息系统]