基于小波变换的图像降噪技术

龙源期刊网 http://www.qikan.com.cn

基于小波变换的图像降噪技术

作者：孙东宁,张丽,闫朝文

来源：《电脑知识与技术》2010年第03期

摘要:小波变换在图像处理中有着重要的应用,在基于小波的图像降噪处理算法中,常常存在着对图像信息的过分滤除和对噪声信息的欠滤出,而使得对图像降噪后不仅没有提高信噪比反而使其降低了,这就不利于对图像进行分析观察。究其原因主要是由于阈值选取和处理方法不恰当引起的。在小波空间Donoho阈值算法的基础上,结合Birge-Massart策略得出的多层阈值图像降噪处理算法,从而达到较好的保留图像的细节有用信息、降低噪声的目的,仿真实验表明对一般受低噪声干扰的图像做降噪处理时,效果较好。 关键词:小波变换;图像降噪;阈值

中图分类号:TP18 文献标识码:A文章编号:1009-3044(2010)03-691-02 Research of Image Noise Reduction Based on Wavelet Transform SUN Dong-ning1, ZHANG Li2, YAN Chao-wen1

(1.Shool of Information and Engineering Shanghai Maritime University, Shanghai 200135, China; 2.Institute of Marine Material Science & Engineering, Shanghai Maritime University, Shanghai 200135, China)

Abstract: Wavelet transform in image processing has important applications, in the image

denoising which based on wavelet processing algorithms, it is often to over-filtered image information and the noise doesn't filter out enough, causing not only failed to post the image noise Instead, but also lowered to improve signal to noise ratio, which is not conducive to observation and analysis of the image.This mainly due to threshold selection and processing methods are not appropriate. In the wavelet space Donoho threshold algorithm based on the Birge-Massart strategy with resulting noise reduction multi-threshold image processing algorithms to achieve the better to retain the image details of useful information and reduce noise, simulation results show that the effect is good when image is noised by the low-noise.

Key words: wavelet transformation; image noise reduction; threshold

人们一直致力于图像去噪这一研究热点,其目的就是最大限度的保持图像的原有信息,降低噪声对图像的影响,提高信噪比。小波分析是一种良好的时频域分析方法,具有多分辨率分析的特性,在时域和频域都能很好的表征信号的局部信息,这使得小波变换在图像信息处理起到了很大的作用。小波阈值是一种非线性的处理方法,它是在小波域内通过对小波系数进行压缩处理

本文来源：https://www.bwwdw.com/article/hkma.html