基于小波的去噪处理.doc

摘 要 随着数码照相机等各类数码产品的普及，图像在采集、转换和传输过程中往往受到成像设备和外界环境噪声干扰等影响产生噪声。[]然而噪声的存在，却严重干扰了图像本来的真正面目。如何消除图像中的噪声，长期以来都是图像处理中古老的课题。 人们会根据图像的特点、噪声的统计特征和频谱分布的规律, 提出和发展了不同的去噪方法。由于图像在去噪处理中存在一个矛盾问题是如何在降低图像噪声和保留图像细节上保持平衡。传统的滤波器对有效地将信号高频和由噪声引起的高频干扰加以区分，将图像的高频滤除，虽然能够达到去噪的处理，但是却破坏了图像细节。 小波变换是一种信号的时频分析方法，有利于各分辨度不同特征的提取，并且比传统的Fourier变换还要快一个数量级。它是在傅立叶变换的基础上发展起来的的一种时频分析方法，是一种有效地分析工具，它具有低熵性、多分辨率、去相关、选基灵活等特性和在信号方面具有的优势，因此具有“数据显微镜”之称的小波变换在图像处理领域受到许多学者的重视。其中，利用小波变换对含噪图像信号进行处理，可以在保留高频信息的同时能够有效地滤除噪声，得到对原始图像信号都得最佳恢复。 本文通过对灰度图像和彩色图像进行去噪处理，分别采用软阈值方法Donoho全局阈值、硬阈值方法Donoho全局阈值、软阈值方法Birge-Massart策略计算阈值、硬阈值方法Birge-Massart策略计算阈值、均值滤波、半软阈值方法、半软阈值与均值滤波相结合的方法分别去除噪声，通过仿真实验发现，本论文提出的半软阈值与均值滤波相结合的方法信噪比最大，去噪效果最明显。 关键词：图像去噪 ； 消除 ； 高频 ；小波变换 ABSTRACT With the popularity of digital cameras and other kinds of digital productsthe image is often in the process of acquisition, conversion and transmission imaging equipment and impact noise such as external environment noise. While the presence of noise, image what serious interference. How to eliminate noise in image is the old project in image processing. For a long time, people according to the characteristics of the images, the statistical characteristic of noise and spectrum distribution of the law, put forward and developed the different denoising method . Due to a conflict problems in image denoising processing is how to reduce image noise and preserving image details on balance. The traditional filter effectively will signal of high frequency and high frequency caused by noise interfere with the distinction, the image of the high frequency filter, while capable of achieving the denoising processing, but it is ruining the image detail. Wavelet transform is a method of signal time-frequency analysis is helpful to distinguish between degrees of different feature extraction, and an order of magnitude faster than the traditional Fourier transform. So with microscope "data," said the wavelet trans