毕业设计（论文）-基于MATLAB的图像消噪处理.doc

毕业设计
基于MATLAB的图像消噪处理
学院名称: 新华学院
专业名称: 电子信息工程
年级: 2006级
学生学号:
学生姓名:
指导教师:
申请日期: 2010-5-10
摘要
随着计算机技术的飞速发展,数字图像处理技术也获得了飞速的发展,并形成了许多分支,其中图像消噪就是图像处理中的一个重要分支。高斯噪声和脉冲噪声在图像形成和传输过程中最为常见,且容易导致图像质量下降,为减小噪声影响,各种线性、非线性滤波方法被广泛使用。本论文在Matlab软件平台下,利用数字滤波器对含高斯噪声和脉冲噪声的图像进行滤波,通过峰值信噪比和主观感受来评价滤波效果。
本论文首先概要的陈述了数字图像的基本理论,包括图像的分类,类型,常见图像的格式,并阐述了图像滤波的意义,图像滤波的方法及图像质量评价标准。
其次,介绍了数字滤波器的基础知识。数字滤波器可以分为两大类,一种是线性滤波器,一种是非线性滤波器。线性滤波器从冲击响应的时域特征可以分为FIR滤波器和IIR滤波器。非线性滤波器主要介绍了中值滤波器,给出了中值滤波的原理和主要特性。
最后,选取标准的Lena图像,对其添加不同强度的高斯噪声和脉冲噪声,分别通过线性的FIR滤波器和非线性的中值滤波器对其进行噪声滤除。文中给出了在Matlab平台下的仿真结果和实验数据。试验结果表明:线性FIR滤波器和非线性中值滤波器对数字图像噪声消除均有一定的效果,能提高滤波图像的峰值信噪比。非线性中值滤波器在滤除脉冲噪声方面,表现尤为突出,随着脉冲噪声强度的增加,PSNR提高的幅度也有所增加。
关键字:图像处理;消噪;FIR滤波器;中值滤波器;峰值信噪比
Abstract
Along with the rapid development puter technology, the digital image processing technology also wins the rapid development, and formed many branches, image denoising is an important branch in image processing. Gaussian noise and impulse noise is mon in image formation and transport process, which easy to cause the image quality, in order to reduce the noise influence, all sorts of linear and nonlinear filtering method is widely used. In this paper, we used the digital filter to wipe out Gaussian noise and impulse noise in the image on the Matlab, through the peak signal to noise ratio and subjective feeling to evaluate the effect of filter.
This paper first briefly stated the basic theory of digital images, including image classification, type, common image formats, and illustrated the significance of image filtering, image filtering method and image quality assessment standards.
Secondly, the paper introduced the basic knowledge of digital filter. Digital filter can be divided into two kinds, one kind is linear filter, and another is nonlinear filters. Linear filter can be divided into FIR filter and IIR filter from time-domain features of shock response. The nonlinear filters mainly introduced the median filter, median filtering principle and main cha