matlab有选择保边缘平滑法的掩模,边界保持平滑滤波方法研究 灰度方差 k近邻平滑.docx...

北京..大学毕业设计(论文) 38/dx/public/application_view.jsp?no像降噪中的边界保持平滑滤波方法研究 姓名：学号：指导教师：摘要当今社会是信息数字化时代,无论是学习,生活都与信息数字紧密相关联,其中数字图像处理在其中占有着举足轻重的地位。21世纪,数字图像处理技术高速发展,并广泛应用于识别领域,医学领域,体育领域等。平滑滤波是图像处理学的基础.最基础的滤波方式是均值滤波和中值滤波,这两种滤波方式对噪声都有抑制作用而且算法简单,但是导致图像变模糊尤其是边缘变模糊是无可避免的。虽然将滤波器加权后,效果有所改善,但理论是近似的,所以效果仍不明显。为了改善边缘的模糊,我们发现只要处理好灰度变化显著的边缘,图像就会达到一个很好的效果,要最大程度保持图片的清晰,希望在进行平滑处理的同时,检测出景物的边界,然后对噪声进行处理。本文用matlab编辑算法实现中值滤波,均值滤波,最小方差滤波,k近邻滤波对噪声的处理,并进行性能的分析和比较。关键词:matlab;中值滤波;均值滤波;最小方差滤波;k近邻滤波Abstact Modern society is digital , whether learning or life is closely associated with digital information, including digital image processing in which occupies a pivotal position .21 century, the rapid development of digital image processing technology widely used in the field of identification, medicalfication, the field of sports. The smoothing filter is the basis of image processing. The most basic filter is the mean filter and median filter, both filtering noise inhibited and the algorithm is simple, but lead to the edge of the image blurred without avoidable. Although the effect is improving ,the theory is approximate, so the effect is still not clear. In order to improve the edge blur, we found that handling the gray-scale variation significant edge, the image will reach a good effect, To the maximum extent to maintain the clarity of the picture, while performing smoothing processing, a scene boundary is detected, then, the noise is processing. In this paper, using matlab algorithm for editing median filtering, mean filtering, minimum variance filtering, k-nearest neighbor filtering noise processing and analysis and comparison of performance.Keywords: matlab; median filter; mean filter; minimum variance filtering; k nearest neighbor filter目录摘要IAbstactII第1章:绪论11.1 课题背景11.2研究目标11.3研究内容11.4论文的组织安排2第2章 平滑滤波的相关知识32.1噪声的相关知识32.1.1 噪声的定义32.1.2在matlab中添加噪声42.2彩色图像的分解52.3平滑的概念62.3.1空间域方法72.3.2频率域方法72.3.3平滑算法72.4峰值信噪比的概念8第3章均值滤波103.1均值滤波的概念及方法103.2均值滤波的效果比较113.3均值滤波的评价13第4章中值滤波144.1中值滤波的概念和方法144.2