兰州大学机构库 >数学与统计学院
图像复原的滤波器方法研究
Alternative TitleFiltering Methods for Image Restoration
杜斌勇
Subtype学士
Thesis Advisor黄玉梅
2010-05-22
Degree Grantor兰州大学
Place of Conferral兰州
Degree Name学士
Keyword图像复原 滤波器 峰值信噪比 均方误差
Abstract   在数字图像的形成过程中,记录到的图像经常要受到外界或图像形成系统中噪声及模糊的污染,我们得到的图像往往不是原始清晰图像,而是受到污染的失真图像。图像恢复的任务就是寻找有效方法,从失真图像得到和原始图像尽量接近的清晰图像。图像复原的滤波器方法是一种非常重要的图像恢复方法。本文我们研究了图像复原的滤波器方法。首先阐述了几种基本的滤波器方法,他们分别是逆滤波法,维纳滤波,均值滤波。在此基础上,我们提出了改进的领域差值方法,构建了改进的滤波器。最后,在MATLAB软件环境之下,实现了这些方法,完成了图像复原的数值实验,并采用均方误差( MSE) ,峰值信噪比( PSNR) 作为衡量标准,通过对图像恢复的结果进行比较,发现我们提出的改进滤波器能较好的去除不同类型的噪声。
Other AbstractDue to the external or internal influence of the image recording systems, images are often contaminated by noise and blur during their formation process. The task of image restoration is to construct effective methods to get an estimated image which is as close as possible to the original image. Filtering method are important methods for image restoration. In this paper, we study the filtering method for restoring images. Firstly, we elaborated several basic filtering method, they are the inverse filtering method, Wiener filtering method and mean filtering method. Then we construct the domain difference based filtering method, which is more efficient for image restoration. Finally, the numerical experiments are carried out for these methods in MATLAB. The mean square error (MSE), peak signal to noise ratio (PSNR) and visual image restoration are compared for different methods, and we found that the proposed filtering is more general to remove different noises.
URL查看原文
Language中文
Document Type学位论文
Identifierhttps://ir.lzu.edu.cn/handle/262010/224620
Collection数学与统计学院
Recommended Citation
GB/T 7714
杜斌勇. 图像复原的滤波器方法研究[D]. 兰州. 兰州大学,2010.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Altmetrics Score
Google Scholar
Similar articles in Google Scholar
[杜斌勇]'s Articles
Baidu academic
Similar articles in Baidu academic
[杜斌勇]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[杜斌勇]'s Articles
Terms of Use
No data!
Social Bookmark/Share
No comment.
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.