《毕业设计（论文）：基于频域的图像复原技术研究》由会员分享，可在线阅读，更多相关《毕业设计（论文）：基于频域的图像复原技术研究（66页珍藏版）》请在金锄头文库上搜索。
1、学士学位论文摘 要图像复原的定义是利用退化过程的先验知识，去恢复已被退化图像的本来面目。图像复原试图利用退化图像的某种先验知识来重建或复原被退化的图像，因此图像复原可以看成图像退化的逆过程，是将图像退化的过程加以估计，建立退化的数学模型后，补偿退化过程造成的失真，以便获得未经干扰退化的原始图像或原始图像的最优估值，从而改善图像质量。目前，图像复原技术已经广泛应用于空间探索、天文观测、物质研究、遥感遥测、军事科学、医学影像、交通监控、刑事侦察等众多领域。在长时间的应用过程中，人们研究了很多退化的模型和复原的方法。目前主流的复原方法有：维纳滤波法，正则化滤波法，Lucy-Richardson 算法
2、以及盲卷积算法。同时，因为图像的锐化处理可以突出图像的轮廓，因此也可用于某些二值图像复原。其原理是将原图像和对他实施拉式算子后的结果组合后产生一个锐化图像。然后对其在数字图像处理方面进行举例分析，并实现锐化效果。对锐化图像进行分析与讨论后，发现其在图像处理应用方面，特别是用来改善因扩散效应的模糊方面特别有效。而维纳滤波是一种常见的图像复原方法，该方法的思想是使复原的图像与原图像的均方误差最小原则采复原图像。本文中，主要通过介绍维纳滤波的基本原理，并结合 MATLAB 中的函数，设计相应的维纳滤波器，实现“含噪”图像的复原，进行了对退化图像复原的仿真实验，在退化图像中加入了噪声进行恢复，实验表明
3、退化图像在有噪声时必须考虑图像的信噪比，噪声的自相关函数进行图像恢复，才能取得较好的复原效果。 关键词:维纳滤波;MATLAB;图像恢复学士学位论文AbstractThe definition of image restoration is the using of priori knowledge about the restoration process, to restore the true colours about the degraded image. Image restoration attempts to use some prior knowledge of degrad
4、ed image reconstruction or restoration to be degraded image, so that the image restoration can be viewed as the inverse process of degraded image, estimate the process of the image restoration,to establish the mathematical model of restoration, restoration caused by the distortion compensation proce
5、ss, in order to obtain the optimal value of interference without the original image restoration or the original image, thereby improving the image quality. At present, the image restoration techniques have been widely used in space exploration, astronomical observation, material research, remote sen
6、sing, military science, medical imaging, traffic monitoring, criminal reconnaissance.In the application process in the long time, people have studied many restoration model and restoration method. The restoration method of mainstream: Wiener filter method, regularization filter, Lucy-Richardson algo
7、rithm and blind convolution method.At the same time,because theimagesharpeningcan highlightthe outline of the image,so it can beused for sometwo valueimage restoration.The principle is theoriginalimage and theimplementation of pulloperatorto himafter the results ofcombinedto produce asharpimage.Then
8、 carries on theexample analysis indigital image processing,and realizes thesharpening effect. In this paper, mainly through the basic principle of the Wiener filter, and combined with the function of MATLAB, design the corresponding Wiener filter, the realization of “noisy“ image restoration, the si
9、mulation experiment of degraded image restoration in degraded image, add noise to restore degraded images, experiments show that when there is noise must be considered the signal-to-noise ratio of the image, the autocorrelation function of noise for image restoration, can obtain the good restoration
10、 effect.Keywords: Wiener filtering; MATLAB; image restoration学士学位论文目 录 1 绪 论.11.1 课题背景和意义.11.2 国内外研究现状.21.3 课题主要内容.42 MATLAB 软件介绍.52.1 MATLAB 的概况.52.2 MATLAB 的语言特点.52.3 MATLAB 的开发环境.72.3.1 命令窗口.72.3.2 M 文件.82.3.3 路径浏览器与指令历史浏览器.92.3.4 图形视窗.112.3.5 帮助和演示系统.113 图像复原理论.133.1 图像退化/复原处理的模型.133.2 噪声的特征.143.3 噪声的分类.143.3.1 按干扰源分类.143.3.2 按对信号的影响分类.153.4 图像锐化的概念.
