基于MATLAB的维纳滤波在图像恢复中的应用.pdf

江 西 理 工 大 学 本 科 毕 业 设 计 （论文） 题 目：基于MATLAB的维纳滤波在图像恢复中的应用 1 摘 要 本文主要研究的是基于MATLAB的维纳滤波在图像恢复中的应用, 在实际的 日常生活中，人们要接触很多图像，画面。而在景物成像这个过程里可能会出现 模糊、失真或混入噪声，最终导致图像质量下降，这种现象称为图像 “退化”。 退化的数字图像会造成图像中的目标很难识别或者图像中的特征无法提取，必须 对其进行恢复。维娜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；退化模型 2 ABSTRACT Main context of this thesis lies in the Wiener filter based on MATLAB being appliedto image restoration.In real life,peoplewillmeet a lot of images and screens. However, in this process, imaging features may appear blurred, distorted or mixed with noise. As a consquence, quality of images is lowered, and this phenomenon is just described as Image "degraded."Degradation in digital images is likely to make it diffcult to identify the target image or to extract the image features, images must be restored, then. Wiener filter is a common method for image restoration, the idea of this method is to minimize the mean square error between restored images and the original ones. This paper mainly introduces the basic principles of Wiener filtering, and function of MATLAB are combined to design the corresponding Wiener filter, which aimes at restoration of "noisy" images. Besides, simulation experiments of degraded images restoration are performed as well. And noise restoration are also taken into account in the experiments.Finally, experiments show that SNR of images should be taken into consideration when there is noise in the degraded images. Combining with noise autocorrelation function for image restoration, we will achieve betterrehabilitationresults. Keywords:Wiener filter;MATLAB