毕业论文移动版

摘要图像复原在图像处理领域中是重点也是难点，是在已知退化图像的基础上，对其质量进行恢复强化。图像复原在天文学、医药学等众多的领域中广泛应用。大气湍流退化图像的恢复是图像复原领域中一个常见问题，其模糊性对天文学图片有很大的影响。图像复原的实质是在确定退化模型的情况下，进行求逆操作。87159

本文主要研究内容包括以下几个方面：

（1）图像复原研究相关情况、理论基础，包括图像中存在的噪声、退化图像的模型、图像复原的原理。

（2）图像质量的评价标准，包括主观评价和客观评价两部分。其中客观标准作为主要评价标准。本文选取均方差、峰值信噪比和图像熵作为后续图像复原算法分析比较的评价标准。

（3）逆滤波、维纳滤波、最小二乘法三种经典图像复原算法的分析比较。选取大气湍流模型退化原始图像并添加噪声，仿真结果表明，最小二乘法优于维纳滤波、逆滤波算法。

（4）维纳滤波算法的改进，使得改进后的维纳滤波算法的K值自动估计。仿真结果表明，改进后的维纳滤波算法优于经典的最小二乘算法。

毕业论文关键词：大气湍流模糊；图像复原；维纳滤波

Abstract  Image restoration is the key and difficult point in the field of image processing，which restores and enhances its quality on the basis of the known degraded images。 Image restoration is widely used in astronomy, medicine and many other fields。 Restoration of atmospheric turbulence degraded images is a common problem in the field of image restoration, which has a great influence on astronomy pictures。 The essence of image restoration is to carry out inverse operation after determining the degradation model。

The main research contents of this paper include the following aspects:

(1) Relevant circumstances and theoretical basis of image restoration research, including noise of image, model of the degraded image, principle of image restoration。

(2) Evaluation criteria of image quality includes two parts: subjective evaluation and objective evaluation。 Objective criteria is the main evaluation criteria。 In this paper, the average variance, the peak signal-to-noise ratio and the image entropy are selected as the evaluation criteria for the analysis and comparison of the following image restoration algorithms。 

(3) Analysis and comparison of three classical image restoration algorithms: Inverse Filtering, Wiener Filtering, Least Square Method。 In the atmospheric turbulence model, the original image is degraded and the noise is added。 The simulation results show that least square method is better than Wiener Filtering and Inverse Filtering。

(4) Improvement of Wiener Filtering algoritm, which estimates K value automatically。 The simulation results show that the improved Wiener Filtering algorithm is better than the classical least squares algorithm。

Keywords: Atmospheric turbulence blur; Image restoration; Wiener filter

目  录

第一章  绪论 1

1。1 研究的背景和意义 1

1。2 图像复原的发展和国内外现状 2

1。3 图像复原的应用领域 2

1。4 本文研究的主要内容及结构安排 3

1。5 本章小结 3

第二章  图像复原的理论基础 5

2。1 图像的噪声 5

2。2 模糊图像的退化模型 6

2。2。1 一般退化模型 (责任编辑：qin)