二O一一 年 六 月 十 日
装 订 线
……………………. ………………. ………………… ……………….……. …………. …………. ………
****大学学士学位论文
目 录
1 绪论 ·································································································································1 1.1研究背景及其意义 ·······································································································1 1.2 噪声模型及噪声分类 ··································································································1 1.2.1 图像噪声的来源 ·······································································································1 1.2.2 噪声的类型 ···············································································································2 1.3 本文的主要工作·········································································································3 2 图像去噪理论与技术 ·····································································································4 2.1 空间域滤波 ··················································································································4 2.1.1 均值滤波去噪 ···········································································································4 2.1.2 维纳滤波去噪 ···········································································································7 2.1.3 中值滤波去噪 ···········································································································8 2.2 变换域滤波 ················································································································ 11 2.3 去噪性能的评价 ········································································································12 2.3.1 主观评价 ·················································································································12 2.3.2 客观评价 ·················································································································12 3 小波理论及其在图像处理中的应用 ···········································································14 3.1 小波函数 ····················································································································14 3.2 连续小波变换 ············································································································14 3.3 离散小波变换 ············································································································15 3.4 多分辨率分析 ············································································································16 3.5 尺度函数??t? ·············································································································17 3.6 MALLAT 算法 ··············································································································18 4 小波阈值去噪算法 ·······································································································22 4.1 小波阈值去噪的基本原理 ························································································22 4.2常用小波基 ·················································································································23 4.3 常用阈值选取方法 ····································································································23 4.4常用的阈值函数 ·········································································································24 4.5 仿真实验与结论 ········································································································27 4.5.1 软门限和硬门限阈值处理比较 ·············································································27 4.5.2 小波阈值去噪仿真 ·································································································28 4.5.3 实验图像对比 ·········································································································29 5 工作总结及展望 ···········································································································30 5.1 论文总结 ····················································································································30 5.2 工作展望 ····················································································································30 参考文献 ···························································································································31 致 谢 ·································································································································32 附 录 ·································································································································33
I
****大学学士学位论文
Contents
1 Foreword··························································································································1 1.1 Research background and significance·········································································1 1.2 Noise model and noise classification············································································1 1.2.1 Image noise source·····································································································1 1.2.2 Noise type···················································································································2 1.3 The main work of this paper··························································································3 2 Image denoising theory and technology···········································································4 2.1 Space domain filtering···································································································4 2.1.1 Average filtering denoising·························································································4 2.1.2 Wiener filtering denoising··························································································7 2.1.3 Median filtering denoising·························································································8 2.2 Transform domain filtering··························································································11 2.3 Evaluation of image denoising properties ··································································12 2.3.1 Subjective evaluation ······························································································12 2.3.2 Objective evaluation·································································································12 3 Wavelet theory and its application in image processing·················································14 3.1 Wavelet function·········································································································14 3.2 Continuous wavelet transform·····················································································14 3.3 Discrete wavelet transform··························································································15 3.4 Multi-resolution analysis·····························································································16 3.5 Scaling function···········································································································17 3.6 Mallat algorithm··········································································································18 4 Wavelet threshold denoising algorithm··········································································22 4.1 Wavelet threshold denoising basic principle·······························································22 4.2 Wavelet base used commonly·····················································································23 4.3 The threshold value method used commonly·····························································23 4.4 Common threshold function························································································24 4.5 Simulation results and conclusions·············································································27 4.5.1 The contrast of soft threshold and hard threshold processing··································27 4.5.2 Wavelet threshold denoising simulation···································································28 4.5.3 The contrast of experimental simulation ·································································29 5 Paper summary and prospect·····················································································30 5.1 Thesis summary ··········································································································30 5.2 Work prospect···········································································································30 References·························································································································31 Acknowledgments ············································································································32 Appendix···························································································································33
II
数字图像去噪算法的研究
【摘要】图像是人们获取信息的重要来源，但在图像的获取、传输和存储过程中经常会引入各种各样的噪声，使图像质量变差。因此，去除图像中的噪声，改善图像的质量，就成为数字图像处理的重要内容。图像去噪指的是利用各种滤波模型，通过传统滤波、小波、偏微分方程等多种方法从已知的含有噪声的图像中去掉噪声部分。图像去噪从整个图像分析的流程上来讲属于图像的预处理阶段，从数字图像处理的技术角度来说属于图像恢复的技术范畴，它的存在有着非常重要的意义。
论文介绍了空间域中几种经典的图像去除噪声的滤波方法，然后对经典的小波阈值去噪算法进行仿真。并在此基础上，提出了一种改进阈值的小波去噪方法。 【关键词】图像去噪、中值滤波、均值滤波、小波变换、阈值去噪
Research on Digital Image Denoising Algorithm
【Abstract】Optical images are very important for people to obtain information, but in the process of obtaining, transmission and storage, a great deal of noise is involved in and the quality goes bad. So image denoising is an important content in the digital signal processing. Image denoising refers to the use of various filter models, including the traditional filtering, wavelets, partial differential equations and other methods known to remove the noise from the noisy image. Image denoising belongs to the image preprocessing stage from the view of image analysis's flow,and belongs to the image restoration from the view of the digital image processing technology. Its existence has very important significance.
In this paper,several typical image denoising methods in space domain are first introduced.Then discussion is focused on classic wavelet threshold algorithm.On this basis,an improved threshold of wavelet denoising method are proposed.
【Key words】Image denoising,Median filtering denoising,Average filtering denoising Wavelet transformation ,Thresholding denoising
山东农业大学学士学位论文
1 绪论
数字图像处理技术涉及行业较广，有航空航天、工业检测、生物医学工程等，所涉及的技术有光学图像、微电子技术、计算机应用和数学分析等。数字图像预处理有去噪、增强分割、特征提取、识别等，图像去噪影响着图像的特征提取、图像融合等后续处理。
图像是人们获取信息的重要来源。据统计，在人类的信息中，视觉信息约占80%，俗语“百闻不如一见”就反映了图像在信息感知中的独到之处。另外，由于不同的成像机理，得到的初始图像中都含有大量不同性质的噪声，这些噪声的存在影响着人们对图像的观察，干扰人们对图像信息的理解。噪声严重的时候，图像几乎变形，更使得图像失去了存储信息的本质意义。显然，对图像进行去噪处理，是正确识别图像信息的必要特征。因此，去除图像中的噪声，改善图像的质量，就成为数字图像处理的重要内容。 1.1研究背景及其意义
图像去噪指的是利用各种滤波模型，通过传统滤波、小波、偏微分方程等多种
[1]方法从已知的含有噪声的图像中去掉噪声部分。图像去噪从整个图像分析的流程上来讲属于图像的预处理阶段，从数字图像处理的技术角度来说属于图像恢复的技术范畴，它的存在有着非常重要的意义[2]。主要表现在：
图像去噪处理，是正确识别图像信息的必要保证。通常初始图像中都含有大量不同性质的噪声，这些噪声影响人们对图像的观察，严重的会使图像发生变形，干扰人们对图像信息的理解。
(2) 图像去噪不仅能提高人视觉识别信息的准确性，而且是对图像作进一步处理的可靠保证。对一幅含有噪声的图像进行特征提取、配准或图像融合等处理，图像去噪是必需的。
(3) 研究图像去噪对数字图像其他处理环节性能的提升也有着促进意义。 任意一个图像处理系统，包括图像的获取、处理、发送、传输、接收、输出，每个环节都存在不同程度的噪声，使图像质量降低。一张模拟图像，当信噪比低于 14.2dB 时，图像分割时误检概率大于 0.5%，参数的估计误差大于 0.6%?3?。因此，研究图像去噪，有着非常重要的理论意义和应用价值[4]。 1.2 噪声模型及噪声分类 1.2.1 图像噪声的来源
1