小波变换在图像分割中的应用..doc

存档日期：_____________________ 存档编号：______________________ 徐 州 师 范 大 学 论文题目: 小波变换在图像分割中的应用 姓 名: 王 婷 婷 学 号： 078333106 专 业： 自 动 化 班 级： 07 自 动 化 指导老师： 李 旭 超 科文学院教务处印刷 摘 要 近年来，对图像分割的研究一直是图像技术研究的焦点。图像分割是一种很重要的图像分析技术，它的目的是把图像分为具有各种特性的区域并把感兴趣的部分提取出来。它融合了多个学科的成果，并且成功应用于工业、农业、医学、军事等领域，得到了广泛的应用。 图像分割的方法有很多种，本文的题目是小波变换在图像分割中的应用。主要介绍了两种小波分割方法，即小波阈值分割方法和小波域马尔可夫随机场模型分割方法，其中通过了直方图、建立模型等手段对这两种方法做出具体的讨论。 经过研究发现，区分真正的噪声和边缘是图像分割的难题之一，然而小波变换则可以解决这一问题，小波变换是一种时-频两域分析的工具。同时还利用Matlab分别对两种方法进行仿真，并得到了有效的结果。根据仿真结果我们可以看出不同分割方法的不同分割效果，从而更好地理解这些方法。 图像分割是一个经典的问题，至今仍没有一种通用的解决方法。 关键词：图像分割 小波变换 阈值 马尔可夫随机场模型 Abstract Recently, the image segmentation is still a focus of image processing. It is a so important technology in the field of image processing, the aim of which separates the object into the regions with different characters, and pick up the interesting regions. It incorporates multiple disciplines, it applied into industry, agriculture, medicine, military and other fields successfully, and been widely used. There are a lot of kinds of image segmentation, the topic of this paper is The application of wavelet transform to the image segmentation. It mainly introduces two kinds of wavelet decomposition methods, they are respectively the wavelet threshold segmentation method and the wavelet domain markov model segmentation method with the airport, Among them through the histogram, model and discuss the means of the two methods. After the study found, distinguish true noise and edge is one of the difficult problems of the image segmentation, however wavelet transform can resolve this problem better, because wavelet transform is a kind of time-frequency analysis tools. Then there are two summations for two methods with Matlab, the result can show the effective phenomenon. According to the simulation result ,we can see the different image segmentation effect of different meth