摘要核磁共振成像技术已成为脑部疾病诊断的最常用的手段之一。精确分割脑部核磁共振图像对脑部疾病的诊断、脑部疾病的分析与研究和脑部解剖等有着非常重要的意义。在各种图像分割方法中，阈值分割因为其实现简单、计算量小、分割结果较好的特点成为国内外专家学者的研究的热点。39975
论文的主要内容包括：介绍了与研究课题相关的背景知识或技术，主要包括核磁共振成像技术的概念和原理以及图像分割相关概念和原理。简要介绍了医学图像分割技术的研究现状，并对现有的一些热门的医学图像分割算法做了简要的介绍和分析，对常用的医学图像分割算法做了分类。
本文简单介绍二文Otsu法，其主要思想是：计算二文直方图中主对角线区域概率之和，来准确分割主对角线区域中背景和目标区域。然而，该方法增加了计算的复杂度，分割效率较低，不利于临床应用。
针对这一问题，本文对二文Otsu法进行优化，其优化思想是采用一种递推算法，找到像素点概率与其领域像素点的概率之间的关系，通过累加的方式计算对角线区域的概率之和，从而避免了二文Otsu法利用穷举法使计算复杂程度高的问题。
本文将优化的二文Otsu算法应用在脑MR图像的分割，为了验证该方法的有效性，进行仿真对比实验，实验结果说明本文方法的分割效果优于二文Otsu算法，分割效率更高。
毕业论文关键词: 图像分割 脑部核磁共振图像 二文Otsu法 阈值分割
MRI technology has become one of the most common means of brain disease diagnosis. Accurate segmentation of brain MR images of the brain disease diagnosis, analysis and research of brain disease and brain anatomy has a very important significance. In various image segmentation methods, threshold segmentation because of its simple, small amount of calculation of the characteristics of a good segmentation has become a hot research experts and scholars at home and abroad.
The main contents include: Introduction to the relevant background knowledge and research or technology, including the concepts and principles of nuclear magnetic resonance imaging technology and image segmentation concepts and principles. A brief introduction to medical image segmentation technology research, and some of the existing popular medical image segmentation algorithm to do a brief introduction and analysis of medical image segmentation algorithm used to do a classification.
This paper briefly describes the two-dimensional Otsu method, the main idea is: two-dimensional histogram calculation area of the main diagonal probabilities to accurately pided main diagonal background area and the target area. However, this method increases the computational complexity, low efficiency pision is not conducive to clinical application.
To solve this problem, this two-dimensional Otsu method to optimize its optimization thought is to use a recursive algorithm to find the relationship between probability and its probability of pixels between pixels field was calculated by summing the diagonal region of way probabilities, thus avoiding a two-dimensional Otsu method makes use of brute-force method to calculate the high complexity of the issue.
In this paper, enhanced 2D Otsu algorithm in brain MR images, in order to verify the validity of the method, comparative simulation experiment results show that this method is superior segmentation dimensional Otsu algorithm, split more efficient.
Keywords: Image segmentation Brain magnetic resonance image 2D Otsu Threshold segmentation
目 录
摘 要 I
Abstract II
1 绪论 1
1.1 引言 1
1.2 脑部核磁共振图像（MRI）的分割技术的难点 2
1.3 国内外的研究现状 2 二维Otsu脑MR图像分割研究:http://www.751com.cn/yixue/lunwen_40713.html