节点文献

医学图像自动分割若干关键技术研究

Research on Key Technics of Automatic Segmentation in Medical Image

【作者】 朱峰

【导师】 宋余庆;

【作者基本信息】 江苏大学 , 模式识别与智能系统, 2006, 硕士

【摘要】 医学图像分割技术是医学图像处理与分析领域的重要课题之一,也是近年来备受研究人员关注的热点问题。医学图像分割的目的是把图像中具有特殊含义的不同区域分割开来,并使分割结果尽可能的接近解剖结构,从而为临床诊疗和病理学研究提供可靠依据。由于人体解剖结构的复杂性、组织器官形状的不规则性及个体之间的差异性,一般图像分割方法直接应用于医学图像并不能得到理想分割效果,为此必须寻找一种有效的医学图像分割方法。 论文在回顾总结国内外医学图像分割方法相关文献的基础上,阐述了医学图像分割的目的、意义。根据医学图像自身特点和特殊应用需求,分析了基于模糊最大熵和基于形变模型两种医学图像自动分割方法,并从分割模型本身与算法执行效率两个方面,对存在的问题提出了改进方法,较好地解决了分割精确度与时间复杂度之间的矛盾。论文主要工作包括: 1.总结分析了现有的医学图像分割方法优缺点,并分析了完全自动分割方法对实际应用的重大意义。 2.针对传统模糊熵公式不满足区域一致性条件等问题,提出模糊加权熵公式,并证明了该公式能满足图像分割五个基本条件,解决了传统方法会导致图像细节被均衡的不足。提出一种对称型抛物线分布作为隶属函数,该分布可降低了参数的维数,提高了运算的效率。

【Abstract】 Recently, medical image segmentation technology is one of the important subjects within medical image processing and analysis research field and worked over throughout the world. The main purpose of medical image segmentation is to divide the image into different regions with special signification and make the results approximate to the anatomic structure, which can provide the credibility gist for clinic diagnose and pathology research. As the complexity of human anatomic structure, the abnormity of tissue shape and the difference among individuals, the commonly image segmentation methods is not fit for the medical image. The availability method must be found to resolve the problem.The investigating statuses in quo and developments about medical image segmentation in domestic and foreign fields are reviewed and summarized. According to medical image characteristic and application demand, the paper discusses the two automatic methods of fuzzy maximum entropy and active contours model, improves the current problem from the model theory to algorithm efficiency and solves the contradiction of segmentation precision and time complexity more better, its main contents include:1. In this work, medical image segmentation methods are compared. Strong points and shortcomings of each method are particularly analyzed, we find automatic medical image segmentation methods is very important for the application of clinic diagnose.2. The Fuzzy weight entropy is proposed to against the

  • 【网络出版投稿人】 江苏大学
  • 【网络出版年期】2007年 02期
  • 【分类号】TP391.41
  • 【被引频次】27
  • 【下载频次】542
节点文献中: 

本文链接的文献网络图示:

本文的引文网络