《论文《超声图像纹理分析算法研究》》-毕业论文.doc

PAGE 南京邮电大学 毕 业 设 计（论 文） 题 目：超声图像纹理分析算法研究 专 业： 生物医学工程 学生姓名： 李 成 班级学号： 指导教师： 杨 浩 指导单位： 地理与生物信息学院 日期：2010年 3 月 12日至 2010年 6 月 17日 摘 要 医学超声图像与计算机X射线断层造影术、磁共振成像、核医学成像并称为现代四大医学影像技术，其中超声成像由于其独具的实时性、无损性、廉价性、可重复性好、灵敏度高等优势而广泛应用于临床诊断和治疗中，它在人体内部组织器官的定量分析、实时监控和治疗规划等方面都具有极大的潜力。运用计算机手段对超声图像进行分析，为医生的诊断提供量化的依据，是非常必要的。 本文将肝脏的超声图像作为研究对象，结合超声理论、纹理分析理论，探讨了肝脏所具有的纹理特性，并运用灰度共生矩阵方法提取肝脏超声图像的纹理特征参数，分别选取正常肝脏、脂肪肝以及肝硬化的超声图像进行纹理分析，依据肝脏超声图像的特点对其结果进行对比分析。进而验证程序的准确性，为今后进一步进行超声图像的分割、分类及自动识别打下基础。 关键词： 超声图像；肝脏；纹理分析；灰度共生矩阵  ABSTRACT Medical ultrasound imaging, computerized tomography ,magnetic resonance imaging and nuclear imaging are the four important medical imaging techniques in modem society. Among them , medical ultrasound imaging is widely used in medical diagnosis and treatment due to its unique characteristic of real-time, noninvasiveness, cheapness, easy repetition, and high sensitivity. It has high potential in quantitive analysis, real-time monitoring and treatment scheduling, etc. At present, the diagnosis is determined naturally, the conclusion of disease is influenced by doctors. Therefore, it is needed eagerly to improve ultrasound image diagnosis by using computer. In this paper, we studied the images of liver ultrasound images by contracting the theory of ultrasound, texture analysis. We discussed the texture characteristics of liver. Subsequently we validated this lows by using GLSM method to extract the texture characters. Choose normal respectively, liver cirrhosis and the fat of ultrasound image in texture, based on the liver of ultrasound image on the results on the comparative analysis. to verify the accuracy, For the further ultrasound image segmentation, classification and automatic identification basis. Keywords: ultrasound images ；liver ；texture analysis；GLCM 目 录 TOC \o 1-3 \h \z \u HYPERLINK \l _Toc263816755 第一章 绪论 PAGEREF _Toc263816755 \h 1