首页 | 本学科首页   官方微博 | 高级检索  
     

结合聚类和改进的C-V演化方程在医学图像分割中的应用
引用本文:罗志宏,冯国灿,成秋生. 结合聚类和改进的C-V演化方程在医学图像分割中的应用[J]. 计算机应用, 2008, 28(9): 2288-2291
作者姓名:罗志宏  冯国灿  成秋生
作者单位:中山大学,计算机科学系,广州,510275;中山大学,数学与计算科学学院,广州,510275;中山大学,数学与计算科学学院,广州,510275;广州市第一人民医院,脑内科,广州,510180
基金项目:国家自然科学基金,教育部科学技术研究重点项目
摘    要:在传统的C-V模型的基础上提出一类改进的C-V演化方程,利用它们与聚类技术相结合对病理状态脑部CT和MR图像进行分割,能获得较理想的实验结果。特别是利用提出的演化方程对一个合成的几何图像进行分割,结果显示能比一些传统的演化方程具有更好的稳健性和准确性。

关 键 词:活动轮廓模型  水平集  图像分割  k-means聚类
收稿时间:2008-04-15

Application of clustering and improved C-V evolution equation in medical image segmentation
LUO Zhi-hong,FENG Guo-can,CHENG Qiu-sheng. Application of clustering and improved C-V evolution equation in medical image segmentation[J]. Journal of Computer Applications, 2008, 28(9): 2288-2291
Authors:LUO Zhi-hong  FENG Guo-can  CHENG Qiu-sheng
Affiliation:LUO Zhi-hong1,2,FENG Guo-can2,CHENG Qiu-sheng 3(1.Department of Computer Science,Sun Yat-Sen University,Guangzhou Guangdong 510275,China,2.School of Mathematics , Computational Sciences,3.Department of Neurology,the First Municipal People's Hospital of Guangzhou City,Guangzhou Guangdong 510180,China)
Abstract:A class of improved C-V evolution equation based on classical C-V model was proposed in this paper. It integrated with k-means clustering to segment pathological tissue of Brain CT and MRI, and satisfactory results were obtained. Specially, experimental result of a synthetic image shows that the proposed method is more accurate and robust than some traditional evolution equations.
Keywords:active contour model  level set  image segmentation  k-means clustering
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《计算机应用》浏览原始摘要信息
点击此处可从《计算机应用》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号