| 基于模糊熵聚类和改进粒子群的MRI脑图像分割研究 |
| |
| 引用本文： | 郑伟,姚纪智,王洁,刘帅奇,张晓丹,马泽鹏. 基于模糊熵聚类和改进粒子群的MRI脑图像分割研究[J]. 激光杂志, 2021, 42(1): 98-103 |
| |
| 作者姓名： | 郑伟 姚纪智 王洁 刘帅奇 张晓丹 马泽鹏 |
| |
| 作者单位： | 河北大学电子信息工程学院,河北保定 071002;河北省数字医疗工程重点实验室,河北保定 071002;河北省机器视觉工程技术研究中心,河北保定 071002;河北大学附属医院,河北保定 071000 |
| |
| 基金项目： | 国家自然科学基金资助项目(No.2014000000);国家自然科学基金(No.61401308、No.61572063);河北省自然科学基金(No.F2016201142;No.F2016201187);河北省教育厅项目(No.QN2016085);河北省机器视觉工程技术研究中心开放课题(No.2018HBMV02);河北大学引进人才科研启动经费(No.2014-303);河北大学高性能计算平台支持。 |
| |
| 摘 要： | 现有医学图像生成过程中无法回避噪声的引入,而目前还未有较好的算法对高噪声的MRI医学图像进行分割,分割归属于聚类问题,聚类常用的方法是模糊聚类,但模糊聚类需要解决对噪声和初始化敏感的问题,提出了一种基于模糊熵聚类和粒子群优化算法的MRI脑图像分割算法.首先在模糊熵聚类算法的基础上进行改进,设计了 一种利用邻域空间信息的...
|
| 关 键 词： | 图像分割 脑MRI 模糊熵聚类 邻域空间信息 粒子群优化算法 |
Segmentation of MRI brain image based on fuzzy entropy clustering and improved particle swarm |
| |
| ZHENG Wei,YAO Jizhi,WANG Jie,LIU Shuaiqi,ZHANG Xiaodan,MA Zepeng. Segmentation of MRI brain image based on fuzzy entropy clustering and improved particle swarm[J]. Laser Journal, 2021, 42(1): 98-103 |
| |
| Authors: | ZHENG Wei YAO Jizhi WANG Jie LIU Shuaiqi ZHANG Xiaodan MA Zepeng |
| |
| Affiliation: | (College of Electronic Information Engineering,Hebei University,Baoding Hebei 071002,China;Hebei Key Laboratory of Digital Medical Engineering,Baoding Hebei 071002,China;Hebei Machine Vision Engineering Technology Research Center,Baoding Hebei 071002,China;Affiliated Hospital of Hebei University,Baoding Hebei 071000,China) |
| |
| Abstract: | The noise introduce is inevitably during the existing medical imaging,and there is no good segmentation algorithm for high-noise MRI medical image so far.The segmentation belongs to clustering problem,and the commonly method is fuzzy clustering which needs to solve the sensitives of noise and initialization.This paper proposes the MRI brain image segmentation algorithm based on fuzzy entropy clustering and particle swarm optimization.Firstly,the improvement is carried out in fuzzy entropy clustering to design a new target function with kernelized fussy entropy clustering based on neighborhood spatial information.Then,a new algorithm based on improved particle swarm optimization is proposed.Finally,the white matter,the gray matter and the cerebrospinal fluid in are segmented by the optimization objective function.The MRI images in Montreal neurological institute database are selected in this article,and the proposed algorithm is compared with several existing clustering segmentation algorithm,the simulation results show that the presented algorithm can solve the noise and initialization problems of fuzzy clustering,and realize the precise segmentation of MRI image with high-noise. |
| |
| Keywords: | image segmentation brain MRI fuzzy entropy clustering neighborhood space information particle swarm optimization algorithm |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
