| 摘要: |
| 为了避免随机选取初始聚类中心点的缺陷，利用最大最小距离的方法确定初始聚类中心点.实验结果表明，和传统的模糊C均值聚类（FCM）算法相比，所提聚类算法具有较高的稳定性和准确性，所分割的胼胝体图像边缘信息更加清晰. |
| 关键词: 模糊C均值聚类(FCM) 最大最小距离 初始聚类中心 胼胝体 |
| DOI：10.3969/J.ISSN.1000-5137.2018.04.013 |
| 分类号:TP391 |
| 基金项目:国家自然科学基金（61373004）；上海师范大学校级基金（A700115001005，Sk201220） |
|
| DTI image segmentation based on the improved fuzzy C-means clustering |
|
Fang Bowen, Zhang Xiangfen, Ma Yan, Li Chuanjiang, Zhang Yuping, Yang Yanqin
|
|
The College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 200234, China
|
| Abstract: |
| In order to avoid the shortcomings of selecting initial clustering center points randomly,we use the principle of maximum and minimum distance to determine the initial clustering center points.Compared with traditional fuzzy C-means clustering (FCM)algorithm.The experimental results show that the accuracy and stability of improved FCM algorithm has been improved,and corpus callosum edge information is clearer. |
| Key words: fuzzy C-means clustering (FCM) maximum and minimum distance initial clustering center corpus callosum |
Editorial Board