摘要
在分析 CT原理的基础上 ,通过对原始 CT数据进行预处理 ,增强所需对象的特征 ,然后以人体组织的 CT值为聚类特征 ,采用 K-均值聚类分割方法对 CT数据进行分割处理 ,将所需的组织与其他组织分离开 ,从而实现轮廓提取 。
Based on the analysis of CT principle, we adopt the CT datum in medicine and make pre process on them, thus strengthening the features of objects we need and restraining those we don't. Then, we make the CT value of human bodies' organizations as clustering features, and adopt K Average Value Clustering Division to divide and process the CT datum. At last, we separate the organizations of which we need from those we don't, thereby making foundation for further profile vectorization.
出处
《武汉理工大学学报》
CAS
CSCD
2003年第4期58-60,共3页
Journal of Wuhan University of Technology
基金
教育部骨干教师项目资助
关键词
CT
聚类
均值
轮廊
CT
cluster
average
profile