作　　者： ; ; ;

机构地区： 华南师范大学计算机学院

出　　处： 《计算机科学》 2006年第10期186-188,共3页

摘　　要： 针对CT医学图像的特点,本文将遗传算法与聚类分析两种工具相结合,应用到医学CT图像分割中。对K均值聚类做了简要分析和评论,在此基础上将遗传算法引入聚类分析中,利用遗传算法搜索的随机性和并行性,从而克服了K均值聚类的局部性和对初始聚类中心的敏感性;并且可以根据分割的要求,合理地调整聚类时的特征向量和权重。试验结果表明上述方法是可行的,达到了较好的分割效果。 Aiming at the characteristic of medical images, this paper integrates genetic algorithm with clustering analysis and applies in medical CT image segmentation. K-means clustering is introduced and remarked firstly. On the basis of systematic analysis of current algorithms, genetic algorithm is inducted into clustering analysis to solve the locality and the sensitiveness of the initial condition of K-means clustering by searching randonly and in parallelism. Besides, according to different request, feature vector and weight factors are adjusted rationally. The example shows that the method is feasible,and good segmentation results have got.

关 键 词： 图像分割 均值聚类 遗传算法

领　　域： [自动化与计算机技术] [自动化与计算机技术]