基于特征散度的图像FCM聚类分割  

FEATURE DIVERGENCE BASED FCM CLUSTERING FOR IMAGE SEGMENTATION

作　　者：薛景浩[1] 章毓晋[1] 林行刚[1] 

机构地区：[1]清华大学电子工程系,北京100084

出　　处：《模式识别与人工智能》1998年第4期462-467,共6页Pattern Recognition and Artificial Intelligence

基　　金：国家自然科学基金

摘　　要：本文提出了一种称为特征散度的新概念和一种基于特征散度的图像模糊C-均值(FCM)聚类分割方法。该方法采用特征散度取代传统的欧氏距离作为图像像素与C-聚类典范值之间的差异性度量。本文将新方法与一些传统的FCM算法同时应用于图像分割,并采用形状测度和均匀测度评价各算法的分割性能,结果表明新方法对不同类型的图像都具有良好的分割性能。In this paper, we introduce a new concept called feature divergence and propose an image segmentation technique using fuzzy c-means (FCM) clustering based on feature divergence. In this new FCM algorithm, traditional Euclidean norm metric is replaced by feature divergence as the measure of dissimilarity between pixel and cluster prototypes. The effectiveness and generality of this new algorithm are evaluated and justfied using uniformity measure and shape measure in the experiments of segmenting various real images along with several existing FCM methods.

关 键 词：特征散度 图像分割 FCM 聚类分割 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]