本文在阐述聚类分析方法的基础上重点研究FCM聚类算法。FCM算法是一种基于划分的聚类算法，它的思想是使得被划分到同一簇的对象之间相似度最大，而不同簇之间的相似度最小。最后基于MATLAB实现了对图像信息的聚类。- This paper describes the basis of cluster analysis methods and focus on the FCM clustering algorithm. FCM algorithm is a clustering algorithm based on division and the idea is to make it to the same cluster is divided into the biggest similarity between the objects, while the minimum similarity between different clusters. Finally, I realize the implementation of image information of the cluster based on MATLAB.