基于K均值算法的数据聚类和图像分割研究

K均值算法利用K个聚类的均值作为聚类中心,通过对比样本到各聚类中心的距离,将样本划分到距离最近的聚类中,从而实现样本的聚类.分析了K均值算法的基本原理和实现步骤,并将其应用于数据聚类和图像分割,取得了较好的聚类效果.最后,针对K均值算法的不足之处,提出了改进措施,提高了K均值算法的聚类性能.

Study of Data Clustering and Image Segmentation Based on K-means Algorithm

K-means algorithm uses the means of K cluster as the cluster centers. By comparing the distance between the sample to all cluster centers,the sample is divided into the nearest cluster so as to realize the sample clustering. The basic principles and steps of K-means algorithm are analyzed,and its application in data clustering and image segmentation is given,which achieved good clustering results. Finally,aiming at the shortcomings of K-means algorithm,the improvement measures are proposed,which can improve the clustering performance of K-means algorithm.