k-means算法是一种动态聚类算法，基本原理如下[24]：首先预先定义分类数k，并随机或按一定的原则选取k个样品作为初始聚类中心；然后按照就近的原则将其余的样品进行归类，得出一个初始的分类方案，并计算各类别的均值来更新聚类中心；再根据新的聚类中心对样品进行重新分类，反复循环此过程，直到聚类中心收敛为止。-K- means algorithm is a dynamic clustering algorithm, the basic principle of [24] as follows: first of all the predefined class number k, and random or according to certain principle to choose k as a sample as the initial cluster centers Then according to the principle of the nearest to categorize the rest of the samples, it is concluded that an initial classification scheme, and various other calculated mean to update the cluster centers The samples again according to the new clustering center to reclassify, recycling the process, until the clustering center of convergence.