说明：  K均值聚类 1、任取[0,255]中K个值作为第一次分类中心 2、把图像（gray）中的每一个点与K个中心值进行比较，按差值分成K组，求出新的K个中心点 3、重复上一步骤直到前后两次算出的中心点值相同 4、按最后得到的中心点把图像依据点到中心点的距离分成K类。
(K-means clustering  1, take any [0,255] in the K value as the first classification center  2, the image (gray) in each of the K center point value, the difference divided by the K group, obtains a new center of K  3, repeat the previous step until  center around twice the value calculated in the same  4, press the center of the resulting image based on the distance to the center is divided into K classes.)