基于Matlab实现的Kmeans聚类算法，并将其运用至灰度图像分割中.zip

clear all; close all; clc im = imread(uigetfile('.bmp')); k = 2; %定义统计图像直方图信息变量 img_hist = zeros(256,1); hist_value = zeros(256,1); %统计直方图不同灰度对应的像素个数 for i=1:256 img_hist(i)=sum(sum(im==(i-1))); end; for i=1:256 hist_value(i)=i-1; end; %聚类类别初始化 cluster = zeros(k,1);%聚类的质心 cluster_count = zeros(k,1);%聚类类别个数统计 for i=1:k cluster(i)=uint8(rand*255); end; old = zeros(k,1);%old存放迭代前的质心 %迭代终止条件 while (sum(sum(abs(old-cluster))) >1e-6) old = cluster; closest_cluster = zeros(256,1); %初始化mindistance，存放所有灰度级到质心1的距离 min_distance = uint8(zeros(256,1)); min_distance = abs(hist_value-cluster(1)); %记录每个灰度级到自己所属类的最小距离 for i=2:k min_distance =min(min_distance, abs(hist_value-cluster(i))); end; %确定每个灰度级所属类别序号 for i=1:k closest_cluster(min_distance==(abs(hist_value-cluster(i)))) = i; end; %计算该类别的数量 for i=1:k cluster_count(i) = sum(img_hist .*(closest_cluster==i)); end; %更新质心 for i=1:k if (cluster_count(i) == 0) cluster(i) = uint8(rand*255); else cluster(i) = uint8(sum(img_hist(closest_cluster==i).*hist_value(closest_cluster==i))/cluster_count(i)); end; end; %为图像分配每个灰度级所处坐标的最终质心 imresult=uint8(zeros(size(im))); for i=1:256 imresult(im==(i-1))=cluster(closest_cluster(i)); end; %对每个灰度级对应坐标分配标签 clustersresult=uint8(zeros(size(im))); for i=1:256 clustersresult(im==(i-1))=closest_cluster(i); end; %对每次迭代结果进行展示 result_image=uint8(zeros(size(im))); for i = 1:128 for j = 1:128 if clustersresult(i,j)==1 result_image(i,j)=0; else result_image(i,j)=255; end end end result_image = uint8(result_image); figure subplot(1,2,1); imshow(im); subplot(1,2,2); imshow(result_image); end