matlab具体实现k均值算法进行彩色图像目标分割。.zip

function [A_new,label] = kmean_func(A,K) A = double(A); Ar = A(:,:,1); Ag = A(:,:,2); Ab = A(:,:,3); [m,n] = size(Ar); r = zeros(1,K); g = zeros(1,K); b = zeros(1,K); for i=1:K r(i) = A(randi([1,m],1),randi([1,n],1),1); g(i) = A(randi([1,m],1),randi([1,n],1),2); b(i) = A(randi([1,m],1),randi([1,n],1),3); end label = zeros(m,n); for k = 1:100 for i = 1:m for j = 1:n temp_value = 0; for dis = 1:K value = (Ar(i,j)-r(dis))^2 + (Ag(i,j)-g(dis))^2 + (Ab(i,j)-b(dis))^2; if temp_value == 0 || temp_value>value temp_value = value; label(i,j) = dis; end end end end for i = 1:K r(i) = mean(Ar(label == i)); g(i) = mean(Ag(label == i)); b(i) = mean(Ab(label == i)); end end A_new = zeros(m,n,3); for i=1:m for j=1:n A_new(i,j,1) = r(label(i,j)); A_new(i,j,2) = g(label(i,j)); A_new(i,j,3) = b(label(i,j)); end end A_new = uint8(A_new); end