matlab自带有histeq函数对图像进行直方图均衡
自己写了一个，改成向量化形式，效率提高了一点，但是比自带的还是差很多，差不多9倍
function D = my_histeq(I) [m,n] = size(I); S = zeros(1,256,\'double\'); % 得到0-255各个灰阶的计数 i = 1:m;j = 1:n; S(I(i,j)+1) = S(I(i,j)+1) + 1; % 得到各灰度阶的概率 S = S./(m*n); % 得到概率累计分布函数 P = zeros(1,256,\'double\'); P(1) = S(1); k = 2:256; P(k) = P(k-1) + S(k);
% 投影到灰度阶 D = zeros(m,n,\'uint8\'); D(i,j) = P(I(i,j)+1)*255+0.5;
原来使用for循环的
function D = my_histeq(I) [m,n] = size(I); S = zeros(1,256,\'double\'); % 得到0-255各个灰阶的计数 for i = 1:m for j = 1:n S(I(i,j)+1) = S(I(i,j)+1) + 1; % 注意下标 end end % 得到各灰度阶的概率 S = S./(m*n); % 得到概率累计分布函数 P = zeros(1,256,\'double\'); for i = 1:256 if i==1 P(i) = S(i); else P(i) = P(i-1) + S(i); end end
D = zeros(m,n,\'uint8\'); for i = 1:m for j=1:n D(i,j) = P(I(i,j)+1)*255+0.5; end end
