PS 图像调整算法——自动色阶 (Auto Levels)

PS 给出的定义: 

Enhance Per Channel Contrast:Maximizes the tonal range in each channel to produce a more dramatic correction.Because each channel is adjusted individually, Enhance Per Channel Contrast may remove or introduce color casts.The Auto Levels command uses this algorithm.

简单来说,就是对R,G,B三个通道的动态范围分别进行拉伸。这个算法还要考虑一个 clipping percentage, 默认值是 0.1%。根据这个定义,可以先编写一个拉伸动态范围的函数:

(函数中默认图像的动态范围是 0-1).

if x<x_min:  y=0;

if x>x_max:  y=1;

if x_min < x< x_max:   y=(x-x_min)/(x_max-x_min); 


function I_out=F_color(I, percent)

%%% the tonal range of the input image is 0-1.
[row, col]=size(I);
I_sort=sort(I(:));
I_out=I;

%%% based on the clipping percentage, 
%%% compute the upper and lower boundaries 

if (percent==0)
    I_min=min(I_sort)
    I_max=max(I_sort)
else
    I_min=I_sort(floor(row*col*percent))
    I_max=I_sort(floor(row*col*(1-percent)))
end

for i=1:row
    for j=1:col
            if(I(i,j)<I_min)
                I_out(i,j)=I_min;
            elseif(I(i,j)>I_max)
                I_out(i,j)=1;
            else
                I_out(i,j)=(I(i,j)-I_min)*(1-I_min)/(I_max-I_min)+I_min;
            end
    end 
end


对三个通道分别进行拉伸,可以得到最终调整后的图像。

clc;
clear all;
Image=imread('8.jpg');
Image=double(Image)/255;
imshow(Image);
R=Image(:,:,1);
G=Image(:,:,2);
B=Image(:,:,3);
percent=0.001;
Image_out=Image;
Image_out(:,:,1)=F_color(R, percent);
Image_out(:,:,2)=F_color(G, percent);
Image_out(:,:,3)=F_color(B, percent);
figure, imshow(Image_out);


原图:



调整后的图:



原文地址:https://www.cnblogs.com/muyuge/p/6152399.html