Histogram normalization of two images using IHS

Question

Tanya 2014-5-11

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/129178-histogram-normalization-of-two-images-using-ihs

评论： Image Analyst 2014-5-12

I want a matlab code that will normalize one image taking other as reference using IHS I already have the seperate I, H and S components. I have to compute standard deviation and mean of H and S and then determine a new value and assign to H and S of target image. My code yields an error

clear;
clc;
img_a = imread('E:\TM\tanya\img3.tif');%reference
img_b = imread('E:\TM\tanya\img4.tif');%Target
%computing components for Target image
[r c] = size(img_a);
img_b_re = imresize(img_b,[r c],'bilinear');
img_b=im2double(img_b);
r=img_b(:,:,1);
g=img_b(:,:,2);
b=img_b(:,:,3);
th=acos((0.5*((r-g)+(r-b)))./((sqrt((r-g).^2+(r-b).*(g-b)))+eps));
H=th;
H(b>g)=2*pi-H(b>g);
H=H/(2*pi);
S=1-3.*(min(min(r,g),b))./(r+g+b+eps);
I=(r+g+b)/3;
hsi=cat(3,H,S,I);
HE=H*2*pi;
HE=histeq(HE);
HE=HE/(2*pi);
SE=histeq(S);
IE=histeq(I);
%figure, imshow(hsi),title('HSI Image');
img_b_re_hsi = hsi;
I = img_b_re_hsi(:,:,3);
hespc = imhist(I);
%RV=cat(3,HE,SE,IE);
 %       C=hsitorgb(RV);
  %      figure, imshow(C),title('RGB Image-HSI Normalized target image');
%computing components for reference image
img_a=im2double(img_a);
[r c] = size(img_b);
img_a_re = imresize(img_a,[r c],'bilinear');
r1=img_a(:,:,1);
g1=img_a(:,:,2);
b1=img_a(:,:,3);
th1=acos((0.5*((r1-g1)+(r1-b1)))./((sqrt((r1-g1).^2+(r1-b1).*(g1-b1)))+eps));
H1=th1;
H1(b1>g1)=2*pi-H(b1>g1);
H1=H1/(2*pi);
S1=1-3.*(min(min(r1,g1),b1))./(r1+g1+b1+eps);
I1=(r1+g1+b1)/3;
hsi1=cat(3,H1,S1,I1);
HE1=H1*2*pi;
HE1=histeq(HE1);
HE1=HE1/(2*pi);
SE1=histeq(S1);
IE1=histeq(I1);
%figure, imshow(hsi1),title('HSI Image 1');
img_a_re_hsi = hsi1;
I = img_a_re_hsi(:,:,3);
hespc1 = imhist(I1);
   SH=std(H);
   %M=mean(H);
  % SH1=std(H1);
   SS=std(S);
   %MS=mean(S);
   meanIm=mean([H(:) ; H1(:)]);
   meanSat=mean([S(:) ; S1(:)]);
   %SS1=std(S1);
   Hnew = (SH*100)/meanIm;
   Snew = (SS*100)/meanSat;
   hsinew = cat(3,Hnew,Snew,I1);
figure, imshow(S3),title('IHS normalised RGB Image'); 
 figure,imshow(img_a),title('Original');
  C=hsitorgb(hsinew);
 figure, imshow(C),title('RGB Image');