how to split darker and brighter pixels using a threshold value?

in=imread('window.jpg');
i_in=im2double(in);
i_in=imresize(i_in,[4 4]);
imshow(i_in);
figure(1);
%%LOGRATHIMIC NORMALIZATION
do=rgb2ycbcr(i_in);
% % Extract color 
%logarithmic Normalization.
 y  =do(:,:,1); % Luminance channel
 cb =do(:,:,2); % chroma blue channel
 cr =do(:,:,3); % chroma red channel
 Lin=0.299*do(:,:,1)+0.587*do(:,:,2)+0.114*do(:,:,3);
x=(max(max(Lin)))+1;
Llog=((log(Lin)+1)/log(x));
figure(2);
imshow(Llog);
A=Llog;
%splitting  darker and bright pixels
m=4;
%For bright areas
for x=1:m
for y=1:m
if A(x,y)>0.5
       b(x,y)=A(x,y);
end
end
end
%for dark areas
for x=1:m
for y=1:m
if A(x,y)<0.5
    d(x,y)=A(x,y);
end
end
end

In the above code i got an error on bright area..In b(x,y) i only need pixels values which is greater than threshold,but i didnt get it..I have attached the Input and Bright values of an image for your reference.can you tell me what mistake i have done..

Answer 1

[Ameer Hamza]

There is no need to use for loop. Are you looking for something like this

close all

in=imread('window.jpeg');
i_in=im2double(in);
% i_in=imresize(i_in);
figure(1);
subplot(321)
imshow(i_in);
title('original')
%%LOGRATHIMIC NORMALIZATION
do=rgb2ycbcr(i_in);
% % Extract color
%logarithmic Normalization.
y  =do(:,:,1); % Luminance channel
cb =do(:,:,2); % chroma blue channel
cr =do(:,:,3); % chroma red channel
Lin=0.299*do(:,:,1)+0.587*do(:,:,2)+0.114*do(:,:,3);
x=(max(max(Lin)))+1;
Llog=((log(Lin)+1)/log(x));
% figure(2);
subplot(322)
imshow(Llog);
title('LOGRATHIMIC NORMALIZATION')
A=Llog;

B = A;
B(B<0.5) = 0;
D = A;
D(D>0.5) = 0;

% figure(3);
subplot(323)
imshow(B)
title('brigh portion')
% figure(4);
subplot(324)
imshow(D)
title('Dark portion')

B_mask = A>0.5;
D_mask = A<0.5;

% figure(5);
subplot(325)
imshow(B_mask)
title('Bright Mask')
% figure(6);
subplot(326)
imshow(D_mask)
title('Dark Mask')

[Rocky]

Thanks a lot!!

[Ameer Hamza]

Glad to be of help.