Adding black and white pixels into an image.

I have a guassian lab where I have to show dead pixles (black and white)

The first part of my code has this where 128= mean and std = 8and the image size is 256*256.

When I run this code i get a pretty gray picture, which was expected.

I = normrnd(128,8,[256,256]);

imshow(I,[0,255]);

Now, i have to find the uniform distibution to replace 5% of the samples with black (i=0) and 5% of the pixels white (i=255). I'm having a hard time doing that.

Answer 1

[KSSV]

I = normrnd(128,8,[256,256]);
imshow(I,[0,255]);
% Pick indices to replace with black and white
[m,n] = size(I) ; 
idx = randperm(m*n,round(m*n*10/100)) ;  % 5% +5% = 10%
% Replace with black 
idx1 = randsample(idx,length(idx)/2) ; 
idx2 = setdiff(idx,idx1); 
I(idx1) = 0 ; % replace with black
I(idx2) = 255 ; % Replace with white 
imshow(I)

[Rafael Rivas]

It looks better than what I had but for the I(idx), should it be I(idx1) and I(idx2)? Otherwise when I run the code, I only get gray and white and no black?

[KSSV]

Yes yes... Typo error... Edited the code...