I want column matrix in for loop

I want to create a matrix where the R (subtracted) matrix is stored in C matrix such that all 9 R(i,j) elements produced from 10 images get stored in first column of matrix c, then next R(i,j) be stored in 2nd column of matrix c and so on.

Here x is row no.

and y is column no.

use whatever image you like with 400X400 pixels

n=10
for x=11:1:13
    for y=11:1:13
        for i=1:1:9
            for k =1:1:(n-1)
                image1 = double(imread(sprintf('image_%.4d.jpg',k)));
                I1=image1(1:400,1:400);
                image2 = double(imread(sprintf('image_%.4d.tiff',k+1)));
                I2=image2(1:400,1:400);
                R= imsubtract(I2,I1);
                
                c(k,i)=R(x,y);
            end
        end
    end
end

Answer 1

[DGM]

I'm not sure, but this might be close to what you want.

n=10
xrange = 11:13;
yrange = 11:13;

for k = 1:(n-1)
    image1 = imread(sprintf('image_%.4d.jpg',k));
    image2 = imread(sprintf('image_%.4d.tiff',k+1));
    
    I1 = double(image1(xrange,yrange));
    I2 = double(image2(xrange,yrange));
    
    c(k,:) = I2(:)-I1(:);
end

[HEMRAJ PATEL]

Sorry, but this didn't work

[DGM]

You'll need to explain how it doesn't work or how it doesn't meet your requirements.

[yanqi liu]

yes，sir，may be use

n=10
xrange = 11:13;
yrange = 11:13;

for k = 1:(n-1)
    image1 = imread(sprintf('image_%.4d.jpg',k));
    image2 = imread(sprintf('image_%.4d.tiff',k+1));
    
    %I1 = double(image1(xrange,yrange));
    %I2 = double(image2(xrange,yrange));
    
    R = imsubtract(I2,I1);
    R = R(xrange,yrange);
    
    c(k,:) = I2(:)-I1(:);
end

[DGM]

That's obviously not going to work if I2,I1 don't exist.

If the goal is to avoid the issues caused by data scaling, I usually recommend to simply use im2double(). I guess I didn't bother.

n=10
xrange = 11:13;
yrange = 11:13;
for k = 1:(n-1)
    image1 = imread(sprintf('image_%.4d.jpg',k));
    image2 = imread(sprintf('image_%.4d.tiff',k+1));
    
    I1 = im2double(image1(xrange,yrange));
    I2 = im2double(image2(xrange,yrange));
    
    c(k,:) = I2(:)-I1(:);
end

or you could use imsubtract()

n=10
xrange = 11:13;
yrange = 11:13;
for k = 1:(n-1)
    image1 = imread(sprintf('image_%.4d.jpg',k));
    image2 = imread(sprintf('image_%.4d.tiff',k+1));
    
    D = imsubtract(image2(xrange,yrange),image1(xrange,yrange));
    
    c(k,:) = D(:);
end

If the image classes differ, that will at least cast and scale them so that they're compatible, but the output scale and class will vary with whatever the inputs are. You could cast D with im2double() or the like, but it seems unnecessary when the output class can be asserted to begin with.

That said, OP hasn't mentioned what the problem actually is.

[HEMRAJ PATEL]

I got it. Thanks