How to save multiple matrices in each iteration of a for loop into a structure?

A = imread('Ish.JPG');
A = im2single(squeeze(mean(A,3)));
A = im2double(A);
[pyrRef,pind] = buildLpyr(A,'auto');
nLevels = 7;
kernelx = [0.0, 0.0, 0.0; 0.5, 0.0, -0.5; 0.0, 0.0, 0.0];
kernely = [0.0, 0.5, 0.0; 0.0, 0.0, 0.0; 0.0, -0.5, 0.0];
for k = 1:nLevels
    subband = pyrBand(pyrRef, pind, k);
    rx(k) = conv2(subband(k),kernelx);
    ry(k) = conv2(subband(k),kernely);
end

Here, for each 'k' (k = 1:7) there will be 3 matrices (subband, rx, ry). So, I want to save this in a structure of diemnsion (7 X 3). How do I do that within this loop?

Answer 1

[Matt J]

clear S
for k = nLevels:-1:1
    S(k).subband = pyrBand(pyrRef, pind, k);
    S(k).rx = conv2(subband(k),kernelx);
    S(k).ry = conv2(subband(k),kernely);
end

[Anisia Anil]

Thank you.

[Anisia Anil]

Value = S(k).subband.*S(k-1).subband + S(k).rx.*S(k-1).rx + S(k).ry.*S(k-1).ry;

If I have to compute this value for each iteration, how do I do that? (Values at S(k-1) are previously defined.) The matrices "subband", "rx", and "ry" have different dimensions. So it is showing "Arrays have incompatible sizes for this operation" when try to implement this equation. Also, the matrix dimensions will be different for each iterations. So, I need to generalize this matrix addition with different dimensions. How do I do that? Can you please help?