Need help in creating mask?

I have matix of size 7071X7801 which contains the pixel values of image (this is Landsat 5 image) and there are 0 values at edges of image

So, I want to select 9 fixed matrices to create a mask based on pixed values by excluding 0 values.

Right now, I am reading one small matrix 11X11 pixel size like:

[A, R]      = geotiffread('...tif'); %reading .tif image
%(200,1414) this center for below matrix
i0  = 200;
j0  = 1414; %center of small 11X11 matrix
[row, col]  = size(A);
A_small = A((i0-5):(i0+5),(j0-5):(j0+5));

Now, I want to automate this process by shifting this matrix 1000 pixel and select new 11X11 size matrix. So, can you guys give me any thoughts, how can I proceed from here?

Answer 1

[Walter Roberson]

blockproc() need not return an array the same size as the nominal input window. It would be valid, for example, to tell it you want a 1000 x 1000 window, but then when you get the data, extract only the first 11 x 11 pixels from it.

But if it is just a matter of removing the 0 pixels, then you can

horzmask = any(TheImage, 1);
hfirst = find(horzmask, 1, 'first');
hlast = find(horzmask, 1, 'last');
vertmask = any(TheImage, 2);
vfirst = find(vertmask, 1, 'first');
vlast = find(vertmask, 1, 'last');

subset = TheImage(vfirst:vlast, hfirst:hlast);