Question

I haven’t fully understood how qtdecomp works…

I = [1    1    1    1    2    3    6    6
     1    1    2    1    4    5    6    8
     1    1    1    1   10   15    7    7
     1    1    1    1   20   25    7    7
    20   22   20   22    1    2    3    4
    20   22   22   20    5    6    7    8
    20   22   20   20    9   10   11   12
    22   22   20   20   13   14   15   16];

S = qtdecomp(I,2);
disp(full(S));

The results of this are:

   4     0     0     0     1     1     2     0
   0     0     0     0     1     1     0     0
   0     0     0     0     1     1     2     0
   0     0     0     0     1     1     0     0
   4     0     0     0     1     1     1     1
   0     0     0     0     1     1     1     1
   0     0     0     0     1     1     1     1
   0     0     0     0     1     1     1     1

in the left bottom 4*4 matrix, maximum value (22) of the block elements minus the minimum value (20) is 2, so when decomposing this part, it will left as is.

When I do this on a uint8 matrix:

I = uint8([...
     1    1    1    1    2    3    6    6
     1    1    2    1    4    5    6    8
     1    1    1    1   10   15    7    7
     1    1    1    1   20   25    7    7
    20   22   20   22    1    2    3    4
    20   22   22   20    5    6    7    8
    20   22   20   20    9   10   11   12
    22   22   20   20   13   14   15   16]);

S = qtdecomp(I,2/255);
disp(full(S));

the answer is just like before. But when I change S to this:

S = qtdecomp(I,1.9/255);

The answer is

 4     0     0     0     1     1     2     0
 0     0     0     0     1     1     0     0
 0     0     0     0     1     1     2     0
 0     0     0     0     1     1     0     0
 4     0     0     0     1     1     1     1
 0     0     0     0     1     1     1     1
 0     0     0     0     1     1     1     1
 0     0     0     0     1     1     1     1

I suppose the left bottom 4*4 matrix should decompose, but why doesn’t it?

Answer 1

What matlab does here is when I is uint8 it multiples the threshold by 255 and rounds it, so 1.9/255 is evaluated to 2.

You can see this by opening the source code for qtdecomp (by pressing ctrl+D) or here. There’s an if/elseif near the end of the file (params{1} = round(255 * params{1});).

You should be able to use S = qtdecomp(I,1/255); to get the result you are looking for.