[Octave-bug-tracker] [bug #39314] convn(x, y, "valid") produces results that are different from Matlab

[Marco cecchetti]

Follow-up Comment #5, bug #39314 (project octave):


Indeed the problem can be reproduced with integer values too and is present
only for arrays with more than 2 dimensions. The bug is inside the convolve_nd
function (liboctave/numeric/oct-convn.cc) and exactly inside the following
loops:


for (octave_idx_type ja = 0; ja < na - nb + 1; ja++)
  for (octave_idx_type jb = 0; jb < nb; jb++)
    convolve_nd<T, R> (a + ma*(ja + jb), ad, acd, b + mb*jb, bd, bcd,
                       c + ldc*ja, ccd, nd-1, inner);


The 'b + mb*jb' argument should be 'b + mb*(nb-jb-1)' in fact convolution is
always a sum of products of this type: a(u1, u2, u3) * b(x1-u1, x2-u2,
x3-u3).

The following octave example is straightforward:


octave:1> v = [1,2]
v =

   1   2

octave:2> vvv = reshape(v,1,1,2)
vvv =

ans(:,:,1) =  1
ans(:,:,2) =  2

octave:3> convn(vvv, vvv, 'valid')
ans =  5
octave:4> convn(vvv, vvv)
ans =

ans(:,:,1) =  1
ans(:,:,2) =  4
ans(:,:,3) =  4


The correct result for convn(vvv, vvv, 'valid') should be vvvv(1,1,2) that is
4 instead we get the same result as we computed dot(v, v).

I attached a patch with the correct implementation. I tested it with the zzz
and hhh as input and it works fine.


(file #28829)
    _______________________________________________________

Additional Item Attachment:

File name: convn.diff                     Size:1 KB


    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?39314>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/