image processing style (2D kernel) convolutions?

[Kyler Laird]

It appears that Numeric only supports 1D convolutions.  I need to
apply a 2D kernel such as
	0  1  0
	1 -4  1
	0  1  0
to an image.  Is there such a capability in Python/Numeric?  It
appears that numarray has convolve2d(), but I don't have it
available where I'm working.

(I could express the problem better but I have a feeling that
people who know the answer know exactly what I mean.  Tell me if
that's not true.)

Thank you.

--kyler