On 15-Jul-09, at 6:40 PM, Vebjorn Ljosa wrote: > On Jul 15, 2009, at 17:51, David Warde-Farley wrote: > >> Suppose I have an 8-bit integer 2-d array, X, and I want a 256x256 >> matrix that tells me how many times a pixel value v was horizontally >> (along second dimension) adjacent to a pixel value b > >> Is there an efficient way to do such a thing with numpy operations? I >> can't think of a way. > > I wrote the following cooccurrence matrix function as part of > computing Haralick texture features. It's not exactly what you ask > for, but close enough that it may give you some ideas.
The Haralick-style features are exactly what I was thinking about. Thanks for the pointer to that code. :) Just curious - I noticed in the comments that you (or someone) said there's a problem in the original paper's definition. Do you have any idea which feature that concerned? David _______________________________________________ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion
