[SciPy-user] regional maxima

[Emanuele Zattin]

Hello everybody,
i am considering to port to Python an application in wrote in Matlab.
My only concern right now is that i need something similar to the function
"imregionalmax":

http://www.mathworks.com/access/helpdesk/help/toolbox/images/index.html?/access/helpdesk/help/toolbox/images/imregionalmax.html&http://www.google.dk/search?q=imregionalmax&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a

My googling so far has taken me to find out pymorphpro (
http://www.mmorph.com/pymorphpro/) which does exactly what i want but
doesn't seem to be free and works with numeric and not numpy.

I can also see that scipy offers some morphology functions but nothing to
spot regional maxima. I wonder if it is possible to achieve this using the
available primitives (erosion, dilation, ...) and i think i found the answer
(or part of it) here:

http://www.qgar.org/doc/QgarLib/html/classqgar_1_1RegionalMaxBinaryImage.html

where this is defined:

*RMAX(f) = f - R(f,E(f))*, where *f* is the given image, *E* its erosion,
and *R(a,b)* the reconstruction by dilatation of *b* using *a*.


Anyhow... my experiments so far were not successful so that's why i'm
writing here. Any suggestion?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20070423/2c40cc0f/attachment.html>