[Insight-users] other ways to create the likelihood images for BoykovAlphaExpansionMRFImageFilter

[Roman Grothausmann]

Dear mailing list members,


The Insight Journal contribution 
(http://www.insight-journal.org/browse/publication/306) sounds very 
interesting concerning my image segmentation problems. Still I don't 
really understand the need for the likelihood images.
Since normally I don't know the mean and its variance of the segments in 
my images I'd like to avoid having to specify this to the filter.
I was hoping that one could just give the coordinates of a subpart of 
the input image (containing mostly the information from a wanted 
segment) and let the filter work with that.
But that alone seems to be not sufficient but only an additional option 
(SetIndexContainer).
Is there any way to do this with the help of ITK?

On page 8. of the contribution (end of sec. 2.2) it says: "From such a
sampling of pixel types, it is possible to use non-parametric echniques, 
e.g. Parzen windowing, to construct the likelihood images."
What is meant by that and how could I implement such parametric-free 
usage of the filter?
Is there some filter in ITK that could construct a likelihood image by 
Parzen windowing for BoykovAlphaExpansionMRFImageFilter?

Any help is very much appreciated
Roman

-- 
Roman Grothausmann

Helmholtz-Zentrum Berlin für Materialien und Energie GmbH
Bereich Funktionale Materialien
Institut für angewandte Materialforschung
Hahn-Meitner-Platz 1
D-14109 Berlin	

Tel.: +49-(0)30-8062-2816
Fax.: +49-(0)30-8062-3059

Vorsitzender des Aufsichtsrats: Prof. Dr. Dr. h.c. mult. Joachim Treusch
Stellvertretende Vorsitzende: Dr. Beatrix Vierkorn-Rudolph
Geschäftsführer: Prof. Dr. Anke Rita Kaysser-Pyzalla, Prof. Dr. Dr. h.c. 
Wolfgang Eberhardt, Dr. Ulrich Breuer
Sitz der Gesellschaft: Berlin
Handelsregister: AG Charlottenburg, 89 HRB 5583