[ITK-users] Modifying itk-Functions (Mattes MI to Normalized MI)

[O Hamo]

Hi,
The
implementation of Mattes Mutual Information is much faster in contrary to
Normalized MI. Like I understood, this is due to the fact how the histograms
respectively the probability distributions are calculated / estimated. Like
Louis suggested[1] Mattes MI could be modified to obtain NMI. I tried that by
changing itkMattesMutualInformationImageToImageMetric (ITK version 4.3.1) at lines
583-617 (and accordingly 862-944) to the code below.

Because I
don’t want to mess up the ITK library on my system, I’d like to ask if someone
could explain, how to test the code and integrate it into the library.
Kid Regards,Omar Hamo

 

[1] http://public.kitware.com/pipermail/insight-users/2006-May/017911.html

 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/insight-users/attachments/20140616/36d9936c/attachment.html>