[Insight-users] Preprocessing intensity images

[Arne Hansen]

Hello.
I have a set of 20 MRI images. I am developing a segmentation method which
is based on intensity distributions off subsamples of all the images. The
problem is that the intensity distribution varies across the images, so I
need to somehow process it so that it becomes standard. I have thought about
just substracting the mean and divide with the standard deviation for each
image to get standardized data, but I cannot see if this will affect the
probability distribution.

So my question is, how are images normally preprocessed in order to have
corresponding intensities across all images, so that an intensity based
algorithm can work on all the images?

I hope you know what I mean, otherwise I will gladly try to explain it
again.

Looking forward to hear from you.

Best of regards
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://public.kitware.com/pipermail/insight-users/attachments/20070508/bc71b135/attachment.html