[Insight-users] Model-to-image Gaussian fitting using optimisation

[Ali -]

Hi,

I have asked this before, but the good answer that Luis provided was not what I was looking for.

I am trying to fit a set of particle-images to a Gaussian model, here is a sample of the images: http://img513.imageshack.us/my.php?image=0077vb0.png

Assume that after some pre-processing, including a binary thresholding, we have each particle-image labeled in a separate region. The question is, how is it possible to fit a Gaussian model to these labeled regions in order to extract their positions and more.

- Would the itk spatial objects be of use?

- Another way of implementing this fitting is to use optimisation approaches, is there any existing filters in itk to do this? If yes, how?

_________________________________________________________________
Celeb spotting – Play CelebMashup and win cool prizes
https://www.celebmashup.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://public.kitware.com/pipermail/insight-users/attachments/20071010/a6db5892/attachment.htm