[Insight-users] quick question: discrete or recursive gaussianfilter

[Miller, James V (Research)]

Someone once did an analysis of the two techniques.  Not sure where that is.  I google search may find it in the mail archives. I think they compared accuracy vs tims.  We should try to track that down and put it on the ITK Wiki or maybe a short paper in the Insight Journal.
 
I think in general, the DiscreteGaussian is more accurate than the RecursiveGaussian. I think this is particularly true for small sigma.
 
For large sigma, the RecursiveGaussian is much much much faster.  I think the RecursiveGaussian is constant time wrt sigma.
 
The DiscreteGaussian supports streaming, so you can smoothing just a piece of an image whereas the RecursiveGaussian has to process the entire image.
 
Jim

-----Original Message-----
From: insight-users-bounces+millerjv=crd.ge.com at itk.org [mailto:insight-users-bounces+millerjv=crd.ge.com at itk.org]On Behalf Of Renaud Isabelle
Sent: Wednesday, August 10, 2005 4:13 PM
To: insight-users at itk.org
Subject: [Insight-users] quick question: discrete or recursive gaussianfilter


Hi,
 
Now, thanks for all your replies, I think I caught the way to use itkDiscreteGaussianImageFilter and itkRecursiveGaussianImageFilter. 
 
But, something is still not obvious for me: 
 
for, small recursive sigma value and/or small discrete variance value, could someone tell me what is better to use between itkRecursiveGaussianImageFilter and itkDiscreteGaussianImageFilter, in terms of time computation and precision results?
 
Isabelle



  _____  

Appel audio GRATUIT partout dans le monde avec le nouveau Yahoo! Messenger
Téléchargez  <http://us.rd.yahoo.com/messenger/mail_taglines/default/*http://fr.messenger.yahoo.com> le ici ! 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://public.kitware.com/pipermail/insight-users/attachments/20050812/70d05fcf/attachment.htm