<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.2900.2668" name=GENERATOR></HEAD>
<BODY>
<DIV><SPAN class=229372312-12082005><FONT face=Arial color=#0000ff
size=2>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.</FONT></SPAN></DIV>
<DIV><SPAN class=229372312-12082005><FONT face=Arial color=#0000ff
size=2></FONT></SPAN> </DIV>
<DIV><SPAN class=229372312-12082005><FONT face=Arial color=#0000ff size=2>I
think in general, the DiscreteGaussian is more accurate than the
RecursiveGaussian. I think this is particularly true for small
sigma.</FONT></SPAN></DIV>
<DIV><SPAN class=229372312-12082005><FONT face=Arial color=#0000ff
size=2></FONT></SPAN> </DIV>
<DIV><SPAN class=229372312-12082005><FONT face=Arial color=#0000ff size=2>For
large sigma, the RecursiveGaussian is much much much faster. I think the
RecursiveGaussian is constant time wrt sigma.</FONT></SPAN></DIV>
<DIV><SPAN class=229372312-12082005><FONT face=Arial color=#0000ff
size=2></FONT></SPAN> </DIV>
<DIV><SPAN class=229372312-12082005><FONT face=Arial color=#0000ff size=2>The
DiscreteGaussian supports streaming, so you can smoothing just a piece of an
image whereas the RecursiveGaussian has to process the entire
image.</FONT></SPAN></DIV>
<DIV><SPAN class=229372312-12082005><FONT face=Arial color=#0000ff
size=2></FONT></SPAN> </DIV>
<DIV><SPAN class=229372312-12082005><FONT face=Arial color=#0000ff
size=2>Jim</FONT></SPAN></DIV>
<BLOCKQUOTE>
<DIV class=OutlookMessageHeader dir=ltr align=left><FONT face=Tahoma
size=2>-----Original Message-----<BR><B>From:</B>
insight-users-bounces+millerjv=crd.ge.com@itk.org
[mailto:insight-users-bounces+millerjv=crd.ge.com@itk.org]<B>On Behalf Of
</B>Renaud Isabelle<BR><B>Sent:</B> Wednesday, August 10, 2005 4:13
PM<BR><B>To:</B> insight-users@itk.org<BR><B>Subject:</B> [Insight-users]
quick question: discrete or recursive gaussianfilter<BR><BR></FONT></DIV>
<DIV>Hi,</DIV>
<DIV> </DIV>
<DIV>Now, thanks for all your replies, I think I caught the way to use
itkDiscreteGaussianImageFilter and itkRecursiveGaussianImageFilter. </DIV>
<DIV> </DIV>
<DIV>But, something is still not obvious for me: </DIV>
<DIV> </DIV>
<DIV>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?</DIV>
<DIV> </DIV>
<DIV>Isabelle</DIV>
<P>
<HR SIZE=1>
<B><FONT color=#ff0000>Appel audio GRATUIT</FONT> partout dans le monde</B>
avec le nouveau Yahoo! Messenger<BR><A
href="http://us.rd.yahoo.com/messenger/mail_taglines/default/*http://fr.messenger.yahoo.com">Téléchargez
le ici !</A> </BLOCKQUOTE></BODY></HTML>