<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<META NAME="Generator" CONTENT="MS Exchange Server version 6.5.6944.0">
<TITLE>Do RecursiveHessian or RecursiveGaussian require isotropic voxels?</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->
<P><FONT SIZE=2>Dear Itk users,<BR>
<BR>
I am playing with vesselness measure based on eigen values of Hessian matrices. I have noticed a quite odd behaviour and I would like to know if it is normal.<BR>
<BR>
Basically the eigenvalues change "a lot" whether or not my input image has or not isotropic voxels. The "a lot" changing is related to the tubes or blobs detection with the Hessian+eigenValues principle, i.e. roughly lambda1 = lambda2 = lambda3 for a blob and lambda1 = 0 and lambda1 << 0 and lambda3 << 0 for a tube.<BR>
<BR>
In my case with a blob and non isotropic image: lambda1 close to 0 and lambda2/3 << 0. -> odd(?)<BR>
In my case with a blob and isotropic image: lambda1 = lambda2 = lambda3 (roughly) -> expected<BR>
<BR>
Knowing that the hessian is computed with the RecursiveGaussianImageFilter that is based on the RecursiveGaussian filter, I wonder whether the pb is here.<BR>
<BR>
A note: my sigma for the detection is close to the radius of the blob, and it is defined in physical units. I expect that the recursive gaussian filter uses sigma in physical units ( which looks to be the case, looking at the distinction variance/sigma, isn'it?).<BR>
<BR>
Any suggestion? Thanks!<BR>
<BR>
Best Regards,<BR>
<BR>
Jerome Schmid<BR>
<BR>
-----------------------------------<BR>
Jerome SCHMID<BR>
Project Manager/ Engineer<BR>
Augmented and Virtual Reality<BR>
MIS Centre<BR>
Prince of Wales Hospital<BR>
Chinese University Of Hong-Kong<BR>
-----------------------------------<BR>
<BR>
</FONT>
</P>
</BODY>
</HTML>