[Insight-users] confirmation of isotropic spacing requirement for levelset filters in ITK

[Luis Ibanez]

Hi Charl,


The FastMarchingLevelSetImageFilter will correctly take spacing into
account. We fixed when we were writing plugins for VolView.


For all the others LevelSetFilters, the UseImageSpacing flag is
available in the base class "FiniteDifferenceImageFilter":

http://www.itk.org/Doxygen/html/classitk_1_1FiniteDifferenceImageFilter.html

for updating the computation of the derivatives.


However, as Karthik pointed out, it remains to be seen whether there
are other places where the spacing should have been taking into account.
(and may have not been used correctly).


If you are short of time, resampling the image, as Karthik suggested
is the safest way to go.


Reviewing these filters for correct support for image spacing
would be an interesting exercise for course projects in an
image processing course. (Hint for Hans)



   Regards,


     Luis



-----------------
Charl Botha wrote:
> Dear all,
> 
> In http://www.itk.org/Doxygen/html/classitk_1_1SegmentationLevelSetImageFilter.html
> under "A NOTE ON DATA SPACING" it says "Input data with anisotropic
> spacing should be resampled to isotropic voxels for use in this
> filter".
> 
> Is this still the case for all filters derived from the
> itkSegmentationLevelSetImageFilter?
> 
> Thanks,
> Charl
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users
>