[Insight-users] Parabolic dilatation correct radius

[Richard Beare]

Hi,
I haven't run anything, but I've had a look at your image and the test
program. The parabolic classes assume a binary input image with values
0 or 1, not 0 or 255. Scaling the input will scale the size of the
radius by the square root of the intensity scale. This is the basis of
the spatially variant operations I mentioned in the earlier message.
Also, be sure to update with the version from github, as there have
been minor changes to the padding for openings and closings and there
is support for ITK4.

On Fri, Dec 16, 2011 at 8:23 AM, Simon ESNEAULT
<simon.esneault at gmail.com> wrote:
> Hi All,
>
> Following that conversation on the mailing list :
> http://www.itk.org/pipermail/insight-users/2009-June/030778.html
> I was convinced that the fastest dilatation filter implementation in ITK is
> the multi-threaded parabolic Morphology
> (http://mima2.jouy.inra.fr/darcs/contrib-itk/parabolicMorphology/).
> And my local test confirmed it all.
>
> But I'm having trouble to set the correct radius in the class
> itk::BinaryDilateParaImageFilter. With the same radius, the result
> highly differs between BinaryDilateParaImageFilter and
> the ground-truth BinaryDilateImageFilter. What is the correct radius that
> should be set in the Parabolic filter in order to get the same result than
> the classical binary dilate ?
>
> Here -> http://esneault.free.fr/itk/TestParabolic.zip <- is a minimal c++
> project and a dataset that reproduce the problem
>
> Step to build and run the project :
> uncompress
> cmake .
> make
> ./TestParabolic
>
> And you should have some result file written. etc
>
> Thanks
>
> Simon
>
> --
> ------------------------------------------------------------------
> Simon Esneault
> 13 rue Vasselot
> 35000 Rennes, France
> Mail : simon.esneault at gmail.com
> ------------------------------------------------------------------
>
>
> _____________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Kitware offers ITK Training Courses, for more information visit:
> http://www.kitware.com/products/protraining.html
>
> Please keep messages on-topic and check the ITK FAQ at:
> http://www.itk.org/Wiki/ITK_FAQ
>
> Follow this link to subscribe/unsubscribe:
> http://www.itk.org/mailman/listinfo/insight-users
>