[Insight-users] anisotropic diffusion

[Luis Ibanez]

Hi Bruno,

Yes, indeed these methods allow for diffusion in the
direction parallel to the edges.

Note however, that this claim has to be qualified as:

         They do more diffusion in the direction
         parallel to edges than in the direction
         perpendicular to the edges.


The conductance value is further modulated / computed
inside the code, according to the direction of the gradients.


See for example, lines 114-156 of the file:

           Insight/Code/BasicFilters/
               itkGradientNDAnisotropicDiffusionFunction.txx


    Regards,


           Luis


---------------------------------------------------------------------------------------
On Fri, Dec 4, 2009 at 8:55 AM, Bruno Arbter
<bruno.arbter at isv.uni-stuttgart.de> wrote:
> Hi,
>
> I started using nonlinear diffusion filters in 3D. As far as I understood,
> the classic Perona-Malik implementation
> (itkGradientNDAnisotropicDiffusionFunction) uses a scalar value for the
> conductance and is strictly speaking isotropic. Further I found the
> level-set-based initkCurvatureNDAnisotropicDiffusionFunction.
>
> But are there any anisotropic (edge-enhancing) diffusion filters that are
> able to stop diffusion perpendicular to edges but allows diffusion parallel
> to edges?
>
> In itkAnisotropicDiffusionFunction.h I found the hint that "several
> variations on the scheme [...] are implemented in Itk"...
>
> Thanks,
> Bruno
>
> _____________________________________
> 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
>