[Insight-users] Segmentation filters

[Luis Ibanez]

Hi Eduardo,

Yes,

You could replace the FastMarching filter with a
DanielssonDistanceMap filter.

Just create a binary image that is set to zeros everywhere,
and only set to ones the pixels corresponding to the seed
points that are now being passed to the FastMarching filter.
Then use that image as input to the Danielsson distance map
filter.


Notice however that:


    A) In some of the examples, the seed points are set
       to initial negative values, in order to be able to
       define the initial level set as the iso-value == 0.


    B) The FastMarching filter is a lot faster than the
       Danielson Filter for the purpose of computing that
       distance map


    C) If you really want to stick to a distance filter,
       then you may want to try the SignedMaurer distance
       filter that is equivalent to the Danielson filter,
       but it executes a lot faster.



    Regards,


        Luis



--------------------
Eduardo - LNCC wrote:
> Hi,
> 
> 
> Can I replace the itkFastMarchingImageFilter class by class 
> itkDanielssonDistanceMapImageFilter without have problems in the 
> segmentation filters below?
> 
>     * itkShapeDetectionLevelSetImageFilter
>     * itkGeodesicActiveContourLeveSetImageFilter
>     * itkThresholdSegmentationLevelSetImageFilter
>     * itkCannySegmentationLevelSetImageFilter
>     * itkLaplacianSegmentationLevelSetImageFilter
> 
> 
> 
> Sorry for my english
> 
> -- 
> Atenciosamente,
> 
> Eduardo Camargo
> Analista de Sistemas
> HeMoLab - http://hemolab.lncc.br/equipe/desenv/eduardo.htm
> Laboratório Nacional de Computação Científica - LNCC
> 
> 
> ------------------------------------------------------------------------
> 
> _____________________________________
> Powered by www.kitware.com
> 
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.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