[Insight-users] Seeded watershed

Gaetan Lehmann gaetan.lehmann at jouy.inra.fr
Thu May 11 08:04:56 EDT 2006


Hi,

We have developped a watershed filter from markers, which should do what  
you want. It is not published yet, but you can already get the code at  
http://voxel.jouy.inra.fr/darcs/contrib-itk/watershed/. The code is  
heavily tested and should not give problems.

You can also use the minima imposition filter published in the insight  
journal  
(http://www.insight-journal.org/InsightJournalManager/view_reviews.php?back=publications.php%3Fjournalid%3D3%26order%3D%26revision_display%3D&pubid=63),  
and then use the itk watershed (See  
http://voxel.jouy.inra.fr/darcs/contrib-itk/watershed/iwsm.cxx for an  
example). Note that this approach have several drawbacks:
- it require to relabel the labels produced by the watershed filter to get  
the same labels than the one in the marker image
- it doesn't work if you have some markers not separated by background  
pixels
- it gives poor results with plateaus
- it's less efficient

I can't find reference to seeded watershed in matlab. Can you give a link  
to have more details ?

Gaetan

On Thu, 11 May 2006 13:12:30 +0200, van de Giessen  
<M.vandeGiessen at tudelft.nl> wrote:

> Dear all,
>
> For my research I've developed a segmentation algorithm that uses a
> seeded watershed. This algorithm is now in Matlab, but I want to speed
> it up, using ITK.
>
> Such a seeded watershed uses a number of labeled seeds (+/- 15 in my
> case) and performs a watershed. The merging of the basins is as follows:
> The original seeds never merge. As soon as the region within a basin
> adjacent to one of the seeds touches the region of the seed, it is
> merged with the region of the seed. In this way the whole image is
> segmented in the same number of regions as the number of seeds.
>
> I cannot find such a filter. Do I overlook something? Is it possible to
> build such a filter efficiently from other existing filters or do I have
> to derive a filter from the watershed image filter myself?
>
> Thanks,
>
> Martijn
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users



-- 
Gaëtan Lehmann
Biologie du Développement et de la Reproduction
INRA de Jouy-en-Josas (France)
tel: +33 1 34 65 29 66    fax: 01 34 65 29 09
http://voxel.jouy.inra.fr


More information about the Insight-users mailing list