[Insight-users] negative propagation in GAC

[Luis Ibanez]

Hi Xabier,

Using negative values for the weights is probably not a good idea.

You may find better results with increasing the Advection weight,
assuming that your level set is already quite close to the edges
of the object that you want to segment.

Please note that when you play with this weights you want to
think on their relative ratios, and to go to values as high as
hundreds.


      Regards,


         Luis


--------------------------
Xabier Artaechevarria wrote:
> Hi all,
> 
> I am trying to use Geodesic Active Contours to refine an initial
> segmentation. I am trying different parameters for propagation,
> curvature and advection scaling. I tried with negative propagation
> scaling but I am not able to understand its effect.
> 
> Does it make sense at all to have negative propagation scaling and, if
> so, what should the effect be?
> 
> Thank you in advance.
> Regards,
> Xabier
> _____________________________________
> 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
>