[Insight-users] Fast Marching for arbitrary interfaces

[Amardeep Singh]

Dear Luis

Thank you very much for your answer!
I have also thought about using the ShapeDetectionLevelSetImageFilter. 
But in my application, I'm afraid, I am
interested in the time-crossing map, initialized from an arbitrary 
front. I need to
use level-sets with an exact narrow-band (for post-processing). So, the 
chamfer distance function which
is currently used is not exact enough in my case. Instead, I would like 
to use fast marching for reinitialization.
Therefore I would like to repeat my question:
Is it correct that the itkIsoContourDistanceImageFilter is a good 
starting point?
What is the use of the "FastMarchingExtensionImageFilter"?

Thanks a lot!

Best regards
Amardeep

Luis Ibanez wrote:
>
> Hi Amardeep
>
>
> The implementation of FastMarching in ITK is initialized using
> seed points.
>
> If you already have an initial front (as opposed to a set of
> seed points) then you can use the
>
>               ShapeDetectionLevelSetImageFilter
>
> and set the curvature weight to zero. In those conditions this
> filter will preform a front propagation similar to Fast Marching.
>
>
>
>     Regards,
>
>
>        Luis
>
>
>
>
> -------------------
> Amardeep Singh wrote:
>> Dear ITK-users
>>
>> In the current implementation of the fast marching filter, the 
>> initial surface that is to be propagated
>> has to be a binary interface (Please, correct me if I'm wrong).
>> I am wondering if there is a way to apply the fast marching filter in 
>> ITK to arbitrary inital fronts as it is described
>> by Sethian in his paper "A fast marching level set method for 
>> monotonically
>> advancing fronts" (PNAS, February 1996). If such a functionality does 
>> not exist at the moment:
>> Would you agree that the itkIsoContourDistanceImageFilter might be a 
>> good starting point to implement the initial
>> calculation of the signed distances to some arbitrary isosurface? 
>> These distances might then be used for the fast marching propagation 
>> as described by Sethian.
>>
>> Thanks a lot!
>>
>> Best regards
>> Amardeep
>> _____________________________________
>> 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
>>
>