[Insight-users] Re: FastMarching on OS X

[Luis Ibanez]

Hi Jon,

Thanks a lot for letting us know,

It is a great relief to get rid of the
numerical instabilities on the Mac !

(Even if that happened as a lateral
effect and almost unconciously  :-) ).

The Sigmoid is certainly a better choice
than the negative exponential. The main
advantage come from the capabilitiy of
controling the mapping from the gradient
magnitude to the speed image.

The same change will be done on the
ShapeDetectionLevelSet example soon.


    Luis


-------------------------------------

Jon Harald Kaspersen wrote:
> Hi Luis,
> 
> I noticed that you have made some changes to the FastMarchingLevelSet 
> application in ITK.
> I am happy to announce that these changes bypasses the problem we have 
> had earlier with this application on OS X. I have reduced the problem to 
> be the use of ExpNegativeFilterType. Changing from ExpNegativeFilterType 
> to SigmoidFilterType cures the problem (some numerical overflow) we 
> earlier had on OS X.
> 
> Regards
> Jon
> ========================================================
> Jon Harald Kaspersen Tel: +47 73 59 75 89
> Ph.D. Mechanical Engineering Mob: +47 93 03 65 90
> Senior Scientist Pager +47 96 84 29 94
> SINTEF Unimed - Ultralyd Fax: +47 73 59 78 73
> N-7465 Trondheim
> NORWAY e-mail: Jon.H.Kaspersen@unimed.sintef.no
> WEB: _http://www.us.unimed.sintef.no/_
> ========================================================