Re: [vmtk-users] Centerlines of regular geometry

[Luca A. <luc...@or...>]

Dear Regine,
 I'm pretty sure the issue is indeed the fact that the Voronoi diagram of a perfect 
cylinder is not a surface but a line, and the algorithm is affected by this. 

In order to generate your centerlines, you need to "shake" things up a bit.
There are a few ways of doing it.

A first try could be to use vmtksurfacedecimation, which should make the surface 
mesh asymmetric.

If that doesn't work,

1. Open the surface file in Paraview
2. Apply the "Random Vectors" filter
3. Apply the "Warp by Vector" filter (if you see your surface garbled, lower
the Maximum Speed parameter in Random Vectors until you only see tiny perturbations
of the surface)
4. Apply the "Extract surface" filter
5. Save the surface as a vtp file.

Now run the surface through vmtkcenterlines.

Hope this helps


Luca



On Apr 17, 2012, at 12:56 PM, Francesco Iannaccone wrote:

> Not sure if this can help. I dont know if this depend on the regular  
> geometry but maybe on the quality of the mesh that is not properly  
> defined. i faced a similar problem . Check if the stl has interseting  
> or duplicated elements , degenerated elements with areas equals to 0.
> 
> ciao
> francesco
> 
> Quoting "Regine Schmidt" <ric...@we...>:
> 
> Dear all,
> 
> I want to calculate the centerlines of a geometry I constructed by
> connecting tubes of different diameters (kind of an ideal
> bifurcation). Unfortunately, I always get the error message: "Unable
> to factor linear system".
> If I am calculating the centerlines for a real vessel-geometry (which
> I extracted from CT data) I have no problem at all.
> I guess this problem appears due to my regular geometry.
>  I already tried Luca's advice from an older thread and added
> "-usetetgen 1", but then the program is just crashing.
> 
>  Is there any way to avoid this and to extract the centerlines?
> 
> Thanks a million for all your helps and ideas,
> Regine
> 
> 	Ihr WEB.DE Postfach immer dabei: die kostenlose WEB.DE Mail App für
> iPhone und Android.
> HTTPS://PRODUKTE.WEB.DE/FREEMAIL_MOBILE_STARTSEITE/[1]
> 
> 
> 
> Links:
> ------
> [1] https://produkte.web.de/freemail_mobile_startseite/
> 
> 
> 
> 
> ------------------------------------------------------------------------------
> Better than sec? Nothing is better than sec when it comes to
> monitoring Big Data applications. Try Boundary one-second 
> resolution app monitoring today. Free.
> http://p.sf.net/sfu/Boundary-dev2dev
> _______________________________________________
> vmtk-users mailing list
> vmt...@li...
> https://lists.sourceforge.net/lists/listinfo/vmtk-users