[Insight-users] Segmentation-Surface Reconstruction- Questions

[Terry Yoo]

A reference for this problem is already online.  See:

http://www.itk.org/HTML/AntiAliasBinaryImageFilterExample.html

Good luck, and Enjoy ITK!

-t.

=================

Bill Lorensen wrote:

> You can use itk::AntiAliasBinaryImageFilter . It creates a levelset from 
> a binary mask. You can use vtkMarchingCubes to extract the 0 level set.
>  
> Bill
> 
>  
> On 6/23/06, *Karthik Krishnan* <Karthik.Krishnan at kitware.com 
> <mailto:Karthik.Krishnan at kitware.com>> wrote:
> 
>     If your purpose is to produce a smooth mesh / contour from a binary
>     segmentation, you could in principle do this both on the image domain
>     and on the mesh domain.
> 
>     1. Apply the Anti-alias filter in ITK and then run marching cubes on
>     it.  The anti-aliasing filter will smooth jagged edges, by introducing
>     some continuity. So the output of your marching cubes won't look
>     jagged.
> 
>     2. Do some polygon anti-aliasing on the mesh. You will have to go to VTK
>     for that.
> 
>     HTH
>     -karthik
> 
>     platon galatis wrote:
> 
>      > Hello,
>      >
>      > Im  a mechanical engineer and my diploma thesis subject is about
>      > segmenting a series of human 2D ct scans, in order to produce a 3D
>      > surface mesh suitable to be exported to CAD-CAE systems for FEA.
>      >
>      > Im trying to conclude on my strategy but i have some questions.
>      >
>      > I've read a lot about Marching cubes,Marching tetrahedra and their
>      > variations. All these algorithms speak for values at nodes of a
>      > regular 3D grid. But what if the input images are binary
>     (black/white)
>      > as a result of a segmentation algorithm from itk?
>      >
>      > If i said that out = 1 ,in=-1 and the boundary=0 then the linear
>      > interpolation would always give the middle of the intersected  edge.
>      > Am i mising something here?
>      >
>      > Can i get results giving as input a thresholded set and by specifying
>      > the isocontour value as the mean value of a set of  points sampled
>      > from the desired boundary (skipping segmentation)?
>      >
>      > I've found the Deformable models method interesting . Would it be a
>      > good practice to produce a binary 3D image with some segmentation
>      > method, pass it to the itk::BinaryMask3DMeshSource for a rough
>     initial
>      > mesh and then use the itk::DeformableMeshFilter to refine it?
>      >
>      >
>      >
>      > Thank you for reading this.
>      >
>      > Any help would be appreciated.
>      >
>      > Sincerely,
>      >
>      > Plato Galatis.
>      >
>      > _________________________________________________________________
>      > Express yourself instantly with MSN Messenger! Download today it's
>      > FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
>      >
>      > _______________________________________________
>      > Insight-users mailing list
>      > Insight-users at itk.org <mailto:Insight-users at itk.org>
>      > http://www.itk.org/mailman/listinfo/insight-users
>      >
>     _______________________________________________
>     Insight-users mailing list
>     Insight-users at itk.org <mailto:Insight-users at itk.org>
>     http://www.itk.org/mailman/listinfo/insight-users
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users


-- 
-------------------------------------------------------------------
Terry S. Yoo, Ph.D.
National Library of Medicine          National Institutes of Health
High Performance Computing and Communications       yoo at nlm.nih.gov
-------------------------------------------------------------------