[Insight-users] Level Set Segmentation for 3D images

Luis Ibanez luis.ibanez at kitware.com
Sat Jun 19 09:31:52 EDT 2004


Hi Thomas,

All the LeveSet filters in ITK are implemented
for N-Dimensions. You can easily instantiate
a FastMarching filter for 4D images.

You will find demo applications on the use of
LevelSets for 3D images in InsightApplications.

For example:

   InsightApplications/
           FastMarchingLevelSet
           LevelSetSegmentation
           LevelSetSurfaceProcessing
           ShapeDetectionLevelSet
           ThresholdSegmentationLevelSetFltkGui

The reason why examples in the SoftwareGuide are
limited to 2D is simply that 2D images are more
illustrative in the paper version of the book.
Should we have presented 3D images, it wasn't clear
what the correct and informative representation of
the Level Set should be : Surface renderings ?,
Volume renderings ?, Orthogonal slices ?...



There are many other segmentation algorithms in ITK.
You will find an overview in the Tutorial sessions

      http://www.itk.org/HTML/Tutorials.htm

in particular in

http://www.itk.org/CourseWare/Training/SegmentationMethodsOverview.pdf


Among those methods you will find for example:

- Region Growing algorithms  (N-Dimensional)
- Fuzzy connectedness        (N-Dimensional)
- Statistical classifiers    (N-Dimensional)
- Watersheds                 (N-Dimensional)
- Deformable Models          (3-D)
- Cellular Algorithms        (2-D / 3-D)
- Atlas-based segmentation   (N-D)
- Model-based segmentation   (N-D)


You are strongly encouraged to read the Chapter
on image segmentation of the ITK SoftwareGuide

     http://www.itk.org/ItkSoftwareGuide.pdf

Chapter 9, pdf-page 341.



   Regards,



       Luis



----------------------------
Thomas - Kuiran Chen wrote:

> Hi Luis,
> 
> The ITK documentation shows that it has specific implementation for level-set segmentation, algorithms of which includes: 
> - fast marching, 
> - shape detection, 
> - geodesic active contours, 
> - threshold level set, 
> - canny edge level set, and 
> - laplacian level set
> 
> Apparently these will work for 2D images, but would  they support 3D images (i.e., a CT volume data)?  The example shows only 2D images though.
> 
> Also, for 3D images segmentation, does ITK have any other specific algorithm implemented?
> 
> Thanks very much!
> 
> Thomas Kuiran Chen 
> 
> 
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users
> 





More information about the Insight-users mailing list