[Insight-users] Using Bacterial Colonies for segmentation : Morphogenesis Application

Luis Ibanez luis . ibanez at kitware . com
Thu, 11 Dec 2003 19:15:04 -0500


Hi Yasser,

Using VTK for visualization is a natural complement to
the image analysis done with ITK.

There is certainly an overhead in the process of converting
and ITK mesh into a VTK one, but for the case of deformable
meshes this overhead is negligeable compared to the burden
of computing the actual deformations in the itk::Mesh.

You may want to look at the example for converting an
itk::Mesh in to a vtkUnstructuredGrid in the file

     InsightApplications/Auxiliary/vtk/itk2vtk.cxx

You can actually simplify this code a lot if you are only
interested in a wireframe representation. (which is the
case for the Morphogenesis example).

Notice also that you don't need to render the mesh at
every iteration of the bacterial colony. You may setup
something like refresing the rendering every 10 iterations
for example.   The same thing is valid for observing the
evolution of a LevelSet...


Please let  us know if you have further questions.



Thanks


     Luis


---------------------------
Yasser Bashir wrote:

>Great. This works perfectly. Thank you Luis.
>
>Questions regarding visualization of the ITKMesh:
>
>If i set up a pipeline between ITK and VTK and perform the visualization in
>VTK, isnt that extremely (un-usably) inefficient?  In general, what is the
>most optimum way of viewing an ITKMesh that is forming or deforming over
>time.  At this point i am just interested in the wireframe, not the
>surfaces.
>
>It would be nice to be able to harness the power of VTK but moving and back
>and forth between the UnstructuredGrid and the ITKMesh will incur a HUGE
>overhead or so i believe.
>
>Regards.
>
>
>
>  
>