[vtkusers] Filter floaters

Bill Lorensen bill.lorensen at gmail.com
Sun Dec 21 15:24:04 EST 2008


Are you directly rendering the data set (Volume Rendering) or
extracting surfaces? If you are extracting surfaces, have you tried a
region growing algorithm?

If you are thesholding the data to extract surfaces, you could apply
vtk's island removal filter. Or, using itk, there are many ways to
solve the problem, including image morphology, connected component
labeling, and several region growing techniques.

I suspect that the GE system is using some form of grayscale
morpohology to filter floaters.

The book "Morphological Image Analysis"
http://www.amazon.com/Morphological-Image-Analysis-Pierre-Soille/dp/3540429883/ref=pd_bbs_sr_1?ie=UTF8&s=books&qid=1229890626&sr=8-1
is outstanding and many of the algorithms described in the book are
implemented in itk.

Bill

On Sat, Dec 20, 2008 at 6:18 PM, Kevin Osborn <planck at netspeed.com.au> wrote:
> I am rendering a 3D DICOM data set acquired as 1mm CT slices from a CT
> angiogram. The images therefore contain a lot of noise which has a similar
> density value to the tissues of interest. The rendered image contains
> multiple small separate objects (floaters). I have tried to smooth the
> images before rendering however that just makes the floaters look smooth
> :-(. I have used a GE medical workstation in the past that has a function to
> filter floaters in a rendered image to remove objects below a certain size.
> I have looked through the VTK class list to try to find a similar filter
> without luck. Does anyone know of a filter that would do the job?
>
> Thanks, Kevin
>
> _______________________________________________
> This is the private VTK discussion list.
> Please keep messages on-topic. Check the FAQ at:
> http://www.vtk.org/Wiki/VTK_FAQ
> Follow this link to subscribe/unsubscribe:
> http://www.vtk.org/mailman/listinfo/vtkusers
>



More information about the vtkusers mailing list