[Insight-users] utilization of itkMRFImageFilter

[jbehar at messel.emse.fr]

Hello ITK community,

I have a question about about the itkMRFImageFilter. I have segmented  
the human heart using the kmeans (Step 1) as well as another pipeline  
in order to isolate the heart from the other organs/artifacts that  
have been segmented with the heart... (Step 2)

I would like to refine the initial classification by introducing  
spacial coherence (between Step 1 and Step 2). In order to do so, I  
tried to use the example "ScalarImageMarkovRandomField1.cxx" with two  
images. The first "image.dicom" is a DICOM image which corresponds to  
a slice of a volume (CT) and the second label.png is the result of the  
segmentation after applying the itkScalarImageKmeansImageFilter.

I also give the initial means (we took 7 clusters): -3024 / 185.926 /  
-689.681 / -480.271 / 50.6863 / -107.3 /423.254.
Those means are returned using kmeansFilter->GetFinalMeans()

Moreover I have add this piece of code at the begining of the Markov  
example in order to process dicom images:

...
   typedef itk::GDCMImageIO           ImageIOType; //*
   ReaderType::Pointer reader = ReaderType::New();

   reader->SetFileName( inputImageFileName );
   ImageIOType::Pointer gdcmImageIO = ImageIOType::New(); //*
   reader->SetImageIO( gdcmImageIO ); //*

   try //*
     {
     reader->Update();
     }
   catch (itk::ExceptionObject & e)
     {
     std::cerr << "exception in file reader " << std::endl;
     std::cerr << e << std::endl;
     return EXIT_FAILURE;
     }
...


And it does not produce the expected result...

If someone can give me a clue on how to use this filter,

Thanks for your help,

Joachim B.

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.