[Insight-users] Bspline 3D deformable registration

[Luis Ibanez]

Hi Shoosh,

Do you want to make multi-resolution on the images ?
or do you want to make it on the BSpline grid ?

Please let us know,

In either case, you are starting from the wrong code.
The example that you posted is appropriate for
the Demons method but not for the BSplineDeformable
Transform.

We can point you to the right framework once we know
if you want to do image multi-resolution or BSpline
multi-resolution.



    Thanks


       Luis


--------------------
shoosh moosh wrote:
> I'm trying to write code to register 3D images using a multiresolution first degree B-spline deformation model. I'm not sure where to include/spedify the B-spline part. The code is attached. 
> Thanks
> 
> 
> 
> _________________________________________________________________
> Explore the seven wonders of the world
> http://search.msn.com/results.aspx?q=7+wonders+world&mkt=en-US&form=QBRE
> 
> 
> ------------------------------------------------------------------------
> 
> /*******************************************************************************
> Program:   Demons Registration using ITK
> Id:        $Id$
> Language:  C++
> *******************************************************************************/
> 
> #include <iostream>
> using std::cerr;
> using std::clog;
> using std::endl;
> 
> #include <cstdlib>
> 
> #include "demons.h"
> 
> // ITK IO includes
> #include "itkImage.h"
> #include "itkImageFileReader.h"
> #include "itkImageFileWriter.h"
> 
> // ITK Registration includes
> #include "itkMultiResolutionPDEDeformableRegistration.h"
> #include "itkMultiResolutionImageRegistrationMethod.h"
> #include "itkMultiResolutionPyramidImageFilter.h"
> #include "itkDemonsRegistrationFilter.h" 
> #include "itkHistogramMatchingImageFilter.h" 
> #include "itkCastImageFilter.h" 
> #include "itkWarpImageFilter.h" 
> #include "itkLinearInterpolateImageFunction.h"
> 
> int demons( std::vector< std::string > args )
> {
>   
>   // Verify the number of parameters in the command line
>   if( args.size() != 4 ){
>     clog << "usage: " << endl << args[0] << " target source deformation" << endl;
>     return EXIT_FAILURE;
>   }
>   
>   // define ITK short-hand types
>   const unsigned int Dimension = 3; 
>   typedef short PixelType;
>   typedef float InternalPixelType; 
>   typedef itk::Image< PixelType, Dimension > ImageType;
>   typedef itk::Image< InternalPixelType, Dimension > InternalImageType;
>   typedef itk::CastImageFilter< ImageType, InternalImageType > ImageCasterType;
> 
>   // setup input file readers
>   typedef itk::ImageFileReader< ImageType >  ReaderType;
>   ReaderType::Pointer targetReader = ReaderType::New();
>   targetReader->SetFileName( args[1] );
>   ReaderType::Pointer sourceReader = ReaderType::New();
>   sourceReader->SetFileName( args[2] );
> 
>   // cast target and source to float
>   ImageCasterType::Pointer targetImageCaster = ImageCasterType::New(); 
>   ImageCasterType::Pointer sourceImageCaster = ImageCasterType::New(); 
>   targetImageCaster->SetInput( targetReader->GetOutput() ); 
>   sourceImageCaster->SetInput( sourceReader->GetOutput() );
>   
>   // match the histograms between source and target
>   typedef itk::HistogramMatchingImageFilter< InternalImageType, InternalImageType > MatchingFilterType; 
>   MatchingFilterType::Pointer matcher = MatchingFilterType::New();
>   matcher->SetInput( sourceImageCaster->GetOutput() ); 
>   matcher->SetReferenceImage( targetImageCaster->GetOutput() );
>   matcher->SetNumberOfHistogramLevels( 1024 ); 
>   matcher->SetNumberOfMatchPoints( 7 );
>   matcher->ThresholdAtMeanIntensityOn();
> 
>   // setup the deformation field and filter
>   typedef itk::Vector< float, Dimension > VectorPixelType; 
>   typedef itk::Image< VectorPixelType, Dimension > DeformationFieldType; 
>   typedef itk::DemonsRegistrationFilter< InternalImageType, 
>     InternalImageType, DeformationFieldType> RegistrationFilterType; 
>   RegistrationFilterType::Pointer filter = RegistrationFilterType::New();
>   filter->SetStandardDeviations( 1.0 );
> 
>   // use multiresolution scheme
>   typedef itk::MultiResolutionPDEDeformableRegistration
>     < InternalImageType, InternalImageType, DeformationFieldType >
>     MultiResRegistrationFilterType;
>   MultiResRegistrationFilterType::Pointer 
>     multires = MultiResRegistrationFilterType::New();
>   multires->SetRegistrationFilter( filter );
>   multires->SetNumberOfLevels( 5 );
>   multires->SetFixedImage( targetImageCaster->GetOutput() );
>   multires->SetMovingImage( matcher->GetOutput() );
>   unsigned int nIterations[5] = { 25, 60, 95, 130, 165 };
>   multires->SetNumberOfIterations( nIterations );
> 
>   // apply the registration filter
>   multires->Update();
> 
>   // write the deformation field
>   typedef itk::ImageFileWriter< DeformationFieldType >  DeformationWriterType;
>   DeformationWriterType::Pointer defwriter = DeformationWriterType::New();
>   defwriter->SetFileName( args[3] );
>   defwriter->SetInput( multires->GetOutput() );
>   defwriter->Update();
> 
>   return EXIT_SUCCESS;
> }
> 
> 
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users