[ITK-users] MutualInformationImageToImageMetric without transformation

[jan]

Dear Carlos,
thank you for your help. i'm still experiencing some difficulties. I tried
your proposed way like this:

    typedef itk::MutualInformationImageToImageMetric< UChar2DImageType,
UChar2DImageType > MetricType;
    typedef itk::TranslationTransform< double, 2 > TransformType;
    typedef itk::GradientDescentOptimizer OptimizerType;
    MetricType::Pointer metric = MetricType::New();
    TransformType::Pointer transform = TransformType::New();

    MetricType::TransformParametersType displacement(
transform->ParametersDimension );
    displacement[0] = 0.0;
    displacement[1] = 0.0;

    metric->SetFixedImage( fixedImage );
    metric->SetMovingImage( movingImage);
    metric->SetTransform( transform );
    metric->GetValue( displacement );

When I am trying to get the MI-value with the displacement parameters, my
program crashes.



i tried another way with an optimizer that does only one iteration:

    typedef itk::TranslationTransform< double, 2 > TransformType;
    typedef itk::GradientDescentOptimizer OptimizerType;
    typedef itk::LinearInterpolateImageFunction< UChar2DImageType, double >
InterpolatorType;
    typedef itk::ImageRegistrationMethod< UChar2DImageType, UChar2DImageType
> RegistrationType;
    typedef itk::MutualInformationImageToImageMetric< UChar2DImageType,
UChar2DImageType > MetricType;
    typedef RegistrationType::ParametersType ParametersType;

    TransformType::Pointer transform = TransformType::New();
    OptimizerType::Pointer optimizer = OptimizerType::New();
    InterpolatorType::Pointer interpolator = InterpolatorType::New();
    RegistrationType::Pointer registration = RegistrationType::New();
    MetricType::Pointer metric = MetricType::New();

    registration->SetOptimizer( optimizer );
    registration->SetTransform( transform );
    registration->SetInterpolator( interpolator );
    registration->SetMetric( metric );

    registration->SetFixedImage( fixedImage );
    registration->SetMovingImage( movingImage );

    ParametersType initialParameters( transform->GetNumberOfParameters() );
    initialParameters[0] = 0.0;
    initialParameters[1] = 0.0;

    registration->SetInitialTransformParameters( initialParameters );

    optimizer->SetNumberOfIterations( 1 );

    try
    {
        registration->Update();
    }
    catch( itk::ExceptionObject & err )
    {
        std::cout << "ExceptionObject caught !" << std::endl;
        std::cout << err << std::endl;
    }

    double bestValue = optimizer->GetValue();

The problem here is that I get the following Error:

     "Description: itk::ERROR:
MutualInformationImageToImageMetric(0x48e0860): Standard deviation is too
small"



Any suggestions for a convenient solution?

Thanks, 
Jan



--
View this message in context: http://itk-insight-users.2283740.n2.nabble.com/MutualInformationImageToImageMetric-without-transformation-tp7586081p7586118.html
Sent from the ITK Insight Users mailing list archive at Nabble.com.