Computes similarity between two objects to be registered. More...

#include <itkGradientDifferenceImageToImageMetric.h>

Inheritance diagram for itk::GradientDifferenceImageToImageMetric< TFixedImage, TMovingImage >:

Collaboration diagram for itk::GradientDifferenceImageToImageMetric< TFixedImage, TMovingImage >:

Public Types
typedef CastFixedImageFilterType::Pointer	CastFixedImageFilterPointer
typedef itk::CastImageFilter < FixedImageType, FixedGradientImageType >	CastFixedImageFilterType
typedef CastMovedImageFilterType::Pointer	CastMovedImageFilterPointer
typedef itk::CastImageFilter < TransformedMovingImageType, MovedGradientImageType >	CastMovedImageFilterType
typedef SmartPointer< const Self >	ConstPointer
typedef Superclass::DerivativeType	DerivativeType
typedef itk::Image< RealType, itkGetStaticConstMacro(FixedImageDimension) >	FixedGradientImageType
typedef FixedGradientImageType::PixelType	FixedGradientPixelType
typedef Superclass::FixedImageConstPointer	FixedImageConstPointer
typedef TFixedImage::PixelType	FixedImagePixelType
typedef Superclass::FixedImageType	FixedImageType
typedef Superclass::MeasureType	MeasureType
typedef itk::Image< RealType, itkGetStaticConstMacro(MovedImageDimension) >	MovedGradientImageType
typedef MovedGradientImageType::PixelType	MovedGradientPixelType
typedef TMovingImage::PixelType	MovedImagePixelType
typedef Superclass::MovingImageConstPointer	MovingImageConstPointer
typedef Superclass::MovingImageType	MovingImageType
typedef SmartPointer< Self >	Pointer
typedef Superclass::RealType	RealType
typedef GradientDifferenceImageToImageMetric	Self
typedef ImageToImageMetric < TFixedImage, TMovingImage >	Superclass
typedef itk::Image < FixedImagePixelType, itkGetStaticConstMacro(FixedImageDimension) >	TransformedMovingImageType
typedef Superclass::TransformJacobianType	TransformJacobianType
typedef itk::ResampleImageFilter < MovingImageType, TransformedMovingImageType >	TransformMovingImageFilterType
typedef Superclass::TransformParametersType	TransformParametersType
typedef Superclass::TransformPointer	TransformPointer
typedef Superclass::TransformType	TransformType
Public Member Functions
virtual ::itk::LightObject::Pointer	CreateAnother (void) const
void	GetDerivative (const TransformParametersType &parameters, DerivativeType &derivative) const
virtual const char *	GetNameOfClass () const
MeasureType	GetValue (const TransformParametersType &parameters) const
void	GetValueAndDerivative (const TransformParametersType &parameters, MeasureType &Value, DerivativeType &derivative) const
virtual void	Initialize (void) throw ( ExceptionObject )
void	WriteGradientImagesToFiles (void) const
Static Public Member Functions
static Pointer	New ()
Static Public Attributes
static const unsigned int	FixedImageDimension = TFixedImage::ImageDimension
static const unsigned int	MovedImageDimension = MovingImageType::ImageDimension
Protected Types
typedef NeighborhoodOperatorImageFilter < FixedGradientImageType, FixedGradientImageType >	FixedSobelFilter
typedef NeighborhoodOperatorImageFilter < MovedGradientImageType, MovedGradientImageType >	MovedSobelFilter
Protected Member Functions
MeasureType	ComputeMeasure (const TransformParametersType &parameters, const double *subtractionFactor) const
void	ComputeMovedGradientRange (void) const
void	ComputeVariance (void) const
Private Member Functions
	GradientDifferenceImageToImageMetric (const Self &)
void	operator= (const Self &)
Private Attributes
CastFixedImageFilterPointer	m_CastFixedImageFilter
CastMovedImageFilterPointer	m_CastMovedImageFilter
double	m_DerivativeDelta
ZeroFluxNeumannBoundaryCondition < FixedGradientImageType >	m_FixedBoundCond
FixedSobelFilter::Pointer	m_FixedSobelFilters [itkGetStaticConstMacro(FixedImageDimension)]
SobelOperator < FixedGradientPixelType, itkGetStaticConstMacro(FixedImageDimension) >	m_FixedSobelOperators [FixedImageDimension]
FixedGradientPixelType	m_MaxFixedGradient [FixedImageDimension]
MovedGradientPixelType	m_MaxMovedGradient [MovedImageDimension]
FixedGradientPixelType	m_MinFixedGradient [FixedImageDimension]
MovedGradientPixelType	m_MinMovedGradient [MovedImageDimension]
ZeroFluxNeumannBoundaryCondition < MovedGradientImageType >	m_MovedBoundCond
MovedSobelFilter::Pointer	m_MovedSobelFilters [itkGetStaticConstMacro(MovedImageDimension)]
SobelOperator < MovedGradientPixelType, itkGetStaticConstMacro(MovedImageDimension) >	m_MovedSobelOperators [MovedImageDimension]
TransformMovingImageFilterType::Pointer	m_TransformMovingImageFilter
MovedGradientPixelType	m_Variance [FixedImageDimension]
virtual void	SetDerivativeDelta (double _arg)
virtual const double &	GetDerivativeDelta ()
	GradientDifferenceImageToImageMetric ()
virtual	~GradientDifferenceImageToImageMetric ()
void	PrintSelf (std::ostream &os, Indent indent) const

Detailed Description

template<class TFixedImage, class TMovingImage>
class itk::GradientDifferenceImageToImageMetric< TFixedImage, TMovingImage >

Computes similarity between two objects to be registered.

This Class is templated over the type of the Images to be compared and over the type of transformation and Iterpolator to be used.

This metric computes the sum of squared differences between pixels in the derivatives of the moving and fixed images after passing the squared difference through a function of type $\frac{1}{1+x}$ .

Warning:: THIS IMAGE METRIC IS CURRENTLY UNDER DEBUGGING. USE AT YOUR OWN RISK.

Spatial correspondance between both images is established through a Transform. Pixel values are taken from the Moving image. Their positions are mapped to the Fixed image and result in general in non-grid position on it. Values at these non-grid position of the Fixed image are interpolated using a user-selected Interpolator.

Implementation of this class is based on: Hipwell, J. H., et. al. (2003), "Intensity-Based 2-D-3D Registration of Cerebral Angiograms,", IEEE Transactions on Medical Imaging, 22(11):1417-1426.

Definition at line 58 of file itkGradientDifferenceImageToImageMetric.h.