Computes a diffeomorphic displacement field as the Lie group exponential of a vector field. More...

#include <itkExponentialDisplacementFieldImageFilter.h>

Inheritance diagram for itk::ExponentialDisplacementFieldImageFilter< TInputImage, TOutputImage >:

Collaboration diagram for itk::ExponentialDisplacementFieldImageFilter< TInputImage, TOutputImage >:

Public Types
typedef SmartPointer< const Self >	ConstPointer
typedef InputImageType::ConstPointer	InputImageConstPointer
typedef InputImageType::Pointer	InputImagePointer
typedef TInputImage	InputImageType
typedef InputPixelType::RealValueType	InputPixelRealValueType
typedef InputImageType::PixelType	InputPixelType
typedef OutputImageType::Pointer	OutputImagePointer
typedef TOutputImage	OutputImageType
typedef OutputImageType::PixelType	OutputPixelType
typedef SmartPointer< Self >	Pointer
typedef ExponentialDisplacementFieldImageFilter	Self
typedef ImageToImageFilter < TInputImage, TOutputImage >	Superclass
Public Member Functions
virtual ::itk::LightObject::Pointer	CreateAnother (void) const
virtual const char *	GetNameOfClass () const
	typedef (Concept::SameDimension< ImageDimension, PixelDimension >) SameDimensionCheck2
	typedef (Concept::SameDimension< ImageDimension, OutputImageDimension >) SameDimensionCheck1
	typedef (Concept::HasNumericTraits< typename OutputPixelType::ValueType >) OutputHasNumericTraitsCheck
	typedef (Concept::SameDimension< ImageDimension, OutputPixelDimension >) SameDimensionCheck3

virtual void	SetMaximumNumberOfIterations (unsigned int _arg)
virtual unsigned int	GetMaximumNumberOfIterations () const

virtual void	SetAutomaticNumberOfIterations (bool _arg)
virtual bool	GetAutomaticNumberOfIterations () const
virtual void	AutomaticNumberOfIterationsOn ()
virtual void	AutomaticNumberOfIterationsOff ()

virtual void	SetComputeInverse (bool _arg)
virtual bool	GetComputeInverse () const
virtual void	ComputeInverseOn ()
virtual void	ComputeInverseOff ()
Static Public Member Functions
static Pointer	New ()
Static Public Attributes

static const unsigned int	ImageDimension = TInputImage::ImageDimension
static const unsigned int	OutputImageDimension = TInputImage::ImageDimension
static const unsigned int	PixelDimension = InputPixelType::Dimension
static const unsigned int	OutputPixelDimension = OutputPixelType::Dimension
Protected Types
typedef AdderType::Pointer	AdderPointer
typedef AddImageFilter < OutputImageType, OutputImageType, OutputImageType >	AdderType
typedef CasterType::Pointer	CasterPointer
typedef CastImageFilter < InputImageType, OutputImageType >	CasterType
typedef DivideByConstantType::Pointer	DivideByConstantPointer
typedef DivideImageFilter < InputImageType, itk::Image < InputPixelRealValueType, ImageDimension > , OutputImageType >	DivideByConstantType
typedef FieldInterpolatorType::OutputType	FieldInterpolatorOutputType
typedef FieldInterpolatorType::Pointer	FieldInterpolatorPointer
typedef VectorLinearInterpolateNearestNeighborExtrapolateImageFunction < OutputImageType, double >	FieldInterpolatorType
typedef InputImageType::RegionType	RegionType
typedef VectorWarperType::Pointer	VectorWarperPointer
typedef WarpVectorImageFilter < OutputImageType, OutputImageType, OutputImageType >	VectorWarperType
Protected Member Functions
	ExponentialDisplacementFieldImageFilter ()
void	GenerateData ()
void	PrintSelf (std::ostream &os, Indent indent) const
virtual	~ExponentialDisplacementFieldImageFilter ()
Private Member Functions
	ExponentialDisplacementFieldImageFilter (const Self &)
void	operator= (const Self &)
Private Attributes
AdderPointer	m_Adder
bool	m_AutomaticNumberOfIterations
CasterPointer	m_Caster
bool	m_ComputeInverse
DivideByConstantPointer	m_Divider
unsigned int	m_MaximumNumberOfIterations
VectorWarperPointer	m_Warper

Detailed Description

template<class TInputImage, class TOutputImage>
class itk::ExponentialDisplacementFieldImageFilter< TInputImage, TOutputImage >

Computes a diffeomorphic displacement field as the Lie group exponential of a vector field.

ExponentialDisplacementFieldImageFilter takes a 'smooth' vector field as input and computes the displacement field that is its exponential.

Given that both the input and output displacement field are represented as discrete images with pixel type vector, the exponential will be only an estimation and will probably not correspond to a perfect exponential. The precision of the exponential can be improved at the price of increasing the computation time (number of iterations).

The method used for computing the exponential displacement field is an iterative scaling and squaring (cf Arsigny et al., "A Log-Euclidean Framework for Statistics on Diffeomorphisms", MICCAI'06).

$exp(\Phi) = exp( \frac{\Phi}{2^N} )^{2^N}$

This filter expects both the input and output images to be of pixel type Vector.

Author:: Tom Vercauteren, INRIA & Mauna Kea Technologies

This implementation was taken from the Insight Journal paper: http://hdl.handle.net/1926/510

Definition at line 63 of file itkExponentialDisplacementFieldImageFilter.h.