Selects a region of space where two independent fronts run towards each other. More...

#include <itkCollidingFrontsImageFilter.h>

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

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

Public Types
typedef SmartPointer< const Self >	ConstPointer
typedef itk::FastMarchingUpwindGradientImageFilter < TInputImage, TOutputImage >	FastMarchingUpwindGradientImageFilterType
typedef FastMarchingUpwindGradientImageFilterType::GradientImageType	GradientImageType
typedef FastMarchingUpwindGradientImageFilterType::IndexType	IndexType
typedef InputImageType::Pointer	InputImagePointer
typedef TInputImage	InputImageType
typedef TInputImage::PixelType	InputPixelType
typedef FastMarchingUpwindGradientImageFilterType::NodeContainer	NodeContainer
typedef FastMarchingUpwindGradientImageFilterType::NodeContainerPointer	NodeContainerPointer
typedef FastMarchingUpwindGradientImageFilterType::NodeType	NodeType
typedef OutputImageType::Pointer	OutputImagePointer
typedef Superclass::OutputImageRegionType	OutputImageRegionType
typedef TOutputImage	OutputImageType
typedef TOutputImage::PixelType	OutputPixelType
typedef FastMarchingUpwindGradientImageFilterType::PixelType	PixelType
typedef SmartPointer< Self >	Pointer
typedef NumericTraits < InputPixelType >::RealType	RealType
typedef CollidingFrontsImageFilter	Self
typedef ImageToImageFilter < TInputImage, TOutputImage >	Superclass
Public Member Functions
virtual void	ApplyConnectivityOff ()
virtual void	ApplyConnectivityOn ()
virtual ::itk::LightObject::Pointer	CreateAnother (void) const
virtual bool	GetApplyConnectivity () const
virtual const char *	GetNameOfClass () const
virtual double	GetNegativeEpsilon () const
NodeContainerPointer	GetSeedPoints1 ()
NodeContainerPointer	GetSeedPoints2 ()
virtual bool	GetStopOnTargets () const
virtual void	SetApplyConnectivity (bool _arg)
virtual void	SetNegativeEpsilon (double _arg)
virtual void	SetStopOnTargets (bool _arg)
virtual void	StopOnTargetsOff ()
virtual void	StopOnTargetsOn ()
	typedef (Concept::HasNumericTraits< InputPixelType >) InputHasNumericTraitsCheck

void	SetSeedPoints1 (NodeContainer *points)

void	SetSeedPoints2 (NodeContainer *points)
Static Public Member Functions
static Pointer	New ()
Static Public Attributes
static const unsigned int	ImageDimension = TOutputImage::ImageDimension
Protected Member Functions
	CollidingFrontsImageFilter ()
void	GenerateData ()
void	PrintSelf (std::ostream &, Indent) const
virtual	~CollidingFrontsImageFilter ()
Private Member Functions
	CollidingFrontsImageFilter (const Self &)
void	operator= (const Self &)
Private Attributes
bool	m_ApplyConnectivity
double	m_NegativeEpsilon
NodeContainerPointer	m_SeedPoints1
NodeContainerPointer	m_SeedPoints2
bool	m_StopOnTargets

Detailed Description

template<typename TInputImage, typename TOutputImage>
class itk::CollidingFrontsImageFilter< TInputImage, TOutputImage >

Selects a region of space where two independent fronts run towards each other.

The filter can be used to quickly segment anatomical structures (e.g. for level set initialization).

The filter uses two instances of FastMarchingUpwindGradientImageFilter to compute the gradients of arrival times of two wavefronts propagating from two sets of seeds. The input of the filter is used as the speed of the two wavefronts. The output is the dot product between the two gradient vector fields.

The filter works on the following basic idea. In the regions where the dot product between the two gradient fields is negative, the two fronts propagate in opposite directions. In the regions where the dot product is positive, the two fronts propagate in the same direction. This can be used to extract the region of space between two sets of points.

If StopOnTargets is On, then each front will stop as soon as all seeds of the other front have been reached. This can markedly speed up the execution of the filter, since wave propagation does not take place on the complete image.

Optionally, a connectivity criterion can be applied to the resulting dot product image. In this case, the only negative region in the output image is the one connected to the seeds.

Author:: Luca Antiga Ph.D. Biomedical Technologies Laboratory, Bioengineering Deparment, Mario Negri Institute, Italy.

Definition at line 61 of file itkCollidingFrontsImageFilter.h.