#include <itkExhaustiveOptimizer.h>

Inheritance diagram for itk::ExhaustiveOptimizer:

Collaboration diagram for itk::ExhaustiveOptimizer:

Public Types
typedef SmartPointer< const Self >	ConstPointer
typedef SmartPointer< Self >	Pointer
typedef ExhaustiveOptimizer	Self
typedef Array< SizeValueType >	StepsType
typedef SingleValuedNonLinearOptimizer	Superclass
Public Member Functions
virtual ::itk::LightObject::Pointer	CreateAnother (void) const
virtual const ParametersType &	GetCurrentIndex ()
virtual const MeasureType &	GetCurrentValue ()
virtual const MeasureType &	GetMaximumMetricValue ()
virtual const ParametersType &	GetMaximumMetricValuePosition ()
virtual const SizeValueType &	GetMaximumNumberOfIterations ()
virtual const MeasureType &	GetMinimumMetricValue ()
virtual const ParametersType &	GetMinimumMetricValuePosition ()
virtual const char *	GetNameOfClass () const
virtual const StepsType &	GetNumberOfSteps ()
virtual const double &	GetStepLength ()
const std::string	GetStopConditionDescription () const
void	ResumeWalking (void)
virtual void	SetNumberOfSteps (StepsType _arg)
virtual void	SetStepLength (double _arg)
virtual void	StartOptimization (void)
void	StartWalking (void)
void	StopWalking (void)
Static Public Member Functions
static Pointer	New ()
Protected Member Functions
void	AdvanceOneStep (void)
	ExhaustiveOptimizer ()
void	IncrementIndex (ParametersType &param)
void	PrintSelf (std::ostream &os, Indent indent) const
virtual	~ExhaustiveOptimizer ()
Protected Attributes
ParametersType	m_CurrentIndex
SizeValueType	m_CurrentIteration
unsigned int	m_CurrentParameter
MeasureType	m_CurrentValue
MeasureType	m_MaximumMetricValue
ParametersType	m_MaximumMetricValuePosition
SizeValueType	m_MaximumNumberOfIterations
MeasureType	m_MinimumMetricValue
ParametersType	m_MinimumMetricValuePosition
StepsType	m_NumberOfSteps
double	m_StepLength
bool	m_Stop
Private Member Functions
	ExhaustiveOptimizer (const Self &)
void	operator= (const Self &)
Private Attributes
std::ostringstream	m_StopConditionDescription

Detailed Description

Optimizer that fully samples a grid on the parametric space.

This optimizer is equivalent to an exahaustive search in a discrete grid defined over the parametric space. The grid is centered on the initial position. The subdivisions of the grid along each one of the dimensions of the parametric space is defined by an array of number of steps.

A typical use is to plot the metric space to get an idea of how noisy it is. An example is given below, where it is desired to plot the metric space with respect to translations along x, y and z in a 3D registration application: Here it is assumed that the transform is Euler3DTransform.

 OptimizerType::StepsType steps( m_Transform->GetNumberOfParameters() );
 steps[1] = 10;
 steps[2] = 10;
 steps[3] = 10;
 m_Optimizer->SetNumberOfSteps( steps );
 m_Optimizer->SetStepLength( 2 );

The optimizer throws IterationEvents after every iteration. We use this to plot the metric space in an image as follows:

  if( itk::IterationEvent().CheckEvent(& event ) )
    {
    IndexType index;
    index[0] = m_Optimizer->GetCurrentIndex()[0];
    index[1] = m_Optimizer->GetCurrentIndex()[1];
    index[2] = m_Optimizer->GetCurrentIndex()[2];
    image->SetPixel( index, m_Optimizer->GetCurrentValue() );
    }

The image size is expected to be 11 x 11 x 11.

If you wish to use different step lengths along each parametric axis, you can use the SetScales() method. This accepts an array, each element represents the number of subdivisions per step length. For instance scales of [0.5 1 4] along with a step length of 2 will cause the optimizer to search the metric space on a grid with x,y,z spacing of [1 2 8].

Physical dimensions of the grid are influenced by both the scales and the number of steps along each dimension, a side of the region is stepLength*(2*numberOfSteps[d]+1)*scaling[d].

Definition at line 82 of file itkExhaustiveOptimizer.h.

Member Typedef Documentation

typedef SmartPointer< const Self > itk::ExhaustiveOptimizer::ConstPointer

Reimplemented from itk::SingleValuedNonLinearOptimizer.

Definition at line 90 of file itkExhaustiveOptimizer.h.

typedef SmartPointer< Self > itk::ExhaustiveOptimizer::Pointer

Reimplemented from itk::SingleValuedNonLinearOptimizer.

Definition at line 89 of file itkExhaustiveOptimizer.h.

typedef ExhaustiveOptimizer itk::ExhaustiveOptimizer::Self

Standard "Self" typedef.

Reimplemented from itk::SingleValuedNonLinearOptimizer.

Definition at line 87 of file itkExhaustiveOptimizer.h.

typedef Array< SizeValueType > itk::ExhaustiveOptimizer::StepsType

Definition at line 92 of file itkExhaustiveOptimizer.h.

typedef SingleValuedNonLinearOptimizer itk::ExhaustiveOptimizer::Superclass

Reimplemented from itk::SingleValuedNonLinearOptimizer.

Definition at line 88 of file itkExhaustiveOptimizer.h.

Constructor & Destructor Documentation

itk::ExhaustiveOptimizer::ExhaustiveOptimizer ( ) [protected]

virtual itk::ExhaustiveOptimizer::~ExhaustiveOptimizer ( ) [inline, protected, virtual]

Definition at line 124 of file itkExhaustiveOptimizer.h.

itk::ExhaustiveOptimizer::ExhaustiveOptimizer ( const Self & ) [private]

Member Function Documentation

void itk::ExhaustiveOptimizer::AdvanceOneStep ( void ) [protected]

Advance to the next grid position.

virtual::itk::LightObject::Pointer itk::ExhaustiveOptimizer::CreateAnother ( void ) const [virtual]

Create an object from an instance, potentially deferring to a factory. This method allows you to create an instance of an object that is exactly the same type as the referring object. This is useful in cases where an object has been cast back to a base class.

Reimplemented from itk::SingleValuedNonLinearOptimizer.

virtual const ParametersType& itk::ExhaustiveOptimizer::GetCurrentIndex ( ) [virtual]

virtual const MeasureType& itk::ExhaustiveOptimizer::GetCurrentValue ( ) [virtual]

virtual const MeasureType& itk::ExhaustiveOptimizer::GetMaximumMetricValue ( ) [virtual]

virtual const ParametersType& itk::ExhaustiveOptimizer::GetMaximumMetricValuePosition ( ) [virtual]

virtual const SizeValueType& itk::ExhaustiveOptimizer::GetMaximumNumberOfIterations ( ) [virtual]

virtual const MeasureType& itk::ExhaustiveOptimizer::GetMinimumMetricValue ( ) [virtual]

virtual const ParametersType& itk::ExhaustiveOptimizer::GetMinimumMetricValuePosition ( ) [virtual]

virtual const char* itk::ExhaustiveOptimizer::GetNameOfClass ( ) const [virtual]

Run-time type information (and related methods).

Reimplemented from itk::SingleValuedNonLinearOptimizer.

virtual const StepsType& itk::ExhaustiveOptimizer::GetNumberOfSteps ( ) [virtual]

virtual const double& itk::ExhaustiveOptimizer::GetStepLength ( ) [virtual]

const std::string itk::ExhaustiveOptimizer::GetStopConditionDescription ( ) const [virtual]

Get the reason for termination

Reimplemented from itk::Optimizer.

void itk::ExhaustiveOptimizer::IncrementIndex ( ParametersType & param ) [protected]

static Pointer itk::ExhaustiveOptimizer::New ( ) [static]

Method for creation through the object factory.

Reimplemented from itk::SingleValuedNonLinearOptimizer.

void itk::ExhaustiveOptimizer::operator= ( const Self & ) [private]

Types inherited from the superclass

Reimplemented from itk::SingleValuedNonLinearOptimizer.

void itk::ExhaustiveOptimizer::PrintSelf	(	std::ostream &	os,
		Indent	indent
	)		const `[protected, virtual]`

Methods invoked by Print() to print information about the object including superclasses. Typically not called by the user (use Print() instead) but used in the hierarchical print process to combine the output of several classes.

Reimplemented from itk::SingleValuedNonLinearOptimizer.

void itk::ExhaustiveOptimizer::ResumeWalking ( void )