itk::OnePlusOneEvolutionaryOptimizer Class Reference
[Numerics, Optimizers]

1+1 evolutionary strategy optimizer. More...

#include <itkOnePlusOneEvolutionaryOptimizer.h>

Inheritance diagram for itk::OnePlusOneEvolutionaryOptimizer:

[legend]Collaboration diagram for itk::OnePlusOneEvolutionaryOptimizer:

[legend]List of all members.

Public Types
typedef OnePlusOneEvolutionaryOptimizer	Self
typedef SingleValuedNonLinearOptimizer	Superclass
typedef SmartPointer< Self >	Pointer
typedef SmartPointer< const Self >	ConstPointer
typedef SingleValuedCostFunction	CostFunctionType
typedef CostFunctionType::Pointer	CostFunctionPointer
typedef Statistics::RandomVariateGeneratorBase	NormalVariateGeneratorType
Public Methods
virtual const char *	GetClassName () const
virtual double	GetGrowthFactor () const
virtual double	GetShrinkFactor () const
virtual double	GetInitialRadius () const
void	SetNormalVariateGenerator (NormalVariateGeneratorType *generator)
void	Initialize (double initialRadius, double grow=-1, double shrink=-1)
MeasureType	GetCurrentCost ()
virtual unsigned int	GetCurrentIteration () const
void	StartOptimization ()
void	StopOptimization ()
virtual void	SetMaximumIteration (unsigned int _arg)
virtual unsigned int	GetMaximumIteration () const
virtual void	SetEpsilon (double _arg)
virtual double	GetEpsilon () const
Static Public Methods
Pointer	New ()
Protected Methods
	OnePlusOneEvolutionaryOptimizer ()
virtual	~OnePlusOneEvolutionaryOptimizer ()
void	PrintSelf (std::ostream &os, Indent indent) const

---

Detailed Description

1+1 evolutionary strategy optimizer.

This optimizer searches for the optimal parameters. It changes its search radius and position using the grow factor ,shrink factor, and isotropic probability function (which is a random unit normal variate generator).

This optimizer needs a cost function and a random unit normal variate generator. The cost function should return cost with new position in parameter space which will be generated by 1+1 evolutionary strategy. Users should plug-in the random unit normal variate generator using SetNormalVariateGenerator method.

The SetEpsilon method is the minimum value for the frobenius_norm of the covariance matrix. If the fnorm is smaller than this value, the optimization process will stop even before it hits the maximum iteration.

Another way to stop the optimization process is calling the StopOptimization method. At next iteration after calling it, the optimization process will stop.

This optimizing scheme was initially developed and implemented by Martin Styner, Univ. of North Carolina at Chapel Hill, and his colleagues.

For more details. refer to the following articles. "Parametric estimate of intensity inhomogeneities applied to MRI" Martin Styner, G. Gerig, Christian Brechbuehler, Gabor Szekely, IEEE TRANSACTIONS ON MEDICAL IMAGING; 19(3), pp. 153-165, 2000, (http://www.cs.unc.edu/~styner/docs/tmi00.pdf)

"Evaluation of 2D/3D bias correction with 1+1ES-optimization" Martin Styner, Prof. Dr. G. Gerig (IKT, BIWI, ETH Zuerich), TR-197 (http://www.cs.unc.edu/~styner/docs/StynerTR97.pdf)

See also:

NormalVariateGenerator

Definition at line 68 of file itkOnePlusOneEvolutionaryOptimizer.h.

---

Member Typedef Documentation

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

Reimplemented from itk::SingleValuedNonLinearOptimizer. Definition at line 76 of file itkOnePlusOneEvolutionaryOptimizer.h.

typedef CostFunctionType::Pointer itk::OnePlusOneEvolutionaryOptimizer::CostFunctionPointer

Reimplemented from itk::SingleValuedNonLinearOptimizer. Definition at line 86 of file itkOnePlusOneEvolutionaryOptimizer.h.

typedef SingleValuedCostFunction itk::OnePlusOneEvolutionaryOptimizer::CostFunctionType

Type of the Cost Function Reimplemented from itk::SingleValuedNonLinearOptimizer. Definition at line 85 of file itkOnePlusOneEvolutionaryOptimizer.h.

typedef Statistics::RandomVariateGeneratorBase itk::OnePlusOneEvolutionaryOptimizer::NormalVariateGeneratorType

Normal random variate generator type. Definition at line 89 of file itkOnePlusOneEvolutionaryOptimizer.h.

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

Reimplemented from itk::SingleValuedNonLinearOptimizer. Definition at line 75 of file itkOnePlusOneEvolutionaryOptimizer.h.

typedef OnePlusOneEvolutionaryOptimizer itk::OnePlusOneEvolutionaryOptimizer::Self

Standard "Self" typedef. Reimplemented from itk::SingleValuedNonLinearOptimizer. Definition at line 73 of file itkOnePlusOneEvolutionaryOptimizer.h.

typedef SingleValuedNonLinearOptimizer itk::OnePlusOneEvolutionaryOptimizer::Superclass

Reimplemented from itk::SingleValuedNonLinearOptimizer. Definition at line 74 of file itkOnePlusOneEvolutionaryOptimizer.h.

---

Constructor & Destructor Documentation

itk::OnePlusOneEvolutionaryOptimizer::OnePlusOneEvolutionaryOptimizer (    )  [protected]

virtual itk::OnePlusOneEvolutionaryOptimizer::~OnePlusOneEvolutionaryOptimizer (    )  [protected, virtual]

---

Member Function Documentation

virtual const char* itk::OnePlusOneEvolutionaryOptimizer::GetClassName (    )  const [virtual]

Run-time type information (and related methods). Reimplemented from itk::SingleValuedNonLinearOptimizer.

MeasureType itk::OnePlusOneEvolutionaryOptimizer::GetCurrentCost (    )  [inline]

Definition at line 121 of file itkOnePlusOneEvolutionaryOptimizer.h.

virtual unsigned int itk::OnePlusOneEvolutionaryOptimizer::GetCurrentIteration (    )  const [virtual]

virtual double itk::OnePlusOneEvolutionaryOptimizer::GetEpsilon (    )  const [virtual]

Set/Get the minimal size of search radius (frobenius_norm of covariance matrix).

virtual double itk::OnePlusOneEvolutionaryOptimizer::GetGrowthFactor (    )  const [virtual]

Get the search radius grow factor in parameter space.

virtual double itk::OnePlusOneEvolutionaryOptimizer::GetInitialRadius (    )  const [virtual]

Get initial search radius in parameter space.

virtual unsigned int itk::OnePlusOneEvolutionaryOptimizer::GetMaximumIteration (    )  const [virtual]

Set/Get maximum iteration limit.

virtual double itk::OnePlusOneEvolutionaryOptimizer::GetShrinkFactor (    )  const [virtual]

Get search radius shrink factor.

void itk::OnePlusOneEvolutionaryOptimizer::Initialize (  double    initialRadius, double    grow = -1, double    shrink = -1 )

Initializes the optimizer. Before running this optimizer, this function should have been called. initialRadius: search radius in parameter space grow: search radius grow factor shrink: searhc radius shrink factor

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

Method for creation through the object factory. Reimplemented from itk::SingleValuedNonLinearOptimizer.

void itk::OnePlusOneEvolutionaryOptimizer::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.

virtual void itk::OnePlusOneEvolutionaryOptimizer::SetEpsilon (  double    _arg )  [virtual]

Set/Get the minimal size of search radius (frobenius_norm of covariance matrix).

virtual void itk::OnePlusOneEvolutionaryOptimizer::SetMaximumIteration (  unsigned int    _arg )  [virtual]

Set/Get maximum iteration limit.

void itk::OnePlusOneEvolutionaryOptimizer::SetNormalVariateGenerator (  NormalVariateGeneratorType *    generator )

void itk::OnePlusOneEvolutionaryOptimizer::StartOptimization (    )  [virtual]

Start optimization. Optimization will stop when it meets either of two termination conditions, the maximum iteration limit or epsilon (minimal search radius) Reimplemented from itk::Optimizer.

void itk::OnePlusOneEvolutionaryOptimizer::StopOptimization (    )  [inline]

when users call StartOptimization, this value will be set false. By calling StopOptimization, this flag will be set true, and optimization will stop at the next iteration. Definition at line 133 of file itkOnePlusOneEvolutionaryOptimizer.h. References itk::MeasureType, and HardConnectedComponentImageFilter::PrintSelf().

---

The documentation for this class was generated from the following file:

• itkOnePlusOneEvolutionaryOptimizer.h

---