itk::Statistics::MixtureModelComponentBase< TSample > Class Template Reference

base class for distribution modules that supports analytical way to update the distribution parameters. More...

#include <itkMixtureModelComponentBase.h>

Inheritance diagram for itk::Statistics::MixtureModelComponentBase:

[legend]Collaboration diagram for itk::Statistics::MixtureModelComponentBase< TSample >:

[legend]List of all members.

[NOHEADER]
virtual const char *	GetClassName () const
Pointer	New ()
Public Types
typedef MixtureModelComponentBase	Self
typedef Object	Superclass
typedef SmartPointer< Self >	Pointer
typedef TSample::MeasurementVectorType	MeasurementVectorType
typedef MembershipFunctionBase< MeasurementVectorType >	MembershipFunctionType
typedef Array< double >	WeightArrayType
typedef Array< double >	ParametersType
Public Methods
virtual void	SetSample (TSample *sample)
TSample *	GetSample ()
MembershipFunctionType *	GetMembershipFunction ()
void	SetMinimalParametersChange (double change)
double	GetMinimalParametersChange ()
virtual void	SetParameters (const ParametersType &parameters)
virtual ParametersType	GetFullParameters ()
void	AreParametersModified (bool flag)
bool	AreParametersModified ()
void	SetWeight (int index, double value)
double	GetWeight (int index)
double	Evaluate (MeasurementVectorType &measurements)
WeightArrayType *	GetWeights ()
virtual void	Update ()
Protected Methods
	MixtureModelComponentBase ()
virtual	~MixtureModelComponentBase ()
void	PrintSelf (std::ostream &os, Indent indent) const
void	CreateWeightArray ()
void	DeleteWeightArray ()
void	SetMembershipFunction (MembershipFunctionType *function)
virtual void	GenerateData ()

---

Detailed Description

template<class TSample>
class itk::Statistics::MixtureModelComponentBase< TSample >

base class for distribution modules that supports analytical way to update the distribution parameters.

This class expects that its subclasses (distribution components) should have analytical expressions for updating its paraters using only the measurement vectors and their associated weights.

This class can be considered as a macro class that encapsulates the storage for the weights, model (subclasses of MembershipFunctionBase), and model parameter estimators (implemenation of analytical expressions).

Subclasses of this class should define their own distribution specific membership function. For example, GaussianMixtureModelComponent class defines and creates a GaussianDensityFunction object for that matter. Subclasses should also cast such membership function object to MembershipFunctionBase object. By doing that, users can get pointers to membership functions from different distributional model

See also:

ExpectationMaximizationMixtureModelEstimator

Definition at line 53 of file itkMixtureModelComponentBase.h.

---

Member Typedef Documentation

template<class TSample> typedef TSample::MeasurementVectorType itk::Statistics::MixtureModelComponentBase< TSample >::MeasurementVectorType

Reimplemented in itk::Statistics::GaussianMixtureModelComponent< TSample >. Definition at line 67 of file itkMixtureModelComponentBase.h.

template<class TSample> typedef MembershipFunctionBase< MeasurementVectorType > itk::Statistics::MixtureModelComponentBase< TSample >::MembershipFunctionType

typedef for the MembershipFunctionBase Reimplemented in itk::Statistics::GaussianMixtureModelComponent< TSample >. Definition at line 71 of file itkMixtureModelComponentBase.h.

template<class TSample> typedef Array< double > itk::Statistics::MixtureModelComponentBase< TSample >::ParametersType

Reimplemented in itk::Statistics::GaussianMixtureModelComponent< TSample >. Definition at line 76 of file itkMixtureModelComponentBase.h.

template<class TSample> typedef SmartPointer<Self> itk::Statistics::MixtureModelComponentBase< TSample >::Pointer

Reimplemented from itk::Object. Reimplemented in itk::Statistics::GaussianMixtureModelComponent< TSample >. Definition at line 60 of file itkMixtureModelComponentBase.h.

template<class TSample> typedef MixtureModelComponentBase itk::Statistics::MixtureModelComponentBase< TSample >::Self

Standard class typedefs. Reimplemented from itk::Object. Reimplemented in itk::Statistics::GaussianMixtureModelComponent< TSample >. Definition at line 58 of file itkMixtureModelComponentBase.h.

template<class TSample> typedef Object itk::Statistics::MixtureModelComponentBase< TSample >::Superclass

Reimplemented from itk::Object. Reimplemented in itk::Statistics::GaussianMixtureModelComponent< TSample >. Definition at line 59 of file itkMixtureModelComponentBase.h.

template<class TSample> typedef Array< double > itk::Statistics::MixtureModelComponentBase< TSample >::WeightArrayType

typedef of strorage for the weights Reimplemented in itk::Statistics::GaussianMixtureModelComponent< TSample >. Definition at line 74 of file itkMixtureModelComponentBase.h.

---

Constructor & Destructor Documentation

template<class TSample> itk::Statistics::MixtureModelComponentBase< TSample >::MixtureModelComponentBase (    )  [protected]

template<class TSample> virtual itk::Statistics::MixtureModelComponentBase< TSample >::~MixtureModelComponentBase (    )  [protected, virtual]

---

Member Function Documentation

template<class TSample> bool itk::Statistics::MixtureModelComponentBase< TSample >::AreParametersModified (    )

returns the value of parameter modified tag

template<class TSample> void itk::Statistics::MixtureModelComponentBase< TSample >::AreParametersModified (  bool    flag )

sets the parameters modified tag. if one or more of the membership funtion's parameters are changed, then flag should be true

template<class TSample> void itk::Statistics::MixtureModelComponentBase< TSample >::CreateWeightArray (    )  [protected]

allocates the weights array

template<class TSample> void itk::Statistics::MixtureModelComponentBase< TSample >::DeleteWeightArray (    )  [protected]

deallocates the weights array

template<class TSample> double itk::Statistics::MixtureModelComponentBase< TSample >::Evaluate (  MeasurementVectorType &    measurements )

returns the membership score of the "measurements" vector

template<class TSample> virtual void itk::Statistics::MixtureModelComponentBase< TSample >::GenerateData (    )  [protected, virtual]

Reimplemented in itk::Statistics::GaussianMixtureModelComponent< TSample >.

template<class TSample> virtual const char* itk::Statistics::MixtureModelComponentBase< TSample >::GetClassName (    )  const [virtual]

Standard Macros Reimplemented from itk::Object. Reimplemented in itk::Statistics::GaussianMixtureModelComponent< TSample >.

template<class TSample> virtual ParametersType itk::Statistics::MixtureModelComponentBase< TSample >::GetFullParameters (    )  [inline, virtual]

Definition at line 98 of file itkMixtureModelComponentBase.h.

template<class TSample> MembershipFunctionType* itk::Statistics::MixtureModelComponentBase< TSample >::GetMembershipFunction (    )

returns the pointer to the membership function object. Subclasses of this class are responsible for creating the actual membership function objects and cast them to MembershipFunctionBase objects

template<class TSample> double itk::Statistics::MixtureModelComponentBase< TSample >::GetMinimalParametersChange (    )  [inline]

Definition at line 93 of file itkMixtureModelComponentBase.h.

template<class TSample> TSample* itk::Statistics::MixtureModelComponentBase< TSample >::GetSample (    )

returns the sample pointer

template<class TSample> double itk::Statistics::MixtureModelComponentBase< TSample >::GetWeight (  int    index )

returns the index-th weight

template<class TSample> WeightArrayType* itk::Statistics::MixtureModelComponentBase< TSample >::GetWeights (    )

returns the pointer to the weights array

template<class TSample> Pointer itk::Statistics::MixtureModelComponentBase< TSample >::New (    )  [static]

Standard Macros Reimplemented from itk::Object. Reimplemented in itk::Statistics::GaussianMixtureModelComponent< TSample >.

template<class TSample> void itk::Statistics::MixtureModelComponentBase< TSample >::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::Object. Reimplemented in itk::Statistics::GaussianMixtureModelComponent< TSample >.

template<class TSample> void itk::Statistics::MixtureModelComponentBase< TSample >::SetMembershipFunction (  MembershipFunctionType *    function )  [protected]

stores the pointer to the membership function. subclasses use this funtion to store their membership function object after dynamic creation

template<class TSample> void itk::Statistics::MixtureModelComponentBase< TSample >::SetMinimalParametersChange (  double    change )  [inline]

Definition at line 90 of file itkMixtureModelComponentBase.h.

template<class TSample> virtual void itk::Statistics::MixtureModelComponentBase< TSample >::SetParameters (  const ParametersType &    parameters )  [virtual]

Reimplemented in itk::Statistics::GaussianMixtureModelComponent< TSample >.

template<class TSample> virtual void itk::Statistics::MixtureModelComponentBase< TSample >::SetSample (  TSample *    sample )  [virtual]

stores the sample pointer Reimplemented in itk::Statistics::GaussianMixtureModelComponent< TSample >.

template<class TSample> void itk::Statistics::MixtureModelComponentBase< TSample >::SetWeight (  int    index, double    value )

sets the index-th weight with the "value"

template<class TSample> virtual void itk::Statistics::MixtureModelComponentBase< TSample >::Update (    )  [virtual]

---

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

• itkMixtureModelComponentBase.h

---