te::rp::ClassifierMaxLikelihoodStrategy::Parameters Class Reference

Classifier Parameters. More...

#include <ClassifierMaxLikelihoodStrategy.h>

Inheritance diagram for te::rp::ClassifierMaxLikelihoodStrategy::Parameters:

Public Types
typedef unsigned int	ClassIDT
	Class ID type definition (zero means invalid ID). More...
typedef std::vector< ClassSampleT >	ClassSamplesContainerT
	Class samples container type definition. More...
typedef std::vector< double >	ClassSampleT
	Class sample type definition. More...
typedef std::map< ClassIDT, ClassSamplesContainerT >	MClassesSamplesCT
	Multi-classes samples container type definition. More...
typedef boost::shared_ptr< MClassesSamplesCT >	MClassesSamplesCTPtr
	A shared pointer to a multi classes samples container type definition. More...

Public Member Functions
AbstractParameters *	clone () const
	Create a clone copy of this instance. More...
const Parameters &	operator= (const Parameters &params)
	Parameters ()
void	reset ()
	Clear all internal allocated resources and reset the parameters instance to its initial state. More...
bool	serialize (AlgorithmParametersSerializer &serializer) const
	Returns a parameter serialization object. More...
	~Parameters ()

Public Attributes
bool	m_applyICM
	Apply a Iterated conditional modes iteractive filtering (default:false);. More...
float	m_icmChangesThreshold
	Minimum porcentage of number of pixel that changes from one iteration to the other [valid values range:1-100, default:10)]. More...
unsigned int	m_icmMaxIterations
	Maximum number of ICM iterations to perform (default 10). More...
MClassesSamplesCTPtr	m_trainSamplesPtr
	A shared pointer to a always-valid structure where trainning samples are stored. More...

Detailed Description

Classifier Parameters.

Definition at line 64 of file ClassifierMaxLikelihoodStrategy.h.

Member Typedef Documentation

ClassIDT

typedef unsigned int te::rp::ClassifierMaxLikelihoodStrategy::Parameters::ClassIDT

Class ID type definition (zero means invalid ID).

Definition at line 68 of file ClassifierMaxLikelihoodStrategy.h.

ClassSamplesContainerT

typedef std::vector< ClassSampleT > te::rp::ClassifierMaxLikelihoodStrategy::Parameters::ClassSamplesContainerT

Class samples container type definition.

Definition at line 72 of file ClassifierMaxLikelihoodStrategy.h.

ClassSampleT

typedef std::vector< double > te::rp::ClassifierMaxLikelihoodStrategy::Parameters::ClassSampleT

Class sample type definition.

Definition at line 70 of file ClassifierMaxLikelihoodStrategy.h.

MClassesSamplesCT

typedef std::map< ClassIDT, ClassSamplesContainerT > te::rp::ClassifierMaxLikelihoodStrategy::Parameters::MClassesSamplesCT

Multi-classes samples container type definition.

Definition at line 74 of file ClassifierMaxLikelihoodStrategy.h.

MClassesSamplesCTPtr

typedef boost::shared_ptr<MClassesSamplesCT> te::rp::ClassifierMaxLikelihoodStrategy::Parameters::MClassesSamplesCTPtr

A shared pointer to a multi classes samples container type definition.

Definition at line 76 of file ClassifierMaxLikelihoodStrategy.h.

Constructor & Destructor Documentation

Parameters()

te::rp::ClassifierMaxLikelihoodStrategy::Parameters::Parameters

(

)

~Parameters()

te::rp::ClassifierMaxLikelihoodStrategy::Parameters::~Parameters

(

)

Member Function Documentation

clone()

AbstractParameters * te::rp::ClassifierMaxLikelihoodStrategy::Parameters::clone ( ) const

virtual

Create a clone copy of this instance.

Returns

A clone copy of this instance.

Note

The caller will take the ownership of the returned pointer.

Implements te::common::AbstractParameters.

operator=()

const Parameters & te::rp::ClassifierMaxLikelihoodStrategy::Parameters::operator=

(

const Parameters &

params

)

reset()

void te::rp::ClassifierMaxLikelihoodStrategy::Parameters::reset ( )

virtual

Clear all internal allocated resources and reset the parameters instance to its initial state.

Implements te::common::AbstractParameters.

serialize()

bool te::rp::ClassifierMaxLikelihoodStrategy::Parameters::serialize ( AlgorithmParametersSerializer &  serializer ) const

virtual

Returns a parameter serialization object.

Parameters

serializer

The output serialization object.

Returns

Returns true if ok, false on erros.

Reimplemented from te::rp::ClassifierStrategyParameters.

Member Data Documentation

m_applyICM

bool te::rp::ClassifierMaxLikelihoodStrategy::Parameters::m_applyICM

Apply a Iterated conditional modes iteractive filtering (default:false);.

Definition at line 80 of file ClassifierMaxLikelihoodStrategy.h.

m_icmChangesThreshold

float te::rp::ClassifierMaxLikelihoodStrategy::Parameters::m_icmChangesThreshold

Minimum porcentage of number of pixel that changes from one iteration to the other [valid values range:1-100, default:10)].

Definition at line 82 of file ClassifierMaxLikelihoodStrategy.h.

m_icmMaxIterations

unsigned int te::rp::ClassifierMaxLikelihoodStrategy::Parameters::m_icmMaxIterations

Maximum number of ICM iterations to perform (default 10).

Definition at line 84 of file ClassifierMaxLikelihoodStrategy.h.

m_trainSamplesPtr

MClassesSamplesCTPtr te::rp::ClassifierMaxLikelihoodStrategy::Parameters::m_trainSamplesPtr

A shared pointer to a always-valid structure where trainning samples are stored.

Definition at line 78 of file ClassifierMaxLikelihoodStrategy.h.

---

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

• /home/castejon/castejon_files/develop/terralib5/git_release/src/terralib/rp/ClassifierMaxLikelihoodStrategy.h