ITK  4.1.0
Insight Segmentation and Registration Toolkit
Public Types | Public Member Functions | Static Public Member Functions | Protected Member Functions | Private Member Functions | Private Attributes
itk::Statistics::MaximumRatioDecisionRule Class Reference

#include <itkMaximumRatioDecisionRule.h>

+ Inheritance diagram for itk::Statistics::MaximumRatioDecisionRule:
+ Collaboration diagram for itk::Statistics::MaximumRatioDecisionRule:

List of all members.

Public Types

typedef
Superclass::ClassIdentifierType 
ClassIdentifierType
typedef
Superclass::MembershipValueType 
MembershipValueType
typedef
Superclass::MembershipVectorType 
MembershipVectorType
typedef SmartPointer< SelfPointer
typedef MembershipValueType PriorProbabilityValueType
typedef
PriorProbabilityVectorType::size_type 
PriorProbabilityVectorSizeType
typedef std::vector
< PriorProbabilityValueType
PriorProbabilityVectorType
typedef MaximumRatioDecisionRule Self
typedef DecisionRule Superclass

Public Member Functions

virtual ::itk::LightObject::Pointer CreateAnother (void) const
virtual ClassIdentifierType Evaluate (const MembershipVectorType &discriminantScores) const
virtual const char * GetNameOfClass () const
virtual const
PriorProbabilityVectorType
GetPriorProbabilities ()
void SetPriorProbabilities (const PriorProbabilityVectorType &p)

Static Public Member Functions

static Pointer New ()

Protected Member Functions

 MaximumRatioDecisionRule ()
void PrintSelf (std::ostream &os, Indent indent) const
virtual ~MaximumRatioDecisionRule ()

Private Member Functions

 MaximumRatioDecisionRule (const Self &)
void operator= (const Self &)

Private Attributes

PriorProbabilityVectorType m_PriorProbabilities

Detailed Description

A decision rule that operates as a frequentist's approximation to Bayes rule.

MaximumRatioDecisionRule returns the class label using a Bayesian style decision rule. The discriminant scores are evaluated in the context of class priors. If the discriminant scores are actual conditional probabilites (likelihoods) and the class priors are actual a priori class probabilities, then this decision rule operates as Bayes rule, returning the class $i$ if $p(x|i) p(i) > p(x|j) p(j)$ for all class $j$. The discriminant scores and priors are not required to be true probabilities.

This class is named the MaximumRatioDecisionRule as it can be implemented as returning the class $i$ if $\frac{p(x|i)}{p(x|j)} > \frac{p(j)}{p(i)}$ for all class $j$.

A priori values need to be set before calling the Evaluate method. If they are not set, a uniform prior is assumed.

See also:
MaximumDecisionRule, MinimumDecisionRule

Definition at line 57 of file itkMaximumRatioDecisionRule.h.


Member Typedef Documentation

Types for class identifiers.

Reimplemented from itk::Statistics::DecisionRule.

Definition at line 76 of file itkMaximumRatioDecisionRule.h.

Types for discriminant values and vectors.

Reimplemented from itk::Statistics::DecisionRule.

Definition at line 69 of file itkMaximumRatioDecisionRule.h.

Reimplemented from itk::Statistics::DecisionRule.

Definition at line 73 of file itkMaximumRatioDecisionRule.h.

Reimplemented from itk::Statistics::DecisionRule.

Definition at line 63 of file itkMaximumRatioDecisionRule.h.

Types for priors and values

Definition at line 79 of file itkMaximumRatioDecisionRule.h.

Definition at line 81 of file itkMaximumRatioDecisionRule.h.

Definition at line 80 of file itkMaximumRatioDecisionRule.h.

Standard class typedefs

Reimplemented from itk::Statistics::DecisionRule.

Definition at line 61 of file itkMaximumRatioDecisionRule.h.

Reimplemented from itk::Statistics::DecisionRule.

Definition at line 62 of file itkMaximumRatioDecisionRule.h.


Constructor & Destructor Documentation

Definition at line 103 of file itkMaximumRatioDecisionRule.h.


Member Function Documentation

virtual::itk::LightObject::Pointer itk::Statistics::MaximumRatioDecisionRule::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::Object.

virtual ClassIdentifierType itk::Statistics::MaximumRatioDecisionRule::Evaluate ( const MembershipVectorType discriminantScores) const [virtual]

Evaluate the decision rule $p(x|i) p(i) > p(x|j) p(j)$. Prior probabilities need to be set before calling Evaluate() using the SetPriorProbabilities() method (otherwise a uniform prior is assumed). Parameter to Evaluate() is the discriminant score in the form of a likelihood $p(x|i)$.

Implements itk::Statistics::DecisionRule.

virtual const char* itk::Statistics::MaximumRatioDecisionRule::GetNameOfClass ( ) const [virtual]

Run-time type information (and related methods)

Reimplemented from itk::Statistics::DecisionRule.

Get the prior probabilities.

Standard New() method support

Reimplemented from itk::Object.

void itk::Statistics::MaximumRatioDecisionRule::operator= ( const Self ) [private]

Mutex lock to protect modification to the reference count

Reimplemented from itk::Statistics::DecisionRule.

void itk::Statistics::MaximumRatioDecisionRule::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.

Set the prior probabilities used in evaluating $p(x|i) p(i) > p(x|j) p(j)$. The likelihoods are set using the Evaluate() method. SetPriorProbabilities needs to be called before Evaluate(). If not set, assumes a uniform prior.


Member Data Documentation

Definition at line 110 of file itkMaximumRatioDecisionRule.h.


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