ITK  4.4.0
Insight Segmentation and Registration Toolkit
Public Types | Public Member Functions | Protected Member Functions | Protected Attributes | Private Member Functions | List of all members
itk::ObjectToObjectOptimizerBase Class Reference

#include <itkObjectToObjectOptimizerBase.h>

+ Inheritance diagram for itk::ObjectToObjectOptimizerBase:
+ Collaboration diagram for itk::ObjectToObjectOptimizerBase:

Detailed Description

Abstract base for object-to-object optimizers.

The goal of this optimizer hierarchy is to work with metrics of any type, i.e. working with any kind of object, such as image or point-set.

Transform parameters are not manipulated directly. Instead, the optimizer retrieves the metric derivative from the metric, modifies the derivative as required, then passes it back to the metric as an update. The metric then processes it as appropriate, typically by passing it to its transform that is being optimized.

SetScales() allows setting of a per-local-parameter scaling array. The gradient of each local parameter is divided by the corresponding scale. If unset, the m_Scales array will be initialized to all 1's. Note that when used with transforms with local support, these scales correspond to each local parameter, and not to each parameter. For example, in a DisplacementFieldTransform of dimensionality N, the Scales is size N, with each element corresponding to a dimension within the transform's displacement field, and is applied to each vector in the displacement field.

SetWeights() allows setting of a per-local-parameter weighting array. If unset, the weights are treated as identity. Weights are multiplied by the gradient at the same time scaling is applied. Weights are different than scales, and may be used, for example, to easily mask out a particular parameter during optimzation to hold it constant. Or they may be used to apply another kind of prior knowledge.

Threading of some optimizer operations may be handled within derived classes, for example in GradientDescentOptimizer.

Derived classes must override StartOptimization, and then call this base class version to perform common initializations.

Definition at line 67 of file itkObjectToObjectOptimizerBase.h.

Public Types

typedef SmartPointer< const SelfConstPointer
typedef MetricType::MeasureType MeasureType
typedef ObjectToObjectMetricBase MetricType
typedef MetricType::Pointer MetricTypePointer
typedef OptimizerParameters
< double > 
typedef SmartPointer< SelfPointer
typedef OptimizerParameters
< double > 
typedef ObjectToObjectOptimizerBase Self
typedef Object Superclass
- Public Types inherited from itk::Object
typedef SmartPointer< const SelfConstPointer
typedef SmartPointer< SelfPointer
typedef Object Self
typedef LightObject Superclass
- Public Types inherited from itk::LightObject
typedef SmartPointer< const SelfConstPointer
typedef SmartPointer< SelfPointer
typedef LightObject Self

Public Member Functions

virtual const MeasureTypeGetCurrentMetricValue ()
const ParametersTypeGetCurrentPosition ()
virtual const char * GetNameOfClass () const
virtual const ThreadIdTypeGetNumberOfThreads ()
virtual const ScalesTypeGetScales ()
virtual const bool & GetScalesAreIdentity ()
const MeasureTypeGetValue ()
virtual const ScalesTypeGetWeights ()
virtual const bool & GetWeightsAreIdentity ()
virtual void SetNumberOfThreads (ThreadIdType number)
virtual void SetScales (ScalesType _arg)
virtual void SetWeights (ScalesType _arg)
virtual void StartOptimization (bool doOnlyInitialization=false)
virtual void SetMetric (MetricType *_arg)
virtual MetricTypeGetModifiableMetric ()
virtual const MetricTypeGetMetric () const
- Public Member Functions inherited from itk::Object
unsigned long AddObserver (const EventObject &event, Command *)
unsigned long AddObserver (const EventObject &event, Command *) const
virtual LightObject::Pointer CreateAnother () const
virtual void DebugOff () const
virtual void DebugOn () const
CommandGetCommand (unsigned long tag)
bool GetDebug () const
MetaDataDictionaryGetMetaDataDictionary (void)
const MetaDataDictionaryGetMetaDataDictionary (void) const
virtual ModifiedTimeType GetMTime () const
virtual const TimeStampGetTimeStamp () const
bool HasObserver (const EventObject &event) const
void InvokeEvent (const EventObject &)
void InvokeEvent (const EventObject &) const
virtual void Modified () const
virtual void Register () const
void RemoveAllObservers ()
void RemoveObserver (unsigned long tag)
void SetDebug (bool debugFlag) const
void SetMetaDataDictionary (const MetaDataDictionary &rhs)
virtual void SetReferenceCount (int)
virtual void UnRegister () const
- Public Member Functions inherited from itk::LightObject
virtual void Delete ()
virtual int GetReferenceCount () const
 itkCloneMacro (Self)
void Print (std::ostream &os, Indent indent=0) const

Protected Member Functions

virtual void PrintSelf (std::ostream &os, Indent indent) const
 ObjectToObjectOptimizerBase ()
virtual ~ObjectToObjectOptimizerBase ()
- Protected Member Functions inherited from itk::Object
 Object ()
bool PrintObservers (std::ostream &os, Indent indent) const
virtual void SetTimeStamp (const TimeStamp &time)
virtual ~Object ()
- Protected Member Functions inherited from itk::LightObject
virtual LightObject::Pointer InternalClone () const
 LightObject ()
virtual void PrintHeader (std::ostream &os, Indent indent) const
virtual void PrintTrailer (std::ostream &os, Indent indent) const
virtual ~LightObject ()

Protected Attributes

MeasureType m_CurrentMetricValue
MetricTypePointer m_Metric
ThreadIdType m_NumberOfThreads
ScalesType m_Scales
bool m_ScalesAreIdentity
ScalesType m_Weights
bool m_WeightsAreIdentity

Private Member Functions

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

Additional Inherited Members

- Static Public Member Functions inherited from itk::Object
static bool GetGlobalWarningDisplay ()
static void GlobalWarningDisplayOff ()
static void GlobalWarningDisplayOn ()
static Pointer New ()
static void SetGlobalWarningDisplay (bool flag)
- Protected Types inherited from itk::LightObject
typedef int InternalReferenceCountType

Member Typedef Documentation

Definition at line 74 of file itkObjectToObjectOptimizerBase.h.

Internal computation value type

Definition at line 96 of file itkObjectToObjectOptimizerBase.h.

Measure type

Definition at line 93 of file itkObjectToObjectOptimizerBase.h.

Metric function type

Definition at line 86 of file itkObjectToObjectOptimizerBase.h.

Definition at line 87 of file itkObjectToObjectOptimizerBase.h.

Number of parameters type

Definition at line 90 of file itkObjectToObjectOptimizerBase.h.

Parameters type.

Definition at line 83 of file itkObjectToObjectOptimizerBase.h.

Definition at line 73 of file itkObjectToObjectOptimizerBase.h.

Scale type.

Definition at line 77 of file itkObjectToObjectOptimizerBase.h.

Standard class typedefs.

Definition at line 71 of file itkObjectToObjectOptimizerBase.h.

Definition at line 72 of file itkObjectToObjectOptimizerBase.h.

Constructor & Destructor Documentation

itk::ObjectToObjectOptimizerBase::ObjectToObjectOptimizerBase ( )

Default constructor

virtual itk::ObjectToObjectOptimizerBase::~ObjectToObjectOptimizerBase ( )

Default constructor

itk::ObjectToObjectOptimizerBase::ObjectToObjectOptimizerBase ( const Self )

Member Function Documentation

virtual const MeasureType& itk::ObjectToObjectOptimizerBase::GetCurrentMetricValue ( )

Accessor for metric value. Returns the value stored in m_CurrentMetricValue from the most recent call to evaluate the metric.

const ParametersType& itk::ObjectToObjectOptimizerBase::GetCurrentPosition ( )

Get a reference to the current position of the optimization. This returns the parameters from the assigned metric, since the optimizer itself does not store a position.

virtual const MetricType* itk::ObjectToObjectOptimizerBase::GetMetric ( ) const

Accessors for Metric

virtual MetricType* itk::ObjectToObjectOptimizerBase::GetModifiableMetric ( )

Accessors for Metric

virtual const char* itk::ObjectToObjectOptimizerBase::GetNameOfClass ( ) const
virtual const ThreadIdType& itk::ObjectToObjectOptimizerBase::GetNumberOfThreads ( )

Get the number of threads set to be used.

virtual const ScalesType& itk::ObjectToObjectOptimizerBase::GetScales ( )

Get current parameters scaling.

virtual const bool& itk::ObjectToObjectOptimizerBase::GetScalesAreIdentity ( )

Get whether scales are identity. Cannot be set

const MeasureType& itk::ObjectToObjectOptimizerBase::GetValue ( )

Deprecated accessor for currently stored metric value for use by classes that support both v4 and v3 optimizers.

See Also
virtual const ScalesType& itk::ObjectToObjectOptimizerBase::GetWeights ( )

Get current parameters weights. This will return an empty array when weights have not been set by user.

virtual const bool& itk::ObjectToObjectOptimizerBase::GetWeightsAreIdentity ( )

Get whether weights are identity. Cannot be set

void itk::ObjectToObjectOptimizerBase::operator= ( const Self )
virtual void itk::ObjectToObjectOptimizerBase::PrintSelf ( std::ostream &  os,
Indent  indent 
) const

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::GradientDescentOptimizerv4, itk::QuasiNewtonOptimizerv4, itk::GradientDescentOptimizerBasev4, itk::MultiStartOptimizerv4, itk::MultiGradientOptimizerv4, itk::GradientDescentLineSearchOptimizerv4, and itk::ConjugateGradientLineSearchOptimizerv4.

virtual void itk::ObjectToObjectOptimizerBase::SetMetric ( MetricType _arg)

Accessors for Metric

virtual void itk::ObjectToObjectOptimizerBase::SetNumberOfThreads ( ThreadIdType  number)

Set the number of threads to use when threading. The default is the global default number of threads returned from itkMultiThreader.

virtual void itk::ObjectToObjectOptimizerBase::SetScales ( ScalesType  _arg)

Set current parameters scaling.

virtual void itk::ObjectToObjectOptimizerBase::SetWeights ( ScalesType  _arg)

Set current parameters weights.

virtual void itk::ObjectToObjectOptimizerBase::StartOptimization ( bool  doOnlyInitialization = false)

Run the optimization.

doOnlyInitializationThis is false by default. It should only be set to true for special cases when the class should be initialized to perform optimization, but no optimization should be run. For example, itkMultiGradientOptimizerv4 needs to do this.
Derived classes must override and call this superclass method, then perform any additional initialization before performing optimization.

Reimplemented in itk::GradientDescentOptimizerv4, itk::MultiStartOptimizerv4, itk::MultiGradientOptimizerv4, itk::QuasiNewtonOptimizerv4, and itk::ConjugateGradientLineSearchOptimizerv4.

Member Data Documentation

MeasureType itk::ObjectToObjectOptimizerBase::m_CurrentMetricValue

Metric measure value at a given iteration, as most recently evaluated.

Definition at line 167 of file itkObjectToObjectOptimizerBase.h.

MetricTypePointer itk::ObjectToObjectOptimizerBase::m_Metric

Definition at line 163 of file itkObjectToObjectOptimizerBase.h.

ThreadIdType itk::ObjectToObjectOptimizerBase::m_NumberOfThreads

Definition at line 164 of file itkObjectToObjectOptimizerBase.h.

ScalesType itk::ObjectToObjectOptimizerBase::m_Scales

Scales. Size is expected to be == metric->GetNumberOfLocalParameters(). See the main documentation for more details.

Definition at line 171 of file itkObjectToObjectOptimizerBase.h.

bool itk::ObjectToObjectOptimizerBase::m_ScalesAreIdentity

Flag to avoid unnecessary arithmetic when scales are identity.

Definition at line 180 of file itkObjectToObjectOptimizerBase.h.

ScalesType itk::ObjectToObjectOptimizerBase::m_Weights

Parameter weights. These are applied to local parameters, at the same time as scales. See main documentation. If not set by user, the array remains empty and treated as identity to simplify the reuse of an optimizer with transforms with different numbers of parameters.

Definition at line 177 of file itkObjectToObjectOptimizerBase.h.

bool itk::ObjectToObjectOptimizerBase::m_WeightsAreIdentity

Flag to avoid unnecessary arithmetic when weights are identity.

Definition at line 183 of file itkObjectToObjectOptimizerBase.h.

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