#include <itkTableLookupSampleClassifier.h>
Inheritance diagram for itk::Statistics::TableLookupSampleClassifier:
[NOHEADER] | |
virtual const char * | GetClassName () const |
Pointer | New () |
Public Types | |
typedef TableLookupSampleClassifier | Self |
typedef SampleClassifier< TSample > | Superclass |
typedef SmartPointer< Self > | Pointer |
typedef Superclass::OutputType | OutputType |
typedef Superclass::MeasurementType | MeasurementType |
typedef Superclass::MeasurementVectorType | MeasurementVectorType |
typedef Index< itkGetStaticConstMacro(MeasurementVectorSize) | CachedMeasurementVectorType ) |
typedef Image< MeasurementType, itkGetStaticConstMacro(MeasurementVectorSize) | LookupTableType ) |
typedef LookupTableType::Pointer | LookupTablePointer |
typedef ImageRegionIteratorWithIndex< LookupTableType > | LookupTableIteratorType |
typedef LookupTableType::RegionType | RegionType |
typedef LookupTableType::SizeType | SizeType |
Public Methods | |
itkStaticConstMacro (MeasurementVectorSize, unsigned int, Superclass::MeasurementVectorSize) | |
void | SetLookupTableLowerBound (MeasurementVectorType lower) |
void | SetLookupTableUpperBound (MeasurementVectorType upper) |
Protected Methods | |
TableLookupSampleClassifier () | |
virtual | ~TableLookupSampleClassifier () |
void | PrintSelf (std::ostream &os, Indent indent) const |
void | PrepareLookupTable () |
void | GenerateData () |
This classifier is identical to the SampleClassifier except that it creates a look-up table which stores a mapping between a measurement vector and its corresponding class labels. After this step, classification of target data becomes a look up operation in this table.
For creation of the look-up table, users should provdes the lower-bound upper-bound of the measurement vectors using SetLookupTableLowerBound and SetLookupTableUpperBound methods.
This classifier is meant to be used with measurement-vectors with low dimensionality (maybe 1 or 2). If users have good knowledge of the possible range of measurement-vectors, by setting the bound fit the range, they might enhance its performenace further.
The first template argument is the type of the target sample data that this classifier will assign a class label for each measurement vector. The second one is the type of a membership value calculator for each. A membership calculator represents a specific knowledge about a class. In other words, it should tell us how "likely" is that a measurement vector (pattern) belong to the class. The third argument is the type of decision rule. The main role of a decision rule is comparing the return values of the membership calculators. However, decision rule can include some prior knowledge that can improve the result.
Before you call the GenerateData method to start the classification process, you should plug in all necessary parts ( one or more membership calculators, a decision rule, and a target sample data). To plug in the decision rule, you use SetDecisionRule method, for the target sample data, SetSample method, and for the membership calculators, use AddMembershipCalculator method.
As the method name indicates, you can have more than one membership calculator. One for each classes. The order you put the membership calculator becomes the class label for the class that is represented by the membership calculator.
The classification result is a MembershipSample.
Definition at line 74 of file itkTableLookupSampleClassifier.h.
|
Lookup table related typedefs Definition at line 97 of file itkTableLookupSampleClassifier.h. |
|
Lookup table related typedefs Definition at line 100 of file itkTableLookupSampleClassifier.h. |
|
Lookup table related typedefs Definition at line 99 of file itkTableLookupSampleClassifier.h. |
|
Lookup table related typedefs Definition at line 98 of file itkTableLookupSampleClassifier.h. |
|
typedefs from TSample object Reimplemented from itk::Statistics::SampleClassifier< TSample >. Definition at line 90 of file itkTableLookupSampleClassifier.h. |
|
Reimplemented from itk::Statistics::SampleClassifier< TSample >. Definition at line 91 of file itkTableLookupSampleClassifier.h. |
|
Common typedefs for classifiers which are Inherited from SampleClassifier class Reimplemented from itk::Statistics::SampleClassifier< TSample >. Definition at line 89 of file itkTableLookupSampleClassifier.h. |
|
Reimplemented from itk::Statistics::SampleClassifier< TSample >. Definition at line 81 of file itkTableLookupSampleClassifier.h. |
|
Lookup table related typedefs Definition at line 101 of file itkTableLookupSampleClassifier.h. |
|
Standard class typedef Reimplemented from itk::Statistics::SampleClassifier< TSample >. Definition at line 79 of file itkTableLookupSampleClassifier.h. |
|
Lookup table related typedefs Definition at line 102 of file itkTableLookupSampleClassifier.h. |
|
Reimplemented from itk::Statistics::SampleClassifier< TSample >. Definition at line 80 of file itkTableLookupSampleClassifier.h. |
|
|
|
Definition at line 113 of file itkTableLookupSampleClassifier.h. References HardConnectedComponentImageFilter::GenerateData(). |
|
Starts the classification process Reimplemented from itk::Statistics::SampleClassifier< TSample >. |
|
Standard macros Reimplemented from itk::Statistics::SampleClassifier< TSample >. |
|
|
|
Standard macros Reimplemented from itk::Statistics::SampleClassifier< TSample >. |
|
|
|
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::Statistics::SampleClassifier< TSample >. |
|
sets the upper boundary for the lookup table construction |
|
sets the upper boundary for the lookup table construction |