class cv::ShapeContextDistanceExtractor
Overview
Implementation of the Shape Context descriptor and matching algorithm. More…
#include <shape_distance.hpp> class ShapeContextDistanceExtractor: public cv::ShapeDistanceExtractor { public: // methods virtual int getAngularBins() const = 0; virtual float getBendingEnergyWeight() const = 0; virtual Ptr<HistogramCostExtractor> getCostExtractor() const = 0; virtual float getImageAppearanceWeight() const = 0; virtual void getImages( OutputArray image1, OutputArray image2 ) const = 0; virtual float getInnerRadius() const = 0; virtual int getIterations() const = 0; virtual float getOuterRadius() const = 0; virtual int getRadialBins() const = 0; virtual bool getRotationInvariant() const = 0; virtual float getShapeContextWeight() const = 0; virtual float getStdDev() const = 0; virtual Ptr<ShapeTransformer> getTransformAlgorithm() const = 0; virtual void setAngularBins(int nAngularBins) = 0; virtual void setBendingEnergyWeight(float bendingEnergyWeight) = 0; virtual void setCostExtractor(Ptr<HistogramCostExtractor> comparer) = 0; virtual void setImageAppearanceWeight(float imageAppearanceWeight) = 0; virtual void setImages( InputArray image1, InputArray image2 ) = 0; virtual void setInnerRadius(float innerRadius) = 0; virtual void setIterations(int iterations) = 0; virtual void setOuterRadius(float outerRadius) = 0; virtual void setRadialBins(int nRadialBins) = 0; virtual void setRotationInvariant(bool rotationInvariant) = 0; virtual void setShapeContextWeight(float shapeContextWeight) = 0; virtual void setStdDev(float sigma) = 0; virtual void setTransformAlgorithm(Ptr<ShapeTransformer> transformer) = 0; };
Inherited Members
public: // methods virtual void clear(); virtual bool empty() const; virtual String getDefaultName() const; virtual void read(const FileNode& fn); virtual void save(const String& filename) const; virtual void write(FileStorage& fs) const; template <typename _Tp> static Ptr<_Tp> load( const String& filename, const String& objname = String() ); template <typename _Tp> static Ptr<_Tp> loadFromString( const String& strModel, const String& objname = String() ); template <typename _Tp> static Ptr<_Tp> read(const FileNode& fn); virtual float computeDistance( InputArray contour1, InputArray contour2 ) = 0; protected: // methods void writeFormat(FileStorage& fs) const;
Detailed Documentation
Implementation of the Shape Context descriptor and matching algorithm.
proposed by Belongie et al. in “Shape Matching and Object Recognition Using Shape Contexts” (PAMI 2002). This implementation is packaged in a generic scheme, in order to allow you the implementation of the common variations of the original pipeline.
Methods
virtual void setAngularBins(int nAngularBins) = 0
Establish the number of angular bins for the Shape Context Descriptor used in the shape matching pipeline.
Parameters:
| nAngularBins | The number of angular bins in the shape context descriptor. |
virtual void setBendingEnergyWeight(float bendingEnergyWeight) = 0
Set the weight of the Bending Energy in the final value of the shape distance. The bending energy definition depends on what transformation is being used to align the shapes. The final value of the shape distance is a user-defined linear combination of the shape context distance, an image appearance distance, and a bending energy.
Parameters:
| bendingEnergyWeight | The weight of the Bending Energy in the final distance value. |
virtual void setCostExtractor(Ptr<HistogramCostExtractor> comparer) = 0
Set the algorithm used for building the shape context descriptor cost matrix.
Parameters:
| comparer | Smart pointer to a HistogramCostExtractor, an algorithm that defines the cost matrix between descriptors. |
virtual void setImageAppearanceWeight(float imageAppearanceWeight) = 0
Set the weight of the Image Appearance cost in the final value of the shape distance. The image appearance cost is defined as the sum of squared brightness differences in Gaussian windows around corresponding image points. The final value of the shape distance is a user-defined linear combination of the shape context distance, an image appearance distance, and a bending energy. If this value is set to a number different from 0, is mandatory to set the images that correspond to each shape.
Parameters:
| imageAppearanceWeight | The weight of the appearance cost in the final distance value. |
virtual void setImages( InputArray image1, InputArray image2 ) = 0
Set the images that correspond to each shape. This images are used in the calculation of the Image Appearance cost.
Parameters:
| image1 | Image corresponding to the shape defined by contours1. |
| image2 | Image corresponding to the shape defined by contours2. |
virtual void setInnerRadius(float innerRadius) = 0
Set the inner radius of the shape context descriptor.
Parameters:
| innerRadius | The value of the inner radius. |
virtual void setOuterRadius(float outerRadius) = 0
Set the outer radius of the shape context descriptor.
Parameters:
| outerRadius | The value of the outer radius. |
virtual void setRadialBins(int nRadialBins) = 0
Establish the number of radial bins for the Shape Context Descriptor used in the shape matching pipeline.
Parameters:
| nRadialBins | The number of radial bins in the shape context descriptor. |
virtual void setShapeContextWeight(float shapeContextWeight) = 0
Set the weight of the shape context distance in the final value of the shape distance. The shape context distance between two shapes is defined as the symmetric sum of shape context matching costs over best matching points. The final value of the shape distance is a user-defined linear combination of the shape context distance, an image appearance distance, and a bending energy.
Parameters:
| shapeContextWeight | The weight of the shape context distance in the final distance value. |
virtual void setStdDev(float sigma) = 0
Set the value of the standard deviation for the Gaussian window for the image appearance cost.
Parameters:
| sigma | Standard Deviation. |
virtual void setTransformAlgorithm(Ptr<ShapeTransformer> transformer) = 0
Set the algorithm used for aligning the shapes.
Parameters:
| transformer | Smart pointer to a ShapeTransformer, an algorithm that defines the aligning transformation. |