LPV Machine learning Library, provides classes for classification/detection/inspection tasks that involves machine learning technology. More...

Data Structures
interface	ILClassifier
	This interface provide functionalities of the machine learning based image classification. More...

interface	ILFeature
	This interface provide functionalities of the image features, which is used for feature extraction from image in machine learning technology. More...

interface	ILModel
	This interface provide functionalities of the machine learning model. More...

interface	ILParameterized
	This interface provide parameterized interface for the machine learning feature and model objects. More...

interface	ILSample
	This interface represents a sample instance, which usually contains an image and an optional region object. More...

interface	ILSampleDatabase
	This interface represents a sample database for training and testing sample organization and management. More...

class	LClassifier

class	LFeature

class	LModel

class	LSample

class	LSampleDatabase

Enumerations
enum	LPVConductivity { LPVConductPMG1 = 0 , LPVConductPMG2 = 1 , LPVConductWeickert = 2 , LPVConductCharbonnier = 3 }
	This enumeration represents the type of the conductivity coefficient used in AKAZE feature. More...

enum	LPVFeatureType { LPVFeatureReserved = 0 , LPVFeatureORB = 1 , LPVFeatureHOG = 2 , LPVFeatureAKAZE = 3 }
	This enumeration represents the type of feature extracted from the image. More...

enum	LPVKNNDistance { LPVKNNDistanceEuclidean = 0 , LPVKNNDistanceManhattan = 1 , LPVKNNDistanceMax = 2 , LPVKNNDistanceHistIntersection = 3 , LPVKNNDistanceHellinger = 4 , LPVKNNDistanceChiSquare = 5 }
	This enumeration represents the distance calculation method used in KNN model for measuring the similarity between samples. More...

enum	LPVKNNMatchAlgo { LPVKNNMatchBruteForce = 0 , LPVKNNMatchKDTree = 1 , LPVKNNMatchKmeans = 2 , LPVKNNMatchComposite = 3 , LPVKNNMatchHierarchical = 4 , LPVKNNMatchAuto = 5 }
	This enumeration represents the algorithm for the nearest searching used in KNN model. More...

enum	LPVKNNWeight { LPVKNNWeightByDistance = 0 , LPVKNNWeightUniformed = 1 }
	This enumeration represents the weighting method for KNN model to voting the final result label from K neighbors. More...

enum	LPVModelType { LPVModelReserved = 0 , LPVModelKNN = 1 , LPVModelSVM = 2 }
	This enumeration represents the type of machine learning model. More...

enum	LPVSampleType { LPVSampleTrain = 1 , LPVSampleTest = 2 }
	This enumeration represents the type of the sample instance. More...

enum	LPVSVMKernel { LPVSVMKernelLinear = 0 , LPVSVMKernelPolynomial = 1 , LPVSVMKernelRBF = 2 , LPVSVMKernelSigmoid = 3 }
	This enumeration represents the type of the kernel used in the SVM model. More...

enum	LPVSVMType { LPVSVMTypeCSupport = 0 , LPVSVMTypeNuSupport = 1 }
	This enumeration represents the type of the SVM model. More...

Detailed Description

LPV Machine learning Library, provides classes for classification/detection/inspection tasks that involves machine learning technology.

This library provides classes for machine learning tasks. To include the definitions of the library's classes, use the following directive:

Example Code

C++ Snippets: MLClassification.cpp
C# Snippets: MLClassification.cs

Enumeration Type Documentation

◆ LPVConductivity

enum LPVConductivity

This enumeration represents the type of the conductivity coefficient used in AKAZE feature.

For all the equations below, \( dL \) is the first order image derivative, \( k \) is the contrast factor.

Enumerator
LPVConductPMG1	Perona & Malik G1: \( \exp{(-dL^2 / k^2)} \)
LPVConductPMG2	Perona & Malik G2: \( 1 / (1 + dL^2 / k^2) \)
LPVConductWeickert	Weickert: \( -3.315 / (dL^2 / k^2)^4 \)
LPVConductCharbonnier	Charbonnier: \( 1 / \sqrt{(1 + dL^2 / k^2)} \)

◆ LPVFeatureType

enum LPVFeatureType

This enumeration represents the type of feature extracted from the image.

Enumerator
LPVFeatureReserved	Not-initialized feature.
LPVFeatureORB	ORB(Oriented FAST and Rotated BRIEF) feature.
LPVFeatureHOG	HOG(Histogram of Oriented Gradients) feature.
LPVFeatureAKAZE	AKAZE(Accelerated-KAZE) feature.

◆ LPVKNNDistance

enum LPVKNNDistance

This enumeration represents the distance calculation method used in KNN model for measuring the similarity between samples.

For all the equations below, x and y is the two feature vectors of the two samples to compare, i is for the i^th value.

Enumerator
LPVKNNDistanceEuclidean	Euclidean(L2): \( \sum{(x_i - y_i)^2} \)
LPVKNNDistanceManhattan	Manhattan(L1): \( \sum{\|x_i - y_i\|} \)
LPVKNNDistanceMax	Maximum: \( \max{\|x_i - y_i\|} \) Only works with BruteForce or K-Means or Hierarchical matching
LPVKNNDistanceHistIntersection	Histogram Intersection: \( \sum{\min{(x_i, y_i)}} \)
LPVKNNDistanceHellinger	Hellinger: \( \sum{(\sqrt{x_i} - \sqrt{y_i})^2} \)
LPVKNNDistanceChiSquare	Chi-Square: \( \sum{((x_i - y_i)^2/(x_i + y_i))} \text{ for } x_i + y_i > 0 \)

◆ LPVKNNMatchAlgo

enum LPVKNNMatchAlgo

This enumeration represents the algorithm for the nearest searching used in KNN model.

Enumerator
LPVKNNMatchBruteForce	The naive Brute-Force algorithm, aka. searching one-by-one, used for small dataset.
LPVKNNMatchKDTree	The samples are organized and searched in several KD-Trees.
LPVKNNMatchKmeans	The samples are organized via K-Means algorithm.
LPVKNNMatchComposite	Composite with KD-Tree and K-Means
LPVKNNMatchHierarchical	The samples are organized via hierarchical clustering
LPVKNNMatchAuto	The samples are organized via auto-tuned indexing

◆ LPVKNNWeight

enum LPVKNNWeight

This enumeration represents the weighting method for KNN model to voting the final result label from K neighbors.

Enumerator
LPVKNNWeightByDistance	The voting is weight by the similarity distance between the sample and prediction feature.
LPVKNNWeightUniformed	Each neighbor has the same weight

◆ LPVModelType

enum LPVModelType

This enumeration represents the type of machine learning model.

Enumerator
LPVModelReserved	Not-initialized model.
LPVModelKNN	KNN(K-Nearest Neighbors) model.
LPVModelSVM	SVM(Support Vector Machine) model.

◆ LPVSampleType

enum LPVSampleType

This enumeration represents the type of the sample instance.

Enumerator
LPVSampleTrain	The sample instance in train set which is used for training
LPVSampleTest	The sample instance in test set which is used for testing

◆ LPVSVMKernel

enum LPVSVMKernel

This enumeration represents the type of the kernel used in the SVM model.

For all the equations below, x and y is the two feature vectors of the two samples to compare, g, c and n are numeric parameters.

Enumerator
LPVSVMKernelLinear	Linear: \( x^Ty \)
LPVSVMKernelPolynomial	Polynomial: \( (g \cdot x^Ty + c)^n \)
LPVSVMKernelRBF	RBF: \( \exp{(-g \cdot \|x - y\|^2)} \)
LPVSVMKernelSigmoid	Sigmoid: \( \tanh{(g \cdot x^Ty + c)} \)

◆ LPVSVMType

enum LPVSVMType

This enumeration represents the type of the SVM model.

Enumerator
LPVSVMTypeCSupport	C-SVM, C-support vector machine, uses the C parameter to optimize the strength of the regularization.
LPVSVMTypeNuSupport	Nu-SVM, Nu-support vector machine, uses the Nu parameter to control the number of support vectors

Data Structures

Enumerations