C# Класс Emgu.CV.Flann.Index

Flann index

Наследование: Emgu.Util.UnmanagedObject

Открытые методы

Метод	Описание
Index ( Matrix values ) : System	Create a linear flann index
Index ( Matrix values, float targetPrecision, float buildWeight, float memoryWeight, float sampleFraction ) : System	Create an auto-tuned flann index
Index ( Matrix values, int numberOfKDTrees ) : System	Create a flann index using multiple KDTrees
Index ( Matrix values, int branching, int iterations, CenterInitType centersInitType, float cbIndex ) : System	Create a flann index using Kmeans
Index ( Matrix values, int numberOfKDTrees, int branching, int iterations, CenterInitType centersInitType, float cbIndex ) : System	Create a flann index using a composition of KDTreee and KMeans tree
KnnSearch ( Matrix queries, Matrix indices, Matrix squareDistances, int knn, int checks ) : void	Perform k-nearest-neighbours (KNN) search
RadiusSearch ( Matrix queries, Matrix indices, Matrix squareDistances, float radius, int checks ) : int	Performs a radius nearest neighbor search for multiple query points

Метод	Описание
DisposeObject ( ) : void	Release the unmanaged memory associated with this Flann Index

Release the unmanaged memory associated with this Flann Index

protected DisposeObject ( ) : void
Результат	void

Create a linear flann index

public Index ( Matrix values ) : System
values	Matrix	A row by row matrix of descriptors
Результат	System

Create an auto-tuned flann index

public Index ( Matrix values, float targetPrecision, float buildWeight, float memoryWeight, float sampleFraction ) : System
values	Matrix	A row by row matrix of descriptors
targetPrecision	float	Precision desired, use 0.9 if not sure
buildWeight	float	build tree time weighting factor, use 0.01 if not sure
memoryWeight	float	index memory weighting factor, use 0 if not sure
sampleFraction	float	what fraction of the dataset to use for autotuning, use 0.1 if not sure
Результат	System

Create a flann index using multiple KDTrees

public Index ( Matrix values, int numberOfKDTrees ) : System
values	Matrix	A row by row matrix of descriptors
numberOfKDTrees	int	The number of KDTrees to be used
Результат	System

Create a flann index using Kmeans

public Index ( Matrix values, int branching, int iterations, CenterInitType centersInitType, float cbIndex ) : System
values	Matrix	A row by row matrix of descriptors
branching	int	Branching factor (for kmeans tree), use 32 for default
iterations	int	Max iterations to perform in one kmeans clustering (kmeans tree), use 11 for deafault
centersInitType	CenterInitType	Algorithm used for picking the initial cluster centers for kmeans tree, use RANDOM for default
cbIndex	float	Cluster boundary index. Used when searching the kmeans tree. Use 0.2 for default
Результат	System

Create a flann index using a composition of KDTreee and KMeans tree

public Index ( Matrix values, int numberOfKDTrees, int branching, int iterations, CenterInitType centersInitType, float cbIndex ) : System
values	Matrix	A row by row matrix of descriptors
numberOfKDTrees	int	The number of KDTrees to be used
branching	int	Branching factor (for kmeans tree), use 32 for default
iterations	int	Max iterations to perform in one kmeans clustering (kmeans tree), use 11 for deafault
centersInitType	CenterInitType	Algorithm used for picking the initial cluster centers for kmeans tree, use RANDOM for default
cbIndex	float	Cluster boundary index. Used when searching the kmeans tree. Use 0.2 for default
Результат	System

Perform k-nearest-neighbours (KNN) search

public KnnSearch ( Matrix queries, Matrix indices, Matrix squareDistances, int knn, int checks ) : void
queries	Matrix	A row by row matrix of descriptors to be query for nearest neighbours
indices	Matrix	The result of the indices of the k-nearest neighbours
squareDistances	Matrix	The square of the Eculidean distance between the neighbours
knn	int	Number of nearest neighbors to search for
checks	int	The number of times the tree(s) in the index should be recursively traversed. A /// higher value for this parameter would give better search precision, but also take more /// time. If automatic configuration was used when the index was created, the number of /// checks required to achieve the specified precision was also computed, in which case /// this parameter is ignored
Результат	void

Performs a radius nearest neighbor search for multiple query points

public RadiusSearch ( Matrix queries, Matrix indices, Matrix squareDistances, float radius, int checks ) : int
queries	Matrix	The query points, one per row
indices	Matrix	Indices of the nearest neighbors found
squareDistances	Matrix	The square of the Eculidean distance between the neighbours
radius	float	The search radius
checks	int	The number of times the tree(s) in the index should be recursively traversed. A /// higher value for this parameter would give better search precision, but also take more /// time. If automatic configuration was used when the index was created, the number of /// checks required to achieve the specified precision was also computed, in which case /// this parameter is ignored
Результат	int