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

Namespace: Emgu.CV.Flann
Assembly: Emgu.CV (in Emgu.CV.dll) Version: 2.2.1.1150 (2.2.1.1150)

Syntax

         
 C#  Visual Basic  Visual C++ 
public Index(
	Matrix<float> values,
	int numberOfKDTrees,
	int branching,
	int iterations,
	CenterInitType centersInitType,
	float cbIndex
)
Public Sub New ( _
	values As Matrix(Of Single), _
	numberOfKDTrees As Integer, _
	branching As Integer, _
	iterations As Integer, _
	centersInitType As CenterInitType, _
	cbIndex As Single _
)
public:
Index(
	Matrix<float>^ values, 
	int numberOfKDTrees, 
	int branching, 
	int iterations, 
	CenterInitType centersInitType, 
	float cbIndex
)

Parameters

values
Matrix<(Of <(<'Single>)>)>
A row by row matrix of descriptors
numberOfKDTrees
Int32
The number of KDTrees to be used
branching
Int32
Branching factor (for kmeans tree), use 32 for default
iterations
Int32
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
Single
Cluster boundary index. Used when searching the kmeans tree. Use 0.2 for default

See Also