org.openimaj.feature

Interface FeatureVector

All Superinterfaces:

Cloneable, Readable, ReadableASCII, ReadableBinary, ReadWriteable, Serializable, Writeable, WriteableASCII, WriteableBinary

All Known Implementing Classes:

ArrayFeatureVector, ByteFV, Document, DoubleFV, EnumFV, FloatFV, Histogram, IntFV, LongFV, MultidimensionalByteFV, MultidimensionalDoubleFV, MultidimensionalFloatFV, MultidimensionalHistogram, MultidimensionalIntFV, MultidimensionalLongFV, MultidimensionalShortFV, OrientedFeatureVector, ShortFV, SparseByteFV, SparseDoubleFV, SparseFloatFV, SparseIntFV, SparseLongFV, SparseShortFV, TernaryPlot.TernaryData

public interface FeatureVector
extends Cloneable, Serializable, ReadWriteable

Interface for objects that represent feature vectors.

Author:

Jonathon Hare (jsh2@ecs.soton.ac.uk)

Method Summary

Methods

Modifier and Type	Method and Description
DoubleFV	asDoubleFV() Convert the FV to a DoubleFV representation
double[]	asDoubleVector() Convert the FV to a 1-dimensional double array representation
Object	getVector() Get the underlying data array.
int	length() Get the length of this vector
DoubleFV	normaliseFV() Normalise the FV to unit length
DoubleFV	normaliseFV(double p) Lp Norm of the FV.
DoubleFV	normaliseFV(double[] min, double[] max) Element-wise normalisation to 0..1 using separated expected minimum and maximum values for each element of the underlying feature vector.
DoubleFV	normaliseFV(double min, double max) Min-Max normalisation of the FV.

Methods inherited from interface org.openimaj.io.ReadableASCII

asciiHeader, readASCII

Methods inherited from interface org.openimaj.io.ReadableBinary

binaryHeader, readBinary

Methods inherited from interface org.openimaj.io.WriteableASCII

asciiHeader, writeASCII

Methods inherited from interface org.openimaj.io.WriteableBinary

binaryHeader, writeBinary

Method Detail

getVector

Object getVector()

Get the underlying data array.

Returns:

underlying data

length

int length()

Get the length of this vector

Returns:

the length of this vector

normaliseFV

DoubleFV normaliseFV(double[] min,
                   double[] max)

Element-wise normalisation to 0..1 using separated expected minimum and maximum values for each element of the underlying feature vector.

Parameters:

min - an array containing the minimum expected values

max - an array containing the maximum expected values

Returns:

copy of the feature vector with each value normalised to 0..1

normaliseFV

DoubleFV normaliseFV(double min,
                   double max)

Min-Max normalisation of the FV. Each element of the underlying feature vector is normalised to 0..1 based on the provided minimum and maximum expected values.

Parameters:

min - the minimum expected value

max - the maximum expected value

Returns:

copy of the feature vector with each value normalised to 0..1

normaliseFV

DoubleFV normaliseFV()

Normalise the FV to unit length

Returns:

a copy of the feature vector as a DoubleFV, normalised to unit length

asDoubleFV

DoubleFV asDoubleFV()

Convert the FV to a DoubleFV representation

Returns:

a copy of the feature vector as a DoubleFV

asDoubleVector

double[] asDoubleVector()

Convert the FV to a 1-dimensional double array representation

Returns:

a copy of the feature vector as a double array

normaliseFV

DoubleFV normaliseFV(double p)

Lp Norm of the FV.

Parameters:

p - the norm to compute

Returns:

feature vector normalised using the Lp norm