org.openimaj.feature.local.matcher

Class VotingKeypointMatcher<T extends Keypoint>

java.lang.Object

org.openimaj.feature.local.matcher.BasicMatcher<T>

org.openimaj.feature.local.matcher.FastBasicKeypointMatcher<T>

org.openimaj.feature.local.matcher.VotingKeypointMatcher<T>

Type Parameters:

T -

All Implemented Interfaces:

LocalFeatureMatcher<T>

public class VotingKeypointMatcher<T extends Keypoint>
extends FastBasicKeypointMatcher<T>
implements LocalFeatureMatcher<T>

Matcher rejects matches with no local support

Author:

Jonathon Hare

Field Summary

Fields

Modifier and Type	Field and Description
protected int	minVote
protected float	singularityDistance

Fields inherited from class org.openimaj.feature.local.matcher.FastBasicKeypointMatcher

modelKeypointsKNN

Fields inherited from class org.openimaj.feature.local.matcher.BasicMatcher

matches, modelKeypoints, thresh

Constructor Summary

Constructors

Constructor and Description
VotingKeypointMatcher(int threshold)
VotingKeypointMatcher(int threshold, int neighbours, int minVote, float singularityDistance)

Method Summary

Modifier and Type	Method and Description
protected boolean	checkSingularity()
protected float	euclideanSqr(Keypoint k1, Keypoint k2)
boolean	findMatches(List<T> keys1) Given a pair of images and their keypoints, pick the first keypoint from one image and find its closest match in the second set of keypoints.
protected List<T>	findModelNeighbours(T kp)
List<Pair<T>>	getAllMatches()
protected float[]	getCentroid()
List<Pair<T>>	getMatches() Get the matches detected by the underlying algorithm
protected int	vote(Pair<T> match)

Methods inherited from class org.openimaj.feature.local.matcher.FastBasicKeypointMatcher

setModelFeatures

Methods inherited from class org.openimaj.feature.local.matcher.BasicMatcher

checkForMatch, setThreshold

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface org.openimaj.feature.local.matcher.LocalFeatureMatcher

setModelFeatures

Field Detail

minVote

protected int minVote

singularityDistance

protected float singularityDistance

Constructor Detail

VotingKeypointMatcher

public VotingKeypointMatcher(int threshold)

Parameters:

threshold - threshold for determining matching keypoints

VotingKeypointMatcher

public VotingKeypointMatcher(int threshold,
                             int neighbours,
                             int minVote,
                             float singularityDistance)

Parameters:

threshold - threshold for determining matching keypoints

neighbours - number of neighbours within which to check for local support

minVote -

singularityDistance -

Method Detail

getMatches

public List<Pair<T>> getMatches()

Description copied from interface: LocalFeatureMatcher

Get the matches detected by the underlying algorithm

Specified by:

getMatches in interface LocalFeatureMatcher<T extends Keypoint>

Overrides:

getMatches in class BasicMatcher<T extends Keypoint>

Returns:

a list of consistent matching keypoints according to the estimated model parameters.

getAllMatches

public List<Pair<T>> getAllMatches()

Returns:

a list of all matches irrespective of whether they fit the model

findMatches

public boolean findMatches(List<T> keys1)

Description copied from class: FastBasicKeypointMatcher

Given a pair of images and their keypoints, pick the first keypoint from one image and find its closest match in the second set of keypoints. Then write the result to a file.

Specified by:

findMatches in interface LocalFeatureMatcher<T extends Keypoint>

Overrides:

findMatches in class FastBasicKeypointMatcher<T extends Keypoint>

Parameters:

keys1 - features from the query

Returns:

whether matches were found

getCentroid

protected float[] getCentroid()

checkSingularity

protected boolean checkSingularity()

vote

protected int vote(Pair<T> match)

euclideanSqr

protected float euclideanSqr(Keypoint k1,
                             Keypoint k2)

findModelNeighbours

protected List<T> findModelNeighbours(T kp)