Module imagingbook.common

Package imagingbook.common.sift

Class SiftMatcher

java.lang.Object

imagingbook.common.sift.SiftMatcher

public class SiftMatcher
extends Object

Instances of this class perform matching between SIFT features. See Secs. 25.5 of [1] for more details.

[1] W. Burger, M.J. Burge, Digital Image Processing – An Algorithmic Introduction, 3rd ed, Springer (2022).

Version:

2022/11/20 removed inner Parameters class, changed constructor

Field Summary

Fields

Modifier and Type	Field	Description
static final VectorNorm.NormType	DefaultNormType
static final double	DefaultRMax

Constructor Summary

Constructors

Constructor	Description
SiftMatcher()	Constructor using default parameters.
SiftMatcher(VectorNorm.NormType normType, double rMax)	Constructor using specific parameters.

Method Summary

Modifier and Type	Method	Description
List<SiftMatch>	match(Collection<SiftDescriptor> setA, Collection<SiftDescriptor> setB)	Finds matches between two sets of SIFT descriptors.

Methods inherited from class java.lang.Object

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

Field Details

DefaultNormType

public static final VectorNorm.NormType DefaultNormType

DefaultRMax

public static final double DefaultRMax

See Also:

• Constant Field Values

Constructor Details

SiftMatcher

public SiftMatcher(VectorNorm.NormType normType,
 double rMax)

Constructor using specific parameters.

Parameters:

normType - the distance norm for comparing feature vectors

rMax - the max. distance ratio between best and second-best match

SiftMatcher

public SiftMatcher()

Constructor using default parameters.

See Also:

• DefaultNormType
• DefaultRMax

Method Details

match

public List<SiftMatch> match(Collection<SiftDescriptor> setA,
 Collection<SiftDescriptor> setB)

Finds matches between two sets of SIFT descriptors. For each descriptor in the first set, the best and second-best fits are searched for in the second set. A valid match instance is created if the distance to the best-matching descriptor is significantly smaller than the distance to the second-best matching descriptor. The resulting list of matches is sorted by increasing match distance.

Parameters:

setA - the first set of SIFT descriptors

setB - the second set of SIFT descriptors

Returns:

a (possibly empty) list of SiftMatch instances