Module imagingbook.common

Package imagingbook.common.image.matching.lucaskanade

Class LucasKanadeForwardMatcher

java.lang.Object

imagingbook.common.image.matching.lucaskanade.LucasKanadeMatcher

imagingbook.common.image.matching.lucaskanade.LucasKanadeForwardMatcher

public class LucasKanadeForwardMatcher
extends LucasKanadeMatcher

Lucas-Kanade (forward-additive) matcher [1], as described in [2]. Also called the "forward-additive" algorithm. See Sec. 24.2 (Alg. 24.1) of [2] for additional details. This version assumes that the origin of R is at its center!

[1] B. D. Lucas and T. Kanade. "An iterative image registration technique with an application to stereo vision". In Proceedings of the 7th International Joint Conference on Artificial Intelligence IJCAI’81, pp. 674–679, Vancouver, BC (1981).
[2] Simon Baker and Iain Matthews, "Lucas-Kanade 20 Years On: A Unifying Framework: Part 1", CMU-RI-TR-02-16 (2002)
[3] W. Burger, M.J. Burge, Digital Image Processing – An Algorithmic Introduction, 3rd ed, Springer (2022).

Version:

2014/02/08

Nested Class Summary

Nested classes/interfaces inherited from class imagingbook.common.image.matching.lucaskanade.LucasKanadeMatcher

LucasKanadeMatcher.Parameters

Constructor Summary

Constructors

Constructor	Description
LucasKanadeForwardMatcher(FloatProcessor I, FloatProcessor R)	Constructor using default parameters.
LucasKanadeForwardMatcher(FloatProcessor I, FloatProcessor R, LucasKanadeMatcher.Parameters params)	Constructor.

Method Summary

Modifier and Type	Method	Description
double	getRmsError()	Measures the RMS intensity difference between the reference image R and the patch in the search image I defined by the current warp Tp.
boolean	hasConverged()	Checks if the matcher has converged.
ProjectiveMapping2D	iterateOnce(ProjectiveMapping2D Tp)	Performs a single matching iteration on the given image pair (I, R).

Methods inherited from class imagingbook.common.image.matching.lucaskanade.LucasKanadeMatcher

getIteration, getMatch, getReferenceMappingTo, getReferencePoints

Methods inherited from class java.lang.Object

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

Constructor Details

LucasKanadeForwardMatcher

public LucasKanadeForwardMatcher(FloatProcessor I,
 FloatProcessor R,
 LucasKanadeMatcher.Parameters params)

Constructor.

Parameters:

I - the search image (of type FloatProcessor).

R - the reference image (of type FloatProcessor)

params - a parameter object of type LucasKanadeMatcher.Parameters.

LucasKanadeForwardMatcher

public LucasKanadeForwardMatcher(FloatProcessor I,
 FloatProcessor R)

Constructor using default parameters.

Parameters:

I - the search image (of type FloatProcessor).

R - the reference image (of type FloatProcessor)

Method Details

hasConverged

public boolean hasConverged()

Description copied from class: LucasKanadeMatcher

Checks if the matcher has converged.

Specified by:

hasConverged in class LucasKanadeMatcher

Returns:

true if minimization criteria have been reached.

getRmsError

public double getRmsError()

Description copied from class: LucasKanadeMatcher

Measures the RMS intensity difference between the reference image R and the patch in the search image I defined by the current warp Tp.

Specified by:

getRmsError in class LucasKanadeMatcher

Returns:

the RMS error under the current warp

iterateOnce

public ProjectiveMapping2D iterateOnce(ProjectiveMapping2D Tp)

Description copied from class: LucasKanadeMatcher

Performs a single matching iteration on the given image pair (I, R).

Specified by:

iterateOnce in class LucasKanadeMatcher

Parameters:

Tp - the warp transformation from the reference image R to the initial search patch, assuming that R is centered at the coordinate origin!

Returns:

a new warp transformation (again assuming that R is centered at the coordinate origin) or null if the iteration was unsuccessful.