Module imagingbook.common

Package imagingbook.common.corners

Class MopsCornerDetector

java.lang.Object

imagingbook.common.corners.GradientCornerDetector

imagingbook.common.corners.MopsCornerDetector

public class MopsCornerDetector
extends GradientCornerDetector

This is an implementation of the corner detector described in [1]. This class extends GradientCornerDetector (where most of the work is done) by defining a specific corner score function and associated threshold. The corner score is defined as the harmonic mean of the local structure tensor's eigenvalues lambda_1, lambda_2. See Sec. 6.3.2 of [2] for additional details.

[1] M. Brown, R. Szeliski, and S. Winder, Multi-image matching using multi-scale oriented patches, in Proc. of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05), 2005, pp. 510–517.
[2] W. Burger, M.J. Burge, Digital Image Processing – An Algorithmic Introduction, 3rd ed, Springer (2022).

Version:

2022/03/30

Nested Class Summary

Nested classes/interfaces inherited from class imagingbook.common.corners.GradientCornerDetector

GradientCornerDetector.Parameters

Field Summary

Fields

Modifier and Type	Field	Description
static double	DEFAULT_THRESHOLD

Fields inherited from class imagingbook.common.corners.GradientCornerDetector

M, N, params, Q, RETAIN_TEMPORARY_DATA, UndefinedScoreValue

Constructor Summary

Constructors

Constructor	Description
MopsCornerDetector(ImageProcessor ip, GradientCornerDetector.Parameters params)

Method Summary

Modifier and Type	Method	Description
protected float	getCornerScore(float A, float B, float C)	Calculates the corner response score for a single image position (u,v) from the elements A, B, C of the local structure matrix.

Methods inherited from class imagingbook.common.corners.GradientCornerDetector

getA, getB, getC, getCorners, getQ

Methods inherited from class java.lang.Object

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

Field Details

DEFAULT_THRESHOLD

public static double DEFAULT_THRESHOLD

Constructor Details

MopsCornerDetector

public MopsCornerDetector(ImageProcessor ip,
 GradientCornerDetector.Parameters params)

Method Details

getCornerScore

protected float getCornerScore(float A,
 float B,
 float C)

Description copied from class: GradientCornerDetector

Calculates the corner response score for a single image position (u,v) from the elements A, B, C of the local structure matrix. The method should normalize the score value, such that 100 is returned for the default threshold value. To be implemented by concrete sub-classes.

Specified by:

getCornerScore in class GradientCornerDetector

Parameters:

A - = Ixx(u,v)

B - = Iyy(u,v)

C - = Ixy(u,v)

Returns:

the corner score

See Also:

• HarrisCornerDetector
• ShiTomasiCornerDetector