boofcv.alg.feature.disparity.impl

Class ImplSelectSparseBasicWta_F32

java.lang.Object

boofcv.alg.feature.disparity.impl.ImplSelectSparseBasicWta_F32

All Implemented Interfaces:

DisparitySparseSelect<float[]>

public class ImplSelectSparseBasicWta_F32
extends java.lang.Object
implements DisparitySparseSelect<float[]>

Selects the disparity with the lowest score with no additional validation. Lack of validation speeds up the code at the cost of reduced signal to noise ratio. This strategy of selecting the lowest score is also known as Winner Take All (WTA).

Author:

Peter Abeles

Constructor Summary

Constructors

Constructor and Description
ImplSelectSparseBasicWta_F32()

Method Summary

Methods

Modifier and Type	Method and Description
double	getDisparity() Returns the found disparity
boolean	select(float[] scores, int maxDisparity) Examines disparity scores and looks for the best correspondence.

Methods inherited from class java.lang.Object

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

Constructor Detail

ImplSelectSparseBasicWta_F32

public ImplSelectSparseBasicWta_F32()

Method Detail

select

public boolean select(float[] scores,
             int maxDisparity)

Description copied from interface: DisparitySparseSelect

Examines disparity scores and looks for the best correspondence. If no correspondence can be found then false is returned.

Specified by:

select in interface DisparitySparseSelect<float[]>

Parameters:

scores - Set of disparity scores.

maxDisparity - Maximum allowed disparity.

Returns:

true if a valid correspondence was found

getDisparity

public double getDisparity()

Description copied from interface: DisparitySparseSelect

Returns the found disparity

Specified by:

getDisparity in interface DisparitySparseSelect<float[]>

Returns:

disparity