Package com.irurueta.numerical.robust

Class MSACRobustEstimator.MSACInliersData

java.lang.Object

com.irurueta.numerical.robust.InliersData

com.irurueta.numerical.robust.MSACRobustEstimator.MSACInliersData

Enclosing class:

MSACRobustEstimator<T>

public static class MSACRobustEstimator.MSACInliersData
extends InliersData

Contains data related to inliers estimated in one iteration.

Field Summary

Fields

Modifier and Type	Field	Description
private double	bestMedianResidual	Best median of error found so far taking into account all provided samples.
private BitSet	inliers	Efficiently stores which samples are considered inliers and which ones aren't.
private boolean	medianResidualImproved	Indicates whether median residual computed in current iteration has improved respect to previous iterations.

Fields inherited from class com.irurueta.numerical.robust.InliersData

numInliers, residuals

Constructor Summary

Constructors

Modifier	Constructor	Description
protected	MSACInliersData(int totalSamples)	Constructor.

Method Summary

Modifier and Type	Method	Description
double	getBestMedianResidual()	Returns best median of error found so far taking into account all provided samples.
BitSet	getInliers()	Returns efficient array indicating which samples are considered inliers and which ones aren't.
boolean	isMedianResidualImproved()	Returns boolean indicating whether median residual computed in current iteration has improved respect to previous iterations.
protected void	update(double bestMedianResidual, BitSet inliers, double[] residuals, int numInliers, boolean medianResidualImproved)	Updates data contained in this instance.

Methods inherited from class com.irurueta.numerical.robust.InliersData

getNumInliers, getResiduals

Methods inherited from class java.lang.Object

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

Field Details

bestMedianResidual

private double bestMedianResidual

Best median of error found so far taking into account all provided samples.

inliers

private BitSet inliers

Efficiently stores which samples are considered inliers and which ones aren't.

medianResidualImproved

private boolean medianResidualImproved

Indicates whether median residual computed in current iteration has improved respect to previous iterations.

Constructor Details

MSACInliersData

protected MSACInliersData(int totalSamples)

Constructor.

Parameters:

totalSamples - total number of samples.

Method Details

getBestMedianResidual

public double getBestMedianResidual()

Returns best median of error found so far taking into account all provided samples.

Returns:

best median of error found so far taking into account all provided samples.

getInliers

public BitSet getInliers()

Returns efficient array indicating which samples are considered inliers and which ones aren't.

Specified by:

getInliers in class InliersData

Returns:

array indicating which samples are considered inliers and which ones aren't.

isMedianResidualImproved

public boolean isMedianResidualImproved()

Returns boolean indicating whether median residual computed in current iteration has improved respect to previous iterations.

Returns:

true if median residual improved, false otherwise.

update

protected void update(double bestMedianResidual,
 BitSet inliers,
 double[] residuals,
 int numInliers,
 boolean medianResidualImproved)

Updates data contained in this instance.

Parameters:

bestMedianResidual - best median of error found so far taking into account all provided samples.

inliers - efficiently stores which samples are considered inliers and which ones aren't.

residuals - residuals obtained for each sample of data.

numInliers - number of inliers found on current iteration.

medianResidualImproved - indicates whether median residual computed in current iteration has improved respect to previous iteration.