org.openimaj.math.model.fit

Class LMedS<I,D,M extends EstimatableModel<I,D>>

java.lang.Object

org.openimaj.math.model.fit.LMedS<I,D,M>

Type Parameters:

I - type of independent data

D - type of dependent data

M - concrete type of model learned

All Implemented Interfaces:

ModelFitting<I,D,M>, RobustModelFitting<I,D,M>

@Reference(type=Article,
           author="Peter J. Rousseeuw",
           title="Least Median of Squares Regression",
           year="1984",
           journal="Journal of the American Statistical Association",
           pages={"871","","880"},
           url="http://www.jstor.org/stable/2288718",
           month="December",
           number="388",
           volume="79")
public class LMedS<I,D,M extends EstimatableModel<I,D>>
extends Object
implements RobustModelFitting<I,D,M>

Least Median of Squares robust model fitting

Author:

Jonathon Hare (jsh2@ecs.soton.ac.uk)

Field Summary

Fields

Modifier and Type	Field and Description
protected M	bestModel
protected boolean	improveEstimate
protected List<IndependentPair<I,D>>	inliers
protected M	model
protected List<IndependentPair<I,D>>	outliers
protected ResidualCalculator<I,D,M>	residualEstimator
protected CollectionSampler<IndependentPair<I,D>>	sampler

Constructor Summary

Constructors

Constructor and Description
LMedS(M model, ResidualCalculator<I,D,M> residualEstimator, boolean impEst) Construct with the given model and residual calculator.
LMedS(M model, ResidualCalculator<I,D,M> residualEstimator, boolean impEst, CollectionSampler<IndependentPair<I,D>> sampler) Construct with the given model and residual calculator.
LMedS(M model, ResidualCalculator<I,D,M> residualEstimator, double outlierProportion, boolean impEst) Construct with the given model, residual calculator and estimated proportion of outliers.
LMedS(M model, ResidualCalculator<I,D,M> residualEstimator, double outlierProportion, boolean impEst, CollectionSampler<IndependentPair<I,D>> sampler) Construct with the given model, residual calculator and estimated proportion of outliers.
LMedS(M model, ResidualCalculator<I,D,M> residualEstimator, double outlierProportion, double inlierNoiseLevel, double degreesOfFreedom, boolean impEst) Construct with the given model, residual calculator and estimated proportion of outliers.
LMedS(M model, ResidualCalculator<I,D,M> residualEstimator, double outlierProportion, double inlierNoiseLevel, double degreesOfFreedom, boolean impEst, CollectionSampler<IndependentPair<I,D>> sampler) Construct with the given model, residual calculator and estimated proportion of outliers.

Method Summary

Modifier and Type	Method and Description
boolean	fitData(List<? extends IndependentPair<I,D>> data) Attempt to fit the given data to the model.
List<? extends IndependentPair<I,D>>	getInliers()
M	getModel()
List<? extends IndependentPair<I,D>>	getOutliers()
int	numItemsToEstimate()

Methods inherited from class java.lang.Object

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

Field Detail

residualEstimator

protected ResidualCalculator<I,D,M extends EstimatableModel<I,D>> residualEstimator

improveEstimate

protected boolean improveEstimate

model

protected M extends EstimatableModel<I,D> model

bestModel

protected M extends EstimatableModel<I,D> bestModel

inliers

protected List<IndependentPair<I,D>> inliers

outliers

protected List<IndependentPair<I,D>> outliers

sampler

protected CollectionSampler<IndependentPair<I,D>> sampler

Constructor Detail

LMedS

public LMedS(M model,
             ResidualCalculator<I,D,M> residualEstimator,
             boolean impEst)

Construct with the given model and residual calculator. The proportion of outliers is assumed to be 0.4. The threshold for determining inliers and outliers is automatically computed. Uniform random sampling is used for creating the subsets.

Parameters:

model - the model to estimate

residualEstimator - the algorithm to compute residuals of the model

impEst - True if we want to perform a final fitting of the model with all inliers, false otherwise

LMedS

public LMedS(M model,
             ResidualCalculator<I,D,M> residualEstimator,
             double outlierProportion,
             boolean impEst)

Construct with the given model, residual calculator and estimated proportion of outliers. The threshold for determining inliers and outliers is automatically computed. Uniform random sampling is used for creating the subsets.

Parameters:

model - the model to estimate

residualEstimator - the algorithm to compute residuals of the model

outlierProportion - Expected proportion of outliers

impEst - True if we want to perform a final fitting of the model with all inliers, false otherwise

LMedS

public LMedS(M model,
             ResidualCalculator<I,D,M> residualEstimator,
             double outlierProportion,
             double inlierNoiseLevel,
             double degreesOfFreedom,
             boolean impEst)

Construct with the given model, residual calculator and estimated proportion of outliers. The given inlier noise level and number of degrees of freedom of the model are used to estimate the optimal threshold for determining inliers versus outliers. Uniform random sampling is used for creating the subsets.

Parameters:

model - the model to estimate

residualEstimator - the algorithm to compute residuals of the model

outlierProportion - Expected proportion of outliers

inlierNoiseLevel - The level of inlier noise (standard deviation of Gaussian noise inherent in the data).

degreesOfFreedom - Number of degrees of freedom of the model

impEst - True if we want to perform a final fitting of the model with all inliers, false otherwise

LMedS

public LMedS(M model,
             ResidualCalculator<I,D,M> residualEstimator,
             boolean impEst,
             CollectionSampler<IndependentPair<I,D>> sampler)

Construct with the given model and residual calculator. The proportion of outliers is assumed to be 0.4. The threshold for determining inliers and outliers is automatically computed.

Parameters:

model - the model to estimate

residualEstimator - the algorithm to compute residuals of the model

impEst - True if we want to perform a final fitting of the model with all inliers, false otherwise

sampler - the sampling algorithm for selecting random subsets

LMedS

public LMedS(M model,
             ResidualCalculator<I,D,M> residualEstimator,
             double outlierProportion,
             boolean impEst,
             CollectionSampler<IndependentPair<I,D>> sampler)

Construct with the given model, residual calculator and estimated proportion of outliers. The threshold for determining inliers and outliers is automatically computed.

Parameters:

model - the model to estimate

residualEstimator - the algorithm to compute residuals of the model

outlierProportion - Expected proportion of outliers

impEst - True if we want to perform a final fitting of the model with all inliers, false otherwise

sampler - the sampling algorithm for selecting random subsets

LMedS

public LMedS(M model,
             ResidualCalculator<I,D,M> residualEstimator,
             double outlierProportion,
             double inlierNoiseLevel,
             double degreesOfFreedom,
             boolean impEst,
             CollectionSampler<IndependentPair<I,D>> sampler)

Construct with the given model, residual calculator and estimated proportion of outliers. The given inlier noise level and number of degrees of freedom of the model are used to estimate the optimal threshold for determining inliers versus outliers.

Parameters:

model - the model to estimate

residualEstimator - the algorithm to compute residuals of the model

outlierProportion - Expected proportion of outliers

inlierNoiseLevel - The level of inlier noise (standard deviation of Gaussian noise inherent in the data).

degreesOfFreedom - Number of degrees of freedom of the model

impEst - True if we want to perform a final fitting of the model with all inliers, false otherwise

sampler - the sampling algorithm for selecting random subsets

Method Detail

fitData

public boolean fitData(List<? extends IndependentPair<I,D>> data)

Description copied from interface: ModelFitting

Attempt to fit the given data to the model.

Specified by:

fitData in interface ModelFitting<I,D,M extends EstimatableModel<I,D>>

Parameters:

data - Data to be fitted

Returns:

true on success, false otherwise

getModel

public M getModel()

Specified by:

getModel in interface ModelFitting<I,D,M extends EstimatableModel<I,D>>

Returns:

the trained model object

getInliers

public List<? extends IndependentPair<I,D>> getInliers()

Specified by:

getInliers in interface RobustModelFitting<I,D,M extends EstimatableModel<I,D>>

Returns:

list of the inliers in the original data

getOutliers

public List<? extends IndependentPair<I,D>> getOutliers()

Specified by:

getOutliers in interface RobustModelFitting<I,D,M extends EstimatableModel<I,D>>

Returns:

list of the outliers in the original data

numItemsToEstimate

public int numItemsToEstimate()

Specified by:

numItemsToEstimate in interface ModelFitting<I,D,M extends EstimatableModel<I,D>>

Returns:

The minimum number of observations required to estimate the model.