viper.comparison.distances
Class Distances.HelperMeasureDistance

java.lang.Object
  viper.comparison.distances.Distances.HelperDistance
      viper.comparison.distances.Distances.HelperMeasureDistance

All Implemented Interfaces:

Distance, MeasureDistance

Enclosing class:

Distances

public static class Distances.HelperMeasureDistance

extends viper.comparison.distances.Distances.HelperDistance

implements MeasureDistance

Converts one of the value distance function objects into a MeasureDistance.

Field Summary

protected Distances.QuickValueDistance

dist

Fields inherited from interface viper.comparison.distances.Distance

BALANCED, CAND_V_TARGS, OVERALL_MEAN, OVERALL_SUM, TARG_V_CANDS

Constructor Summary

Distances.HelperMeasureDistance(Distances.QuickValueDistance d, java.lang.String name, int type, java.lang.String explanation, boolean isDistance)
Constructs a new measure distance from the given function object and parameters.

Distances.HelperMeasureDistance(Distances.SmartValueDistance d, java.lang.String name, int type, java.lang.String explanation, boolean isDistance)
Constructs a new measure distance from the given function object and parameters.

Distances.HelperMeasureDistance(Distances.ValueDistance d, java.lang.String name, int type, java.lang.String explanation, boolean isDistance)
Constructs a new measure distance from the given function object and parameters.

Method Summary

boolean	equals(java.lang.Object o)
java.lang.Number	getDistance(Measurable.Difference D) Gets the distance using cached distance data.
java.lang.Number	getDistance(Measurable alpha, Measurable beta, CanonicalFileDescriptor cfd) Compute the distance.
java.lang.Number	getDistance(Measurable alpha, Measurable beta, Measurable blackout, Measurable ignore, CanonicalFileDescriptor cfd) Compute the distance.
java.lang.String	getExplanation() An explanatory message, like "Pixels matched", or "Localized Object Count Recall".
int	getType() Get what type of metric this is, per the list of static int types.
int	hashCode()
boolean	isDistance() Determines if this is a distance or a similarity quotient.
java.lang.String	toString() Get the EPF name of the metric.

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Methods inherited from interface viper.comparison.distances.Distance

getDistance, getExplanation, getType, isDistance, toString

Field Detail

dist

protected Distances.QuickValueDistance dist

Constructor Detail

Distances.HelperMeasureDistance

public Distances.HelperMeasureDistance(Distances.ValueDistance d,
                                       java.lang.String name,
                                       int type,
                                       java.lang.String explanation,
                                       boolean isDistance)

Constructs a new measure distance from the given function object and parameters.

Parameters:

d - the distance functor

name - the name of the distance metric

type - the category of distance, e.g. Distance.BALANCED

explanation - a long explanation of the distance, for helper documentation and verbose output

isDistance - determines if the distance is really a distance or if it is a score (bigger is better)

Distances.HelperMeasureDistance

public Distances.HelperMeasureDistance(Distances.SmartValueDistance d,
                                       java.lang.String name,
                                       int type,
                                       java.lang.String explanation,
                                       boolean isDistance)

Constructs a new measure distance from the given function object and parameters.

Parameters:

d - the distance functor

name - the name of the distance metric

type - the category of distance, e.g. Distance.BALANCED

explanation - a long explanation of the distance, for helper documentation and verbose output

isDistance - determines if the distance is really a distance or if it is a score (bigger is better)

Distances.HelperMeasureDistance

public Distances.HelperMeasureDistance(Distances.QuickValueDistance d,
                                       java.lang.String name,
                                       int type,
                                       java.lang.String explanation,
                                       boolean isDistance)

Constructs a new measure distance from the given function object and parameters.

Parameters:

d - the distance functor

name - the name of the distance metric

type - the category of distance, e.g. Distance.BALANCED

explanation - a long explanation of the distance, for helper documentation and verbose output

isDistance - determines if the distance is really a distance or if it is a score (bigger is better)

Method Detail

getDistance

public java.lang.Number getDistance(Measurable alpha,
                                    Measurable beta,
                                    CanonicalFileDescriptor cfd)

Description copied from interface: MeasureDistance

Compute the distance. Only defined for static attributes and the like (read: FrameSpan).

Specified by:

getDistance in interface MeasureDistance

Parameters:

alpha - the alpha value

beta - the beta value

cfd - the file descriptor that alpha and beta are defined within

Returns:

the distance between alpha and beta

getDistance

public java.lang.Number getDistance(Measurable alpha,
                                    Measurable beta,
                                    Measurable blackout,
                                    Measurable ignore,
                                    CanonicalFileDescriptor cfd)
                             throws IgnoredValueException

Description copied from interface: MeasureDistance

Compute the distance. Only defined for static attributes and the like (read: FrameSpan). This one includes blackout and ignore parameters - blackouts apply to regions that are losses, and ignore are for regions that are don't-care. For example, "scene" data may be marked as don't-care, or frames that don't meat a certain metric are qualified as blackout.

Specified by:

getDistance in interface MeasureDistance

Parameters:

alpha - the first value

beta - the second value

blackout - the bad region

ignore - the region to ignore

cfd - the file on which the measurables are defined

Returns:

the distance between alpha and beta, given the constraints

Throws:

IgnoredValueException - when the ignored region is such that it prevents evaluation of the distance.

getType

public int getType()

Description copied from interface: Distance

Get what type of metric this is, per the list of static int types.

Specified by:

getType in interface Distance

Returns:

one of BALANCED, CAND_V_TARGS, etc.

getExplanation

public java.lang.String getExplanation()

Description copied from interface: Distance

An explanatory message, like "Pixels matched", or "Localized Object Count Recall".

Specified by:

getExplanation in interface Distance

Returns:

this distance's explanatory message

toString

public java.lang.String toString()

Description copied from interface: Distance

Get the EPF name of the metric.

Specified by:

toString in interface Distance

getDistance

public java.lang.Number getDistance(Measurable.Difference D)

Description copied from interface: Distance

Gets the distance using cached distance data. Has two major benefits: possible consolidation of IgnoredValueExceptions and savings on recomputing, for example, the area of polygons. Calling getDistance (alpha, beta, blackout, ignore) should have the same effect as getDistance (alpha.getDifference (beta, blackout, ignore))

Specified by:

getDistance in interface Distance

Parameters:

D - the distance of the difference

Returns:

the distance

hashCode

public int hashCode()

equals

public boolean equals(java.lang.Object o)

isDistance

public boolean isDistance()

Description copied from interface: Distance

Determines if this is a distance or a similarity quotient. Similarity quotients are better the larger they are, but distances should be minimized.

Specified by:

isDistance in interface Distance

Returns:

true when 0 indicates a perfect match, and distance varies proportionally to difference