de.unihalle.informatik.MiToBo.tracking.multitarget.distributions.abstracts
Class AbstractAssociationDistribution<S extends Copyable<?>,T extends Copyable<?>>

java.lang.Object
  extended by de.unihalle.informatik.MiToBo.tracking.multitarget.distributions.abstracts.AbstractAssociationDistribution<S,T>
Type Parameters:
S - Type of discrete variables in the multi target observation
T - Type of discrete variables in the multi target state
All Implemented Interfaces:
SamplingDistribution<DataAssociation>
Direct Known Subclasses:
AssociationDistribution

public abstract class AbstractAssociationDistribution<S extends Copyable<?>,T extends Copyable<?>>
extends java.lang.Object
implements SamplingDistribution<DataAssociation>

Abstract class for association distributions that allow sampling of the association variables for a set of observations in a multi-target tracking framework.

Author:
Oliver Gress

Field Summary
protected  DataAssociationFactory assocfactory
          factory for the data association object
protected  LogProbabilityDensityFunction clutterdistrib
          spatial clutter distribution
protected  double[][] log_pzc
          storage for the log likelihood values of the observations for possible associations to avoid recomputation
protected  int M
          number of observations
protected  int N
          number of targets
protected  LogProbabilityDensityFunction newborndistrib
          spatial distribution of observations from newborn targets
protected  AbstractMultiObservationDistributionIndep<S,T> obsdistrib
          observation/measurement likelihood P(z_m^t | c_m, Z^{1:t-1}, c_{1:m-1}
protected  java.util.Random rand
          random generator for sampling
protected  AbstractMultiState<S> Z
          observations
 
Constructor Summary
AbstractAssociationDistribution(java.util.Random rand, AbstractMultiState<S> Z, AbstractMultiObservationDistributionIndep<S,T> observationDistrib, LogProbabilityDensityFunction spatialClutterDistrib, LogProbabilityDensityFunction spatialNewbornDistrib, DataAssociationFactory assocFactory)
          Constructor
 
Method Summary
abstract  DataAssociation drawSample()
          Generate a new sample from this density.
abstract  DataAssociation drawSampleDebug(DataAssociation groundtruth, java.io.OutputStream ostream)
           
 void setNewObservations(AbstractMultiState<S> Z, AbstractMultiObservationDistributionIndep<S,T> observationDistrib)
           
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Field Detail

assocfactory

protected DataAssociationFactory assocfactory
factory for the data association object


clutterdistrib

protected LogProbabilityDensityFunction clutterdistrib
spatial clutter distribution


log_pzc

protected double[][] log_pzc
storage for the log likelihood values of the observations for possible associations to avoid recomputation


M

protected int M
number of observations


N

protected int N
number of targets


newborndistrib

protected LogProbabilityDensityFunction newborndistrib
spatial distribution of observations from newborn targets


obsdistrib

protected AbstractMultiObservationDistributionIndep<S extends Copyable<?>,T extends Copyable<?>> obsdistrib
observation/measurement likelihood P(z_m^t | c_m, Z^{1:t-1}, c_{1:m-1}


rand

protected java.util.Random rand
random generator for sampling


Z

protected AbstractMultiState<S extends Copyable<?>> Z
observations

Constructor Detail

AbstractAssociationDistribution

public AbstractAssociationDistribution(java.util.Random rand,
                                       AbstractMultiState<S> Z,
                                       AbstractMultiObservationDistributionIndep<S,T> observationDistrib,
                                       LogProbabilityDensityFunction spatialClutterDistrib,
                                       LogProbabilityDensityFunction spatialNewbornDistrib,
                                       DataAssociationFactory assocFactory)
Constructor

Parameters:
rand -
Z - vector of observations
observationDistrib - observation likelihood that can be evaluated for each observation independently
spatialClutterDistrib -
assocFactory -
Method Detail

drawSample

public abstract DataAssociation drawSample()
Description copied from interface: SamplingDistribution
Generate a new sample from this density. This method should create a new object.

Specified by:
drawSample in interface SamplingDistribution<DataAssociation>
Returns:
new sample object

drawSampleDebug

public abstract DataAssociation drawSampleDebug(DataAssociation groundtruth,
                                                java.io.OutputStream ostream)

setNewObservations

public void setNewObservations(AbstractMultiState<S> Z,
                               AbstractMultiObservationDistributionIndep<S,T> observationDistrib)