Developer Documentation
RAVL, Recognition And Vision Library
USER HOME PAGE CLASS LIST CONTENTS
Ravl - Math - Statistics - FMeanCovarianceC<unsigned int N>
 

  PUBLIC
FMeanCovarianceC::FMeanCovarianceC(void)
FMeanCovarianceC::FMeanCovarianceC(const FVectorC &)
FMeanCovarianceC::FMeanCovarianceC(const FMeanC &)
FMeanCovarianceC::FMeanCovarianceC(SizeT,const FVectorC &,const FMatrixC &)
FMeanCovarianceC::Copy(void) const
FMeanCovarianceC::FMeanCovarianceC(const SArray1dC> &)
FMeanCovarianceC::Number(void) const
FMeanCovarianceC::Mean(void) const
FMeanCovarianceC::Covariance(void) const
FMeanCovarianceC::SetZero(void)
FMeanCovarianceC::operator +=(const FVectorC &)
FMeanCovarianceC::operator -=(const FVectorC &)
FMeanCovarianceC::operator +=(const FMeanC &)
FMeanCovarianceC::operator -=(const FMeanC &)
FMeanCovarianceC::operator +=(const FMeanCovarianceC &)
FMeanCovarianceC::operator -=(const FMeanCovarianceC &)
FMeanCovarianceC::Add(const FVectorC &,const FVectorC &)
FMeanCovarianceC::Remove(const FVectorC &,const FVectorC &)
FMeanCovarianceC::SetSum(const FMeanCovarianceC &,const FMeanCovarianceC &)
FMeanCovarianceC::operator *(const FMeanCovarianceC &) const
FMeanCovarianceC::operator >>(istream &,FMeanCovarianceC &)

   FMeanCovarianceC<unsigned int N>   
 
Mean and covariance together in N-D space
 
include "Ravl/FMeanCovariance.hh"
Created:01/01/1994 
Source file:Ravl/Math/Statistics/MeanCovariance/FMeanCovariance.hh
User Level:Normal
Library:RavlMath
In Scope:RavlN

Comments:
The class FMeanCovarianceC serves for computation of the mean and the covariance matrix of a set of fixed dimensional data points.

Derived Classes: Variables:
FMeanC m;
The mean vector of this data set.

FMatrixC cov;
the covariance matrix of this data set.

Methods:
FMeanCovarianceC()
Constructor.
Creates zero mean and zero covariance matrix representing the fixed-dimensional set containing no data points.

FMeanCovarianceC(const FVectorC<N> & point)
Constructor
Creates the mean vector and zero covariance matrix representing the data set containing just one data point.

FMeanCovarianceC(const FMeanC<N> & mean)
Creates the mean vector and zero covariance matrix representing the data set represented by the 'mean'.

FMeanCovarianceC(SizeT n,const FVectorC<N> & mean,const FMatrixC<N,N> & ncov)
Creates the mean vector and zero covariance matrix representing the data set containing 'n' points and represented by the 'mean' and the covariance matrix 'cov'.

FMeanCovarianceC<N> Copy() const
Returns a new physical copy of this object.

FMeanCovarianceC(const SArray1dC<FVectorC<N>> & data)
Compute the mean and covariance of an array of vectors.

Information about an object
SizeT Number() const
Returns the number of data points which are represented by this object.

const FVectorC<N> & Mean() const
Access the mean.
Returns the mean vector of data points which are represented by this object.

const FMatrixC<N,N> & Covariance() const
Access the covariance.
Returns the covariance matrix of data points which are represented by this object. Object modification -------------------

const FMeanCovarianceC<N> & SetZero()
Total initialization of this object resulting in the representation the empty set of data points.

const FMeanCovarianceC<N> & operator +=(const FVectorC<N> & point)
Adds one point to the set of data points.

const FMeanCovarianceC<N> & operator -=(const FVectorC<N> & point)
Removes one point from the set of data points.
Be carefull to remove a point which was already added to the set, otherwise the representation will not describe a real set.

const FMeanCovarianceC<N> & operator +=(const FMeanC<N> & mean)
Adds a number of data poits represented by the 'mean' and zero covariance matrix to this set.

const FMeanCovarianceC<N> & operator -=(const FMeanC<N> & mean)
Removes a number of data poits represented by the 'mean' and zero covariance matrix from this set.
Be carefull to remove points which were already added to the set, otherwise the representation will not describe a real set.

const FMeanCovarianceC<N> & operator +=(const FMeanCovarianceC<N> & meanCov)
Adds a number of data points represented by the 'meanCov' structure to this set.

const FMeanCovarianceC<N> & operator -=(const FMeanCovarianceC<N> & meanCov)
Removes a number of data points represented by the 'meanCov' structure from this set.
Be carefull to remove points which were already added to the set, otherwise the representation will not describe a real set.

const FMeanCovarianceC<N> & Add(const FVectorC<N> & point,const FVectorC<N> & var)
Updates the mean and the covariance matrix by adding one N-d point whose coordinates are known with the error described by the diagonal convariance matrix represented byt the vector 'var'.

const FMeanCovarianceC<N> & Remove(const FVectorC<N> & point,const FVectorC<N> & var)
Updates the mean and the covariance matrix by removing one N-d point whose coordinates are known with the error described by the diagonal convariance matrix represented byt the vector 'var'. Be carefull to remove the point which was already added to the set, otherwise the representation will not describe a real set.

const FMeanCovarianceC<N> & SetSum(const FMeanCovarianceC<N> & meanCov1,const FMeanCovarianceC<N> & meanCov2)
This object is set to be the union of two set of data points 'meanCov1' and 'meanCov2'.

FMeanCovarianceC<N> operator *(const FMeanCovarianceC<N> & oth) const
Calculate the product of the two probability density functions.
This assumes the estimates of the distributions are accurate. (The number of samples is ignored)

istream & operator >>(istream & inS,FMeanCovarianceC<N> & meanCov)


Maintainer:Radek Marik, Created: 01/01/1994, Documentation by CxxDoc: Tue Aug 13 09:59:30 2002