Bial::FuzzyCMeans< D > Class Template Reference

#include <FuzzyCMeans.hpp>

Collaboration diagram for Bial::FuzzyCMeans< D >:

Public Member Functions
	FuzzyCMeans (const Feature< D > &new_feats, int new_clusters, float new_m, double new_epsilon, size_t nmax_iterations=30)
	Basic constructor. More...
Matrix< double >	Run (bool verbose=false)
	Computes fuzzy c-means from 3D image pixels, according to the choosen feature vector. More...
Matrix< double >	GetMembership () const
	Returns the matrix containing the fuzzy degree of membership computed by fuzzy C-Means. More...
void	UniformCentroidInitialization ()
	Initializes centroid positions of the clusters with uniform space between them. More...
template<class O >
O &	PrintIteration (O &os, int itr, double max_change) const
	Prints Fuzzy c-means iteration status to output stream os. More...

Detailed Description

template<class D>
class Bial::FuzzyCMeans< D >

Definition at line 27 of file FuzzyCMeans.hpp.

Constructor & Destructor Documentation

template<class D>

Bial::FuzzyCMeans< D >::FuzzyCMeans	(	const Feature< D > &	new_feats,
		int	new_clusters,
		float	new_m,
		double	new_epsilon,
		size_t	nmax_iterations = 30
	)

Basic constructor.

Date

2012/Jun/25

Parameters

feats	3D image feature vector.
clusters	number of expected clusters.
m	power of degree of membership. m > 1. Suggested: 2
epsilon	termination criterion. 0.0 < epsilon < 1.0. Values closer to 0.0, are more precise, but demands more computational effort.

Returns

none.

Warning

none.

Member Function Documentation

template<class D>

Matrix< double > Bial::FuzzyCMeans< D >::GetMembership

(

)

const

Returns the matrix containing the fuzzy degree of membership computed by fuzzy C-Means.

Date

2012/Jun/26

Parameters

none.

Returns

The fuzzy degree of membership.

Warning

none.

template<class D>

template<class O >

O& Bial::FuzzyCMeans< D >::PrintIteration	(	O &	os,
		int	itr,
		double	max_change
	)		const

Prints Fuzzy c-means iteration status to output stream os.

Date

2012/Aug/20

Parameters

os	an output stream.
itr	Number of the iteration.
max_change	Maximum change from last iteration.

Returns

The output stream.

Warning

none.

template<class D>

Matrix< double > Bial::FuzzyCMeans< D >::Run

(

bool

verbose = false

)

Computes fuzzy c-means from 3D image pixels, according to the choosen feature vector.

Date

2012/Jun/25

Parameters

verbose

verbose mode.

Returns

The fuzzy degree of membership.

Warning

It is the simplest method that does not takes into account image inhomogeneity. If centroid positions are not initialized, void UniformCentroidInitialization() is called.

template<class D>

void Bial::FuzzyCMeans< D >::UniformCentroidInitialization

(

)

Initializes centroid positions of the clusters with uniform space between them.

Date

2012/Jun/25

Parameters

none.

Returns

none.

Warning

none.

---

The documentation for this class was generated from the following file:

• bial/inc/FuzzyCMeans.hpp