TRTK::Fit< T > Class Template Reference

Base class for all fitting classes. More...

#include <Fit.hpp>

Inheritance diagram for TRTK::Fit< T >:

List of all members.

Public Types
enum	Error {   DATA_POINTS_TOO_SIMILAR, INFINITY_NOT_AVAILABLE, NAN_NOT_AVAILABLE, NOT_ENOUGH_POINTS,   UNKNOWN_ERROR, WRONG_POINT_SIZE }
typedef Eigen::Array< T, Eigen::Dynamic, 1 >	ArrayXT
	General-purpose array of arbitrary size with value type T.
typedef Eigen::Matrix< T, Eigen::Dynamic, Eigen::Dynamic >	MatrixXT
	Matrix of arbitrary size with value type T.
typedef Eigen::Matrix< T, Eigen::Dynamic, 1 >	VectorXT
	Column vector of arbitrary size with value type T.
typedef Eigen::Matrix< T, 2, 1 >	Vector2T
	2D column vector with value type T.
typedef Eigen::Matrix< T, 3, 1 >	Vector3T
	3D column vector with value type T.
typedef Eigen::Matrix< T, 4, 1 >	Vector4T
	4D column vector with value type T.
typedef Eigen::Matrix< T, 2, 2 >	Matrix2T
	2 x 2 matrix with value type T.
typedef Eigen::Matrix< T, 3, 3 >	Matrix3T
	3 x 3 matrix with value type T.
Public Member Functions
virtual	~Fit ()
	Destructor.
virtual void	compute ()=0
	Performs the regression analysis.
virtual T	getDistanceTo (const Coordinate< T > &point) const =0
	Returns the distance/deviation of the given data point to the estimated model.
virtual T	getRMS () const =0
	Returns the root mean square error of the model fitting.
virtual unsigned	getNumberPointsRequired () const =0
	Returns the minimum number of data points required to compute a model.
virtual void	setPoints (const std::vector< Coordinate< T > > &)=0
	Sets the data point set.
Public Attributes
EIGEN_MAKE_ALIGNED_OPERATOR_NEW typedef T	value_type
	Internally used value type (should be a floating point type).

---

Detailed Description

template<class T>
class TRTK::Fit< T >

Base class for all fitting classes.

Template Parameters:

T	floating point type

Every fitting algorithm must provide the method setPoints() with which the data points can be passed. In general, a model is only computed by calling the member function compute(). Bear that in mind before calling the getter functions. After having estimated the regression model, the estimation error can be determined by getRMS(). The minimum number of points needed to be able to compute a model is given by getMinimumNumberOfPoints().

See also:

Fit2D, Fit3D

Author:

Christoph Haenisch, Fabian Killus

Version:

0.2.0

Date:

2012-03-29

Definition at line 53 of file Fit.hpp.

---

Member Enumeration Documentation

template<class T>

enum TRTK::Fit::Error

Enumerator:

DATA_POINTS_TOO_SIMILAR	The data points are to similar.
INFINITY_NOT_AVAILABLE	The type T cannot represent infinity (see std::numeric_limits<T>).
NAN_NOT_AVAILABLE	The type T cannot represent NaN (see std::numeric_limits<T>).
NOT_ENOUGH_POINTS	More points are required to estimate the transformation.
UNKNOWN_ERROR	An unknown error occured.
WRONG_POINT_SIZE	One or more points have a wrong size.

Definition at line 69 of file Fit.hpp.

---

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

• D:/Programmierung/C/TRTK/include/TRTK/Fit.hpp