lti::ransacEstimator Class Reference

This class estimates a transform using the Ransac algorithm. More...

#include <ltiRansacEstimator.h>

Inheritance diagram for lti::ransacEstimator:

[legend]

Collaboration diagram for lti::ransacEstimator:

[legend]

List of all members.

Classes
class	parameters
	The parameters for the class ransacEstimator. More...
Public Member Functions
	ransacEstimator ()
	ransacEstimator (const parameters &par)
	ransacEstimator (const ransacEstimator &other)
virtual	~ransacEstimator ()
virtual const char *	getTypeName () const
virtual bool	apply (const matrix< fpoint > &src, fvector &dest) const
virtual bool	apply (const matrix< fpoint > &src, fvector &dest, fvector &error) const
virtual bool	apply (const matrix< dpoint > &src, dvector &dest) const
virtual bool	apply (const matrix< dpoint > &src, dvector &dest, dvector &error) const
ransacEstimator &	copy (const ransacEstimator &other)
ransacEstimator &	operator= (const ransacEstimator &other)
virtual functor *	clone () const
const parameters &	getParameters () const

Detailed Description

This class estimates a transform using the Ransac algorithm.

The Ransac algorithm maximizes the number of inliers. At each iteration a subset of points/correspondences is drawn from which the transform is computed.

Theoretically the Ransac algorithm copes with up to 90% outliers. It is advised though, to verify the estimated transform and repeat the ransac estimation in case of a bad estimation. For the homography estimators this may be performed by the homographyVerfication (coming up soon).

Constructor & Destructor Documentation

lti::ransacEstimator::ransacEstimator ( )

Default constructor.

lti::ransacEstimator::ransacEstimator ( const parameters & par )

Construct a functor using the given parameters.

lti::ransacEstimator::ransacEstimator ( const ransacEstimator & other )

Copy constructor.

Parameters:

other

the object to be copied

virtual lti::ransacEstimator::~ransacEstimator ( ) [virtual]

Destructor.

Member Function Documentation

virtual bool lti::ransacEstimator::apply	(	const matrix< dpoint > &	src,
		dvector &	dest,
		dvector &	error
	)			const `[virtual]`

Estimates a transform from the supplied point sets.

All points of one point set give a matrix row, whereas all elements of a specifec correspondence stand in a matrix column.

Parameters:

	src	std::vector<vector<dpoint> > with the point sets.
	dest	dvector the estimated transform.
	error	fvector containing the deviation of each point from the estimated transform. Usually this is the residual, also a if a squared residual is computed inside the transform estimator.

Returns:: true if apply successful or false otherwise.

Implements lti::monteCarloEstimator.

virtual bool lti::ransacEstimator::apply	(	const matrix< dpoint > &	src,
		dvector &	dest
	)			const `[virtual]`

Estimates a transform from the supplied point sets.

All points of one point set give a matrix row, whereas all elements of a specifec correspondence stand in a matrix column.

Parameters:

	src	std::vector<vector<dpoint> > with the point sets.
	dest	dvector the estimated transform.

Returns:: true if apply successful or false otherwise.

Implements lti::monteCarloEstimator.

virtual bool lti::ransacEstimator::apply	(	const matrix< fpoint > &	src,
		fvector &	dest,
		fvector &	error
	)			const `[virtual]`

Estimates a transform from the supplied point sets.

All points of one point set give a matrix row, whereas all elements of a specifec correspondence stand in a matrix column.

Parameters:

	src	std::vector<vector<dpoint> > with the point sets.
	dest	fvector the estimated transform.
	error	fvector containing the deviation of each point from the estimated transform. Usually this is the residual, also a if a squared residual is computed inside the transform estimator.