imaging::BoundaryDiscretizer< N > Class Template Reference
[Shape Module]

Abstract class interface for discretizations of shape boundaries. More...

#include <BoundaryDiscretizer.hpp>

List of all members.

Public Member Functions
	BoundaryDiscretizer (size_t n_points)
size_t	n_points () const
virtual void	evaluate (size_t i, ublas::fixed_vector< float_t, SHAPE_DIMENSION > &point, ublas::fixed_vector< float_t, SHAPE_DIMENSION > &normal, float_t &curvature) const =0
ublas::fixed_vector< float_t, SHAPE_DIMENSION >	operator() (size_t i, ublas::fixed_vector< float_t, SHAPE_DIMENSION > &normal, float_t &curvature) const
ublas::fixed_vector< float_t, SHAPE_DIMENSION >	operator() (size_t i, ublas::fixed_vector< float_t, SHAPE_DIMENSION > &normal) const
ublas::fixed_vector< float_t, SHAPE_DIMENSION >	operator() (size_t i) const
template<class const_accessort_t>
const_accessort_t::data_t	integrate (const const_accessort_t &image) const
float_t	compute_boundary_area () const
template<class const_vector_accessor_t>
float_t	integrate_vector_field (const const_vector_accessor_t &vector_field) const
Box< SHAPE_DIMENSION >	compute_bounding_box () const
Static Public Attributes
static const size_t	SHAPE_DIMENSION = N
Protected Attributes
size_t	_n_points

---

Detailed Description

template<size_t N>
class imaging::BoundaryDiscretizer< N >

Abstract class interface for discretizations of shape boundaries.

Objects derived from this class can be evaluated in a finite number of sample points on the boundary of a shape. In addition to the coordinates of the sample point they must provide the outer normal in that point. The length of the normal must be such that the lengthes of all normals sum up to the area of the shape boundary. In other words, the length of the normal corresponds to the area of the infinitesimal boundary element around the normal.

---

Constructor & Destructor Documentation

template<size_t N>

imaging::BoundaryDiscretizer< N >::BoundaryDiscretizer

(

size_t

n_points

)

[inline]

Construct a discretization with n_points discretization points.

---

Member Function Documentation

template<size_t N>

size_t imaging::BoundaryDiscretizer< N >::n_points

(

)

const [inline]

Returns the number of discretization points.

References imaging::BoundaryDiscretizer< N >::_n_points.

Referenced by imaging::BoundaryDiscretizer< N >::compute_boundary_area(), imaging::BoundaryDiscretizer< N >::compute_bounding_box(), imaging::compute_intersection_volume(), imaging::BoundaryDiscretizer< N >::integrate(), imaging::BoundaryDiscretizer< N >::integrate_vector_field(), and imaging::triangulate_shape().

template<size_t N>

virtual void imaging::BoundaryDiscretizer< N >::evaluate	(	size_t	i,
		ublas::fixed_vector< float_t, SHAPE_DIMENSION > &	point,
		ublas::fixed_vector< float_t, SHAPE_DIMENSION > &	normal,
		float_t &	curvature
	)			const [pure virtual]

Sets point to the coordinates of the i-th disretization point and stores the boundary normal in this point in normal. The normal must be scaled to the area of the infinitesimal boundary element around the point. This function must be implemented in classes derived from BoundaryDiscretizer. For the actual evaluation of a discretization point the user should use operator().

Referenced by imaging::BoundaryDiscretizer< N >::operator()().

template<size_t N>

ublas::fixed_vector<float_t, SHAPE_DIMENSION> imaging::BoundaryDiscretizer< N >::operator()	(	size_t	i,
		ublas::fixed_vector< float_t, SHAPE_DIMENSION > &	normal,
		float_t &	curvature
	)			const [inline]

Evaluates the coordinates of the i-th discretization point, sets normal to the boundary normal and curvature to the curvature in this point. The normal is scaled to the area of the infinitesimal boundary element around the point.

References imaging::BoundaryDiscretizer< N >::evaluate().

Referenced by imaging::BoundaryDiscretizer< N >::operator()().

template<size_t N>

ublas::fixed_vector<float_t, SHAPE_DIMENSION> imaging::BoundaryDiscretizer< N >::operator()	(	size_t	i,
		ublas::fixed_vector< float_t, SHAPE_DIMENSION > &	normal
	)			const [inline]

Evaluates the coordinates of the i-th discretization point and sets normal to the boundary normal in this point. The normal is scaled to the area of the infinitesimal boundary element around the point.

References imaging::BoundaryDiscretizer< N >::operator()().

template<size_t N>

ublas::fixed_vector<float_t, SHAPE_DIMENSION> imaging::BoundaryDiscretizer< N >::operator()

(

size_t

i

)

const [inline]

Evaluates the coordinates of the i-th discretization point.

References imaging::BoundaryDiscretizer< N >::operator()().

template<size_t N>

template<class const_accessort_t>

const_accessort_t::data_t imaging::BoundaryDiscretizer< N >::integrate

(

const const_accessort_t &

image

)

const [inline]

Integrate an image over a boundary discretization.

References imaging::BoundaryDiscretizer< N >::n_points().

template<size_t N>

float_t imaging::BoundaryDiscretizer< N >::compute_boundary_area

(

)

const [inline]

Compute the (n-1)-dimensional boundary area of a boundary discretization.

References imaging::BoundaryDiscretizer< N >::n_points().

template<size_t N>

template<class const_vector_accessor_t>

float_t imaging::BoundaryDiscretizer< N >::integrate_vector_field

(

const const_vector_accessor_t &

vector_field

)

const [inline]

Integrate a vector field over a boundary discretization.

References imaging::BoundaryDiscretizer< N >::n_points().

template<size_t N>

Box<SHAPE_DIMENSION> imaging::BoundaryDiscretizer< N >::compute_bounding_box

(

)

const [inline]

Compute the bounding box of a boundary discretization.

References imaging::BoundaryDiscretizer< N >::n_points(), and imaging::BoundaryDiscretizer< N >::SHAPE_DIMENSION.

---

Member Data Documentation

template<size_t N>

size_t imaging::BoundaryDiscretizer< N >::_n_points [protected]

Number of discretization points.

Referenced by imaging::BoundaryDiscretizer< N >::n_points().

template<size_t N>

const size_t imaging::BoundaryDiscretizer< N >::SHAPE_DIMENSION = N [static]

The spatial dimension of the shape boundary. N is 2 for planar shapes and 3 for volumes.

Referenced by imaging::BoundaryDiscretizer< N >::compute_bounding_box().

---

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

• imaging2/shape/BoundaryDiscretizer.hpp