Function

CGAL::sibson_gradient_fitting

Definition

The function sibson_gradient_fitting approximates the gradient of a function at a point p given natural neighbor coordinates for p and its neighbors' function values. The approximation method is described in [Sib81]. Further functions are provided to fit the gradient for all data points that lie inside the convex hull of the data points. One function exists for each type of natural neighbor coordinates.

#include <CGAL/sibson_gradient_fitting.h>

template < class ForwardIterator, class Functor, class Traits>
typename Traits::Vector_d
sibson_gradient_fitting ( ForwardIterator first,
ForwardIterator beyond,
typename std::iterator_traits<ForwardIterator>::value_type::second_type norm,
typename std::iterator_traits<ForwardIterator>::value_type::first_type p,
Functor f,
Traits traits)
This function estimates the gradient of a function at the point p given natural neighbor coordinates of p in the range [. first, beyond.) and the function values of the neighbors provided by the functor f. norm is the normalization factor of the barycentric coordinates.

Requirements

  1. ForwardIterator::value_type is a pair of point/coordinate value, thus ForwardIterator::value_type::first_type is equivalent to a point and ForwardIterator::value_type::second_type is a number type.
  2. Functor::argument_type must be equivalent to ForwardIterator::value_type::first_type and Functor::result_type is the function value type. It must provide a multiplication and addition operation with the type ForwardIterator::value_type::second_type.
  3. Traits is a model of the concept GradientFittingTraits .

template < class Dt, class OutputIterator, class Functor, class Traits>
OutputIterator sibson_gradient_fitting_nn_2 ( Dt dt, OutputIterator out, Functor f, Traits traits)
estimates the function gradients at all vertices of dt that lie inside the convex hull using the coordinates computed by the function natural_neighbor_coordinates_2. OutputIterator::value_type is a pair associating a point to a vector. The sequence of point/gradient pairs computed by this function is placed starting at out. The function returns an iterator that is placed past-the-end of the resulting sequence. The requirements are the same as above. The template class Dt must be equivalent to Delaunay_triangulation_2<Gt, Tds>.

template < class Rt, class OutputIterator, class Functor, class Traits>
OutputIterator sibson_gradient_fitting_rn_2 ( Rt rt, OutputIterator out, Functor f, Traits traits)
estimates the function gradients at all vertices of rt that lie inside the convex hull using the coordinates computed by the function regular_neighbor_coordinates_2. OutputIterator::value_type is a pair associating a point to a vector. The sequence of point/gradient pairs computed by this function is placed starting at out. The function returns an iterator that is placed past-the-end of the resulting sequence. The requirements are the same as above. The template class Rt must be equivalent to Regular_triangulation_2<Gt, Tds>.

See Also

CGAL::linear_interpolation
CGAL::sibson_c1_interpolation
CGAL::farin_c1_interpolation
CGAL::quadratic_interpolation
CGAL::Interpolation_gradient_fitting_traits_2<K>
CGAL::natural_neighbor_coordinates_2
CGAL::regular_neighbor_coordinates_2
CGAL::surface_neighbor_coordinates_3

Implementation

This function implements Sibson's gradient estimation method based on natural neighbor coordinates [Sib81].