#include <CGAL/Shape_detection/Region_growing/Region_growing_on_point_set/K_neighbor_query.h>

Definition

K nearest neighbors search in a set of Kernel::Point_2 or Kernel::Point_3.

This class returns the K nearest neighbors of a query point in a point set.

Template Parameters

GeomTraits	must be a model of `Kernel`.
InputRange	must be a model of `ConstRange` whose iterator type is `RandomAccessIterator`.
PointMap	must be an `LvaluePropertyMap` whose key type is the value type of the input range and value type is `Kernel::Point_2` or `Kernel::Point_3`.

Is Model Of:: NeighborQuery

Examples:: Shape_detection/region_growing_on_point_set_3.cpp.

Initialization
	K_neighbor_query (const InputRange &input_range, const std::size_t k=12, const PointMap point_map=PointMap())
	initializes a Kd-tree with input points. More...

Access
void	operator() (const std::size_t query_index, std::vector< std::size_t > &neighbors) const
	implements `NeighborQuery::operator()()`. More...

Constructor & Destructor Documentation

◆ K_neighbor_query()

template<typename GeomTraits , typename InputRange , typename PointMap >

CGAL::Shape_detection::Point_set::K_neighbor_query< GeomTraits, InputRange, PointMap >::K_neighbor_query	(	const InputRange &	input_range,
		const std::size_t	k = `12`,
		const PointMap	point_map = `PointMap()`
	)

initializes a Kd-tree with input points.

Parameters

input_range	an instance of `InputRange` with 2D or 3D points
k	the number of returned neighbors per each query point. Default is 12.
point_map	an instance of `PointMap` that maps an item from `input_range` to `Kernel::Point_2` or to `Kernel::Point_3`

Precondition: input_range.size() > 0; k > 0

Member Function Documentation

◆ operator()()

template<typename GeomTraits , typename InputRange , typename PointMap >

void CGAL::Shape_detection::Point_set::K_neighbor_query< GeomTraits, InputRange, PointMap >::operator()	(	const std::size_t	query_index,
		std::vector< std::size_t > &	neighbors
	)		const

implements NeighborQuery::operator()().

This operator finds indices of the k closest points to the point with the index query_index using a Kd-tree. These indices are returned in neighbors.