lvr2::KDTree Class Reference

a kd-Tree Implementation for nearest Neighbor searches More...

#include <KDTree.hpp>

Inheritance diagram for lvr2::KDTree:

Public Types
using	KDTreePtr = std::shared_ptr< KDTree >
using	Neighbor = Point *
using	Point = Vector3< PointT >
using	PointT = float

Public Member Functions
template<typename T >
bool	nearestNeighbor (const Vector3< T > &point, Neighbor &neighbor, double &distance, double maxDistance=std::numeric_limits< double >::infinity()) const
	Finds the nearest neighbor of 'point' that is within 'maxDistance' (defaults to infinity). The resulting neighbor is written into 'neighbor' (or nullptr if none is found). More...
virtual	~KDTree ()=default

Static Public Member Functions
static std::shared_ptr< KDTree >	create (SLAMScanPtr scan, int maxLeafSize=20)
	Creates a new KDTree from the given Scan. More...
static size_t	nearestNeighbors (KDTreePtr tree, SLAMScanPtr scan, Neighbor *neighbors, double maxDistance, Vector3d &centroid_m, Vector3d &centroid_d)
	Finds the nearest neighbors of all points in a Scan using a pre-generated KDTree. More...
static size_t	nearestNeighbors (KDTreePtr tree, SLAMScanPtr scan, KDTree::Neighbor *neighbors, double maxDistance)
	Finds the nearest neighbors of all points in a Scan using a pre-generated KDTree. More...

Protected Member Functions
	KDTree ()=default
	KDTree (const KDTree &&)=delete
virtual void	nnInternal (const Point &point, Neighbor &neighbor, double &maxDist) const =0

Protected Attributes
boost::shared_array< Point >	points

Friends
class	KDNode

Detailed Description

a kd-Tree Implementation for nearest Neighbor searches

Definition at line 51 of file KDTree.hpp.

Member Typedef Documentation

KDTreePtr

using lvr2::KDTree::KDTreePtr = std::shared_ptr<KDTree>

Definition at line 57 of file KDTree.hpp.

Neighbor

using lvr2::KDTree::Neighbor = Point*

Definition at line 56 of file KDTree.hpp.

Point

using lvr2::KDTree::Point = Vector3<PointT>

Definition at line 55 of file KDTree.hpp.

PointT

using lvr2::KDTree::PointT = float

Definition at line 54 of file KDTree.hpp.

Constructor & Destructor Documentation

~KDTree()

virtual lvr2::KDTree::~KDTree ( )

virtualdefault

KDTree() [1/2]

lvr2::KDTree::KDTree ( )

protecteddefault

KDTree() [2/2]

lvr2::KDTree::KDTree ( const KDTree &&  )

protecteddelete

Member Function Documentation

create()

KDTreePtr lvr2::KDTree::create ( SLAMScanPtr  scan, int  maxLeafSize = 20  )

static

Creates a new KDTree from the given Scan.

Parameters

points	The Point Cloud
n	The number of points in 'points'
maxLeafSize	The maximum number of points to use for a Leaf in the Tree

Definition at line 153 of file KDTree.cpp.

nearestNeighbor()

template<typename T >

bool lvr2::KDTree::nearestNeighbor ( const Vector3< T > &  point, Neighbor &  neighbor, double &  distance, double  maxDistance = std::numeric_limits<double>::infinity()  ) const

inline

Finds the nearest neighbor of 'point' that is within 'maxDistance' (defaults to infinity). The resulting neighbor is written into 'neighbor' (or nullptr if none is found).

Parameters

point	The Point whose neighbor is searched
neighbor	A Pointer that is set to the neighbor or nullptr if none is found
distance	The final distance between point and neighbor
maxDistance	The maximum distance allowed between neighbors. Setting this value significantly speeds up the search.

Returns

bool true if a neighbors was found, false otherwise

Definition at line 80 of file KDTree.hpp.

nearestNeighbors() [1/2]

size_t lvr2::KDTree::nearestNeighbors ( KDTreePtr  tree, SLAMScanPtr  scan, KDTree::Neighbor *  neighbors, double  maxDistance, Vector3d &  centroid_m, Vector3d &  centroid_d  )

static

Finds the nearest neighbors of all points in a Scan using a pre-generated KDTree.

Parameters

tree	The KDTree to search in
scan	The Scan to search for
neighbors	An array to store the results in. neighbors[i] is set to a Pointer to the neighbor of points[i] or nullptr if none was found
maxDistance	The maximum Distance for a Neighbor
centroid_m	Will be set to the average of all Points in 'neighbors'
centroid_d	Will be set to the average of all Points in 'points' that have neighbors

Returns

size_t The number of neighbors that were found

Definition at line 176 of file KDTree.cpp.

nearestNeighbors() [2/2]

size_t lvr2::KDTree::nearestNeighbors ( KDTreePtr  tree, SLAMScanPtr  scan, KDTree::Neighbor *  neighbors, double  maxDistance  )

static

Finds the nearest neighbors of all points in a Scan using a pre-generated KDTree.

Parameters

tree	The KDTree to search in
scan	The Scan to search for
neighbors	An array to store the results in. neighbors[i] is set to a Pointer to the neighbor of points[i] or nullptr if none was found
maxDistance	The maximum Distance for a Neighbor

Returns

size_t The number of neighbors that were found

Definition at line 198 of file KDTree.cpp.

nnInternal()

virtual void lvr2::KDTree::nnInternal ( const Point &  point, Neighbor &  neighbor, double &  maxDist  ) const

protectedpure virtual

Implemented in lvr2::KDLeaf, and lvr2::KDNode.

Friends And Related Function Documentation

KDNode

friend class KDNode

friend

Definition at line 130 of file KDTree.hpp.

Member Data Documentation

points

boost::shared_array<Point> lvr2::KDTree::points

protected

Definition at line 132 of file KDTree.hpp.

---

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

• KDTree.hpp
• KDTree.cpp