Package org.ddogleg.nn.alg.searches

Class KdTreeSearchNBbf<P>

java.lang.Object

org.ddogleg.nn.alg.searches.KdTreeSearchBestBinFirst<P>

org.ddogleg.nn.alg.searches.KdTreeSearchNBbf<P>

All Implemented Interfaces:

KdTreeSearchN<P>

public class KdTreeSearchNBbf<P>
extends KdTreeSearchBestBinFirst<P>
implements KdTreeSearchN<P>

Implementation of KdTreeSearchBestBinFirst which searches for the N nearest-neighbors.

Nested Class Summary

Nested classes/interfaces inherited from class org.ddogleg.nn.alg.searches.KdTreeSearchBestBinFirst

KdTreeSearchBestBinFirst.Helper

Field Summary

Fields inherited from class org.ddogleg.nn.alg.searches.KdTreeSearchBestBinFirst

bestDistanceSq, maxNodesSearched, N, numNodesSearched

Constructor Summary

Constructors

Constructor	Description
KdTreeSearchNBbf(KdTreeDistance<P> distance, int maxNodesSearched)	Configures the search

Method Summary

Modifier and Type	Method	Description
protected boolean	canImprove(double distanceSq)	Checks to see if it is possible for this distance to improve upon the current best
protected void	checkBestDistance(KdTree.Node node, P target)	Checks to see if the current node's point is the closet point found so far
KdTreeSearchN<P>	copy()	Creates a copy of this search with the same configuration.
void	findNeighbor(P target, int searchN, DogArray<KdTreeResult> results)	Searches for the N nearest-neighbors to the target.
void	setTree(Object tree)	Specifies the tree which will be searched.

Methods inherited from class org.ddogleg.nn.alg.searches.KdTreeSearchBestBinFirst

_findClosest, addToQueue, searchNode, setMaxDistance, setTree, setTrees

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface org.ddogleg.nn.alg.KdTreeSearchN

setMaxDistance

Constructor Details

KdTreeSearchNBbf

public KdTreeSearchNBbf(KdTreeDistance<P> distance,
 int maxNodesSearched)

Configures the search

Parameters:

maxNodesSearched - Maximum number of nodes it will search. Used to limit CPU time.

Method Details

setTree

public void setTree(Object tree)

Description copied from interface: KdTreeSearchN

Specifies the tree which will be searched. The type of object is implementation specific

Specified by:

setTree in interface KdTreeSearchN<P>

Parameters:

tree - search tree

findNeighbor

public void findNeighbor(P target,
 int searchN,
 DogArray<KdTreeResult> results)

Description copied from interface: KdTreeSearchN

Searches for the N nearest-neighbors to the target. The results are added to the 'results' list.

Specified by:

findNeighbor in interface KdTreeSearchN<P>

Parameters:

target - Point whose nearest neighbor is being searched for

searchN - Number of closest points it will find. Must be >= 1

results - Storage for the found neighbors.

copy

public KdTreeSearchN<P> copy()

Description copied from interface: KdTreeSearchN

Creates a copy of this search with the same configuration. workspace isn't copied

Specified by:

copy in interface KdTreeSearchN<P>

checkBestDistance

protected void checkBestDistance(KdTree.Node node,
 P target)

Checks to see if the current node's point is the closet point found so far

Specified by:

checkBestDistance in class KdTreeSearchBestBinFirst<P>

canImprove

protected boolean canImprove(double distanceSq)

Description copied from class: KdTreeSearchBestBinFirst

Checks to see if it is possible for this distance to improve upon the current best

Specified by:

canImprove in class KdTreeSearchBestBinFirst<P>

Parameters:

distanceSq - The distance being considered

Returns:

true if it can be better or false if not