Point Cloud Library (PCL)  1.7.0
octree_search.h
1 /*
2  * Software License Agreement (BSD License)
3  *
4  * Point Cloud Library (PCL) - www.pointclouds.org
5  * Copyright (c) 2010-2012, Willow Garage, Inc.
6  *
7  * All rights reserved.
8  *
9  * Redistribution and use in source and binary forms, with or without
10  * modification, are permitted provided that the following conditions
11  * are met:
12  *
13  * * Redistributions of source code must retain the above copyright
14  * notice, this list of conditions and the following disclaimer.
15  * * Redistributions in binary form must reproduce the above
16  * copyright notice, this list of conditions and the following
17  * disclaimer in the documentation and/or other materials provided
18  * with the distribution.
19  * * Neither the name of Willow Garage, Inc. nor the names of its
20  * contributors may be used to endorse or promote products derived
21  * from this software without specific prior written permission.
22  *
23  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
24  * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
25  * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
26  * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
27  * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
28  * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
29  * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
30  * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
31  * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
32  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
33  * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
34  * POSSIBILITY OF SUCH DAMAGE.
35  *
36  * $Id$
37  */
38 
39 #ifndef PCL_OCTREE_SEARCH_H_
40 #define PCL_OCTREE_SEARCH_H_
41 
42 #include <pcl/point_cloud.h>
43 #include <pcl/point_types.h>
44 
45 #include "octree_pointcloud.h"
46 
47 namespace pcl
48 {
49  namespace octree
50  {
51  /** \brief @b Octree pointcloud search class
52  * \note This class provides several methods for spatial neighbor search based on octree structure
53  * \note typename: PointT: type of point used in pointcloud
54  * \ingroup octree
55  * \author Julius Kammerl (julius@kammerl.de)
56  */
57  template<typename PointT, typename LeafContainerT = OctreeContainerPointIndices , typename BranchContainerT = OctreeContainerEmpty >
58  class OctreePointCloudSearch : public OctreePointCloud<PointT, LeafContainerT, BranchContainerT>
59  {
60  public:
61  // public typedefs
62  typedef boost::shared_ptr<std::vector<int> > IndicesPtr;
63  typedef boost::shared_ptr<const std::vector<int> > IndicesConstPtr;
64 
66  typedef boost::shared_ptr<PointCloud> PointCloudPtr;
67  typedef boost::shared_ptr<const PointCloud> PointCloudConstPtr;
68 
69  // Boost shared pointers
70  typedef boost::shared_ptr<OctreePointCloudSearch<PointT, LeafContainerT, BranchContainerT> > Ptr;
71  typedef boost::shared_ptr<const OctreePointCloudSearch<PointT, LeafContainerT, BranchContainerT> > ConstPtr;
72 
73  // Eigen aligned allocator
74  typedef std::vector<PointT, Eigen::aligned_allocator<PointT> > AlignedPointTVector;
75 
77  typedef typename OctreeT::LeafNode LeafNode;
78  typedef typename OctreeT::BranchNode BranchNode;
79 
80  /** \brief Constructor.
81  * \param[in] resolution octree resolution at lowest octree level
82  */
83  OctreePointCloudSearch (const double resolution) :
84  OctreePointCloud<PointT, LeafContainerT, BranchContainerT> (resolution)
85  {
86  }
87 
88  /** \brief Empty class constructor. */
89  virtual
91  {
92  }
93 
94  /** \brief Search for neighbors within a voxel at given point
95  * \param[in] point point addressing a leaf node voxel
96  * \param[out] point_idx_data the resultant indices of the neighboring voxel points
97  * \return "true" if leaf node exist; "false" otherwise
98  */
99  bool
100  voxelSearch (const PointT& point, std::vector<int>& point_idx_data);
101 
102  /** \brief Search for neighbors within a voxel at given point referenced by a point index
103  * \param[in] index the index in input cloud defining the query point
104  * \param[out] point_idx_data the resultant indices of the neighboring voxel points
105  * \return "true" if leaf node exist; "false" otherwise
106  */
107  bool
108  voxelSearch (const int index, std::vector<int>& point_idx_data);
109 
110  /** \brief Search for k-nearest neighbors at the query point.
111  * \param[in] cloud the point cloud data
112  * \param[in] index the index in \a cloud representing the query point
113  * \param[in] k the number of neighbors to search for
114  * \param[out] k_indices the resultant indices of the neighboring points (must be resized to \a k a priori!)
115  * \param[out] k_sqr_distances the resultant squared distances to the neighboring points (must be resized to \a k
116  * a priori!)
117  * \return number of neighbors found
118  */
119  inline int
120  nearestKSearch (const PointCloud &cloud, int index, int k, std::vector<int> &k_indices,
121  std::vector<float> &k_sqr_distances)
122  {
123  return (nearestKSearch (cloud[index], k, k_indices, k_sqr_distances));
124  }
125 
126  /** \brief Search for k-nearest neighbors at given query point.
127  * \param[in] p_q the given query point
128  * \param[in] k the number of neighbors to search for
129  * \param[out] k_indices the resultant indices of the neighboring points (must be resized to k a priori!)
130  * \param[out] k_sqr_distances the resultant squared distances to the neighboring points (must be resized to k a priori!)
131  * \return number of neighbors found
132  */
133  int
134  nearestKSearch (const PointT &p_q, int k, std::vector<int> &k_indices,
135  std::vector<float> &k_sqr_distances);
136 
137  /** \brief Search for k-nearest neighbors at query point
138  * \param[in] index index representing the query point in the dataset given by \a setInputCloud.
139  * If indices were given in setInputCloud, index will be the position in the indices vector.
140  * \param[in] k the number of neighbors to search for
141  * \param[out] k_indices the resultant indices of the neighboring points (must be resized to \a k a priori!)
142  * \param[out] k_sqr_distances the resultant squared distances to the neighboring points (must be resized to \a k
143  * a priori!)
144  * \return number of neighbors found
145  */
146  int
147  nearestKSearch (int index, int k, std::vector<int> &k_indices, std::vector<float> &k_sqr_distances);
148 
149  /** \brief Search for approx. nearest neighbor at the query point.
150  * \param[in] cloud the point cloud data
151  * \param[in] query_index the index in \a cloud representing the query point
152  * \param[out] result_index the resultant index of the neighbor point
153  * \param[out] sqr_distance the resultant squared distance to the neighboring point
154  * \return number of neighbors found
155  */
156  inline void
157  approxNearestSearch (const PointCloud &cloud, int query_index, int &result_index, float &sqr_distance)
158  {
159  return (approxNearestSearch (cloud.points[query_index], result_index, sqr_distance));
160  }
161 
162  /** \brief Search for approx. nearest neighbor at the query point.
163  * \param[in] p_q the given query point
164  * \param[out] result_index the resultant index of the neighbor point
165  * \param[out] sqr_distance the resultant squared distance to the neighboring point
166  */
167  void
168  approxNearestSearch (const PointT &p_q, int &result_index, float &sqr_distance);
169 
170  /** \brief Search for approx. nearest neighbor at the query point.
171  * \param[in] query_index index representing the query point in the dataset given by \a setInputCloud.
172  * If indices were given in setInputCloud, index will be the position in the indices vector.
173  * \param[out] result_index the resultant index of the neighbor point
174  * \param[out] sqr_distance the resultant squared distance to the neighboring point
175  * \return number of neighbors found
176  */
177  void
178  approxNearestSearch (int query_index, int &result_index, float &sqr_distance);
179 
180  /** \brief Search for all neighbors of query point that are within a given radius.
181  * \param[in] cloud the point cloud data
182  * \param[in] index the index in \a cloud representing the query point
183  * \param[in] radius the radius of the sphere bounding all of p_q's neighbors
184  * \param[out] k_indices the resultant indices of the neighboring points
185  * \param[out] k_sqr_distances the resultant squared distances to the neighboring points
186  * \param[in] max_nn if given, bounds the maximum returned neighbors to this value
187  * \return number of neighbors found in radius
188  */
189  int
190  radiusSearch (const PointCloud &cloud, int index, double radius, std::vector<int> &k_indices,
191  std::vector<float> &k_sqr_distances, unsigned int max_nn = 0)
192  {
193  return (radiusSearch (cloud.points[index], radius, k_indices, k_sqr_distances, max_nn));
194  }
195 
196  /** \brief Search for all neighbors of query point that are within a given radius.
197  * \param[in] p_q the given query point
198  * \param[in] radius the radius of the sphere bounding all of p_q's neighbors
199  * \param[out] k_indices the resultant indices of the neighboring points
200  * \param[out] k_sqr_distances the resultant squared distances to the neighboring points
201  * \param[in] max_nn if given, bounds the maximum returned neighbors to this value
202  * \return number of neighbors found in radius
203  */
204  int
205  radiusSearch (const PointT &p_q, const double radius, std::vector<int> &k_indices,
206  std::vector<float> &k_sqr_distances, unsigned int max_nn = 0) const;
207 
208  /** \brief Search for all neighbors of query point that are within a given radius.
209  * \param[in] index index representing the query point in the dataset given by \a setInputCloud.
210  * If indices were given in setInputCloud, index will be the position in the indices vector
211  * \param[in] radius radius of the sphere bounding all of p_q's neighbors
212  * \param[out] k_indices the resultant indices of the neighboring points
213  * \param[out] k_sqr_distances the resultant squared distances to the neighboring points
214  * \param[in] max_nn if given, bounds the maximum returned neighbors to this value
215  * \return number of neighbors found in radius
216  */
217  int
218  radiusSearch (int index, const double radius, std::vector<int> &k_indices,
219  std::vector<float> &k_sqr_distances, unsigned int max_nn = 0) const;
220 
221  /** \brief Get a PointT vector of centers of all voxels that intersected by a ray (origin, direction).
222  * \param[in] origin ray origin
223  * \param[in] direction ray direction vector
224  * \param[out] voxel_center_list results are written to this vector of PointT elements
225  * \param[in] max_voxel_count stop raycasting when this many voxels intersected (0: disable)
226  * \return number of intersected voxels
227  */
228  int
229  getIntersectedVoxelCenters (Eigen::Vector3f origin, Eigen::Vector3f direction,
230  AlignedPointTVector &voxel_center_list, int max_voxel_count = 0) const;
231 
232  /** \brief Get indices of all voxels that are intersected by a ray (origin, direction).
233  * \param[in] origin ray origin
234  * \param[in] direction ray direction vector
235  * \param[out] k_indices resulting point indices from intersected voxels
236  * \param[in] max_voxel_count stop raycasting when this many voxels intersected (0: disable)
237  * \return number of intersected voxels
238  */
239  int
240  getIntersectedVoxelIndices (Eigen::Vector3f origin, Eigen::Vector3f direction,
241  std::vector<int> &k_indices,
242  int max_voxel_count = 0) const;
243 
244 
245  /** \brief Search for points within rectangular search area
246  * \param[in] min_pt lower corner of search area
247  * \param[in] max_pt upper corner of search area
248  * \param[out] k_indices the resultant point indices
249  * \return number of points found within search area
250  */
251  int
252  boxSearch (const Eigen::Vector3f &min_pt, const Eigen::Vector3f &max_pt, std::vector<int> &k_indices) const;
253 
254  protected:
255  //////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
256  // Octree-based search routines & helpers
257  //////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
258  /** \brief @b Priority queue entry for branch nodes
259  * \note This class defines priority queue entries for the nearest neighbor search.
260  * \author Julius Kammerl (julius@kammerl.de)
261  */
263  {
264  public:
265  /** \brief Empty constructor */
267  node (), point_distance (0), key ()
268  {
269  }
270 
271  /** \brief Constructor for initializing priority queue entry.
272  * \param _node pointer to octree node
273  * \param _key octree key addressing voxel in octree structure
274  * \param[in] _point_distance distance of query point to voxel center
275  */
276  prioBranchQueueEntry (OctreeNode* _node, OctreeKey& _key, float _point_distance) :
277  node (_node), point_distance (_point_distance), key (_key)
278  {
279  }
280 
281  /** \brief Operator< for comparing priority queue entries with each other.
282  * \param[in] rhs the priority queue to compare this against
283  */
284  bool
286  {
287  return (this->point_distance > rhs.point_distance);
288  }
289 
290  /** \brief Pointer to octree node. */
291  const OctreeNode* node;
292 
293  /** \brief Distance to query point. */
295 
296  /** \brief Octree key. */
298  };
299 
300  //////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
301  /** \brief @b Priority queue entry for point candidates
302  * \note This class defines priority queue entries for the nearest neighbor point candidates.
303  * \author Julius Kammerl (julius@kammerl.de)
304  */
306  {
307  public:
308 
309  /** \brief Empty constructor */
311  point_idx_ (0), point_distance_ (0)
312  {
313  }
314 
315  /** \brief Constructor for initializing priority queue entry.
316  * \param[in] point_idx an index representing a point in the dataset given by \a setInputCloud
317  * \param[in] point_distance distance of query point to voxel center
318  */
319  prioPointQueueEntry (unsigned int& point_idx, float point_distance) :
320  point_idx_ (point_idx), point_distance_ (point_distance)
321  {
322  }
323 
324  /** \brief Operator< for comparing priority queue entries with each other.
325  * \param[in] rhs priority queue to compare this against
326  */
327  bool
328  operator< (const prioPointQueueEntry& rhs) const
329  {
330  return (this->point_distance_ < rhs.point_distance_);
331  }
332 
333  /** \brief Index representing a point in the dataset given by \a setInputCloud. */
335 
336  /** \brief Distance to query point. */
338  };
339 
340  /** \brief Helper function to calculate the squared distance between two points
341  * \param[in] point_a point A
342  * \param[in] point_b point B
343  * \return squared distance between point A and point B
344  */
345  float
346  pointSquaredDist (const PointT& point_a, const PointT& point_b) const;
347 
348  //////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
349  // Recursive search routine methods
350  //////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
351 
352  /** \brief Recursive search method that explores the octree and finds neighbors within a given radius
353  * \param[in] point query point
354  * \param[in] radiusSquared squared search radius
355  * \param[in] node current octree node to be explored
356  * \param[in] key octree key addressing a leaf node.
357  * \param[in] tree_depth current depth/level in the octree
358  * \param[out] k_indices vector of indices found to be neighbors of query point
359  * \param[out] k_sqr_distances squared distances of neighbors to query point
360  * \param[in] max_nn maximum of neighbors to be found
361  */
362  void
363  getNeighborsWithinRadiusRecursive (const PointT& point, const double radiusSquared,
364  const BranchNode* node, const OctreeKey& key,
365  unsigned int tree_depth, std::vector<int>& k_indices,
366  std::vector<float>& k_sqr_distances, unsigned int max_nn) const;
367 
368  /** \brief Recursive search method that explores the octree and finds the K nearest neighbors
369  * \param[in] point query point
370  * \param[in] K amount of nearest neighbors to be found
371  * \param[in] node current octree node to be explored
372  * \param[in] key octree key addressing a leaf node.
373  * \param[in] tree_depth current depth/level in the octree
374  * \param[in] squared_search_radius squared search radius distance
375  * \param[out] point_candidates priority queue of nearest neigbor point candidates
376  * \return squared search radius based on current point candidate set found
377  */
378  double
379  getKNearestNeighborRecursive (const PointT& point, unsigned int K, const BranchNode* node,
380  const OctreeKey& key, unsigned int tree_depth,
381  const double squared_search_radius,
382  std::vector<prioPointQueueEntry>& point_candidates) const;
383 
384  /** \brief Recursive search method that explores the octree and finds the approximate nearest neighbor
385  * \param[in] point query point
386  * \param[in] node current octree node to be explored
387  * \param[in] key octree key addressing a leaf node.
388  * \param[in] tree_depth current depth/level in the octree
389  * \param[out] result_index result index is written to this reference
390  * \param[out] sqr_distance squared distance to search
391  */
392  void
393  approxNearestSearchRecursive (const PointT& point, const BranchNode* node, const OctreeKey& key,
394  unsigned int tree_depth, int& result_index, float& sqr_distance);
395 
396  /** \brief Recursively search the tree for all intersected leaf nodes and return a vector of voxel centers.
397  * This algorithm is based off the paper An Efficient Parametric Algorithm for Octree Traversal:
398  * http://wscg.zcu.cz/wscg2000/Papers_2000/X31.pdf
399  * \param[in] min_x octree nodes X coordinate of lower bounding box corner
400  * \param[in] min_y octree nodes Y coordinate of lower bounding box corner
401  * \param[in] min_z octree nodes Z coordinate of lower bounding box corner
402  * \param[in] max_x octree nodes X coordinate of upper bounding box corner
403  * \param[in] max_y octree nodes Y coordinate of upper bounding box corner
404  * \param[in] max_z octree nodes Z coordinate of upper bounding box corner
405  * \param[in] a
406  * \param[in] node current octree node to be explored
407  * \param[in] key octree key addressing a leaf node.
408  * \param[out] voxel_center_list results are written to this vector of PointT elements
409  * \param[in] max_voxel_count stop raycasting when this many voxels intersected (0: disable)
410  * \return number of voxels found
411  */
412  int
413  getIntersectedVoxelCentersRecursive (double min_x, double min_y, double min_z, double max_x, double max_y,
414  double max_z, unsigned char a, const OctreeNode* node,
415  const OctreeKey& key, AlignedPointTVector &voxel_center_list,
416  int max_voxel_count) const;
417 
418 
419  /** \brief Recursive search method that explores the octree and finds points within a rectangular search area
420  * \param[in] min_pt lower corner of search area
421  * \param[in] max_pt upper corner of search area
422  * \param[in] node current octree node to be explored
423  * \param[in] key octree key addressing a leaf node.
424  * \param[in] tree_depth current depth/level in the octree
425  * \param[out] k_indices the resultant point indices
426  */
427  void
428  boxSearchRecursive (const Eigen::Vector3f &min_pt, const Eigen::Vector3f &max_pt, const BranchNode* node,
429  const OctreeKey& key, unsigned int tree_depth, std::vector<int>& k_indices) const;
430 
431  /** \brief Recursively search the tree for all intersected leaf nodes and return a vector of indices.
432  * This algorithm is based off the paper An Efficient Parametric Algorithm for Octree Traversal:
433  * http://wscg.zcu.cz/wscg2000/Papers_2000/X31.pdf
434  * \param[in] min_x octree nodes X coordinate of lower bounding box corner
435  * \param[in] min_y octree nodes Y coordinate of lower bounding box corner
436  * \param[in] min_z octree nodes Z coordinate of lower bounding box corner
437  * \param[in] max_x octree nodes X coordinate of upper bounding box corner
438  * \param[in] max_y octree nodes Y coordinate of upper bounding box corner
439  * \param[in] max_z octree nodes Z coordinate of upper bounding box corner
440  * \param[in] a
441  * \param[in] node current octree node to be explored
442  * \param[in] key octree key addressing a leaf node.
443  * \param[out] k_indices resulting indices
444  * \param[in] max_voxel_count stop raycasting when this many voxels intersected (0: disable)
445  * \return number of voxels found
446  */
447  int
448  getIntersectedVoxelIndicesRecursive (double min_x, double min_y, double min_z,
449  double max_x, double max_y, double max_z,
450  unsigned char a, const OctreeNode* node, const OctreeKey& key,
451  std::vector<int> &k_indices,
452  int max_voxel_count) const;
453 
454  /** \brief Initialize raytracing algorithm
455  * \param origin
456  * \param direction
457  * \param[in] min_x octree nodes X coordinate of lower bounding box corner
458  * \param[in] min_y octree nodes Y coordinate of lower bounding box corner
459  * \param[in] min_z octree nodes Z coordinate of lower bounding box corner
460  * \param[in] max_x octree nodes X coordinate of upper bounding box corner
461  * \param[in] max_y octree nodes Y coordinate of upper bounding box corner
462  * \param[in] max_z octree nodes Z coordinate of upper bounding box corner
463  * \param a
464  */
465  inline void
466  initIntersectedVoxel (Eigen::Vector3f &origin, Eigen::Vector3f &direction,
467  double &min_x, double &min_y, double &min_z,
468  double &max_x, double &max_y, double &max_z,
469  unsigned char &a) const
470  {
471  // Account for division by zero when direction vector is 0.0
472  const float epsilon = 1e-10f;
473  if (direction.x () == 0.0)
474  direction.x () = epsilon;
475  if (direction.y () == 0.0)
476  direction.y () = epsilon;
477  if (direction.z () == 0.0)
478  direction.z () = epsilon;
479 
480  // Voxel childIdx remapping
481  a = 0;
482 
483  // Handle negative axis direction vector
484  if (direction.x () < 0.0)
485  {
486  origin.x () = static_cast<float> (this->min_x_) + static_cast<float> (this->max_x_) - origin.x ();
487  direction.x () = -direction.x ();
488  a |= 4;
489  }
490  if (direction.y () < 0.0)
491  {
492  origin.y () = static_cast<float> (this->min_y_) + static_cast<float> (this->max_y_) - origin.y ();
493  direction.y () = -direction.y ();
494  a |= 2;
495  }
496  if (direction.z () < 0.0)
497  {
498  origin.z () = static_cast<float> (this->min_z_) + static_cast<float> (this->max_z_) - origin.z ();
499  direction.z () = -direction.z ();
500  a |= 1;
501  }
502  min_x = (this->min_x_ - origin.x ()) / direction.x ();
503  max_x = (this->max_x_ - origin.x ()) / direction.x ();
504  min_y = (this->min_y_ - origin.y ()) / direction.y ();
505  max_y = (this->max_y_ - origin.y ()) / direction.y ();
506  min_z = (this->min_z_ - origin.z ()) / direction.z ();
507  max_z = (this->max_z_ - origin.z ()) / direction.z ();
508  }
509 
510  /** \brief Find first child node ray will enter
511  * \param[in] min_x octree nodes X coordinate of lower bounding box corner
512  * \param[in] min_y octree nodes Y coordinate of lower bounding box corner
513  * \param[in] min_z octree nodes Z coordinate of lower bounding box corner
514  * \param[in] mid_x octree nodes X coordinate of bounding box mid line
515  * \param[in] mid_y octree nodes Y coordinate of bounding box mid line
516  * \param[in] mid_z octree nodes Z coordinate of bounding box mid line
517  * \return the first child node ray will enter
518  */
519  inline int
520  getFirstIntersectedNode (double min_x, double min_y, double min_z, double mid_x, double mid_y, double mid_z) const
521  {
522  int currNode = 0;
523 
524  if (min_x > min_y)
525  {
526  if (min_x > min_z)
527  {
528  // max(min_x, min_y, min_z) is min_x. Entry plane is YZ.
529  if (mid_y < min_x)
530  currNode |= 2;
531  if (mid_z < min_x)
532  currNode |= 1;
533  }
534  else
535  {
536  // max(min_x, min_y, min_z) is min_z. Entry plane is XY.
537  if (mid_x < min_z)
538  currNode |= 4;
539  if (mid_y < min_z)
540  currNode |= 2;
541  }
542  }
543  else
544  {
545  if (min_y > min_z)
546  {
547  // max(min_x, min_y, min_z) is min_y. Entry plane is XZ.
548  if (mid_x < min_y)
549  currNode |= 4;
550  if (mid_z < min_y)
551  currNode |= 1;
552  }
553  else
554  {
555  // max(min_x, min_y, min_z) is min_z. Entry plane is XY.
556  if (mid_x < min_z)
557  currNode |= 4;
558  if (mid_y < min_z)
559  currNode |= 2;
560  }
561  }
562 
563  return currNode;
564  }
565 
566  /** \brief Get the next visited node given the current node upper
567  * bounding box corner. This function accepts three float values, and
568  * three int values. The function returns the ith integer where the
569  * ith float value is the minimum of the three float values.
570  * \param[in] x current nodes X coordinate of upper bounding box corner
571  * \param[in] y current nodes Y coordinate of upper bounding box corner
572  * \param[in] z current nodes Z coordinate of upper bounding box corner
573  * \param[in] a next node if exit Plane YZ
574  * \param[in] b next node if exit Plane XZ
575  * \param[in] c next node if exit Plane XY
576  * \return the next child node ray will enter or 8 if exiting
577  */
578  inline int
579  getNextIntersectedNode (double x, double y, double z, int a, int b, int c) const
580  {
581  if (x < y)
582  {
583  if (x < z)
584  return a;
585  else
586  return c;
587  }
588  else
589  {
590  if (y < z)
591  return b;
592  else
593  return c;
594  }
595 
596  return 0;
597  }
598 
599  };
600  }
601 }
602 
603 #define PCL_INSTANTIATE_OctreePointCloudSearch(T) template class PCL_EXPORTS pcl::octree::OctreePointCloudSearch<T>;
604 
605 #endif // PCL_OCTREE_SEARCH_H_