Point Cloud Library (PCL) 1.12.0
Loading...
Searching...
No Matches
kdtree_flann.hpp
1/*
2 * Software License Agreement (BSD License)
3 *
4 * Point Cloud Library (PCL) - www.pointclouds.org
5 * Copyright (c) 2009-2011, Willow Garage, Inc.
6 * Copyright (c) 2012-, Open Perception, Inc.
7 *
8 * All rights reserved.
9 *
10 * Redistribution and use in source and binary forms, with or without
11 * modification, are permitted provided that the following conditions
12 * are met:
13 *
14 * * Redistributions of source code must retain the above copyright
15 * notice, this list of conditions and the following disclaimer.
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials provided
19 * with the distribution.
20 * * Neither the name of the copyright holder(s) nor the names of its
21 * contributors may be used to endorse or promote products derived
22 * from this software without specific prior written permission.
23 *
24 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
25 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
26 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
27 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
28 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
29 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
30 * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
31 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
32 * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
33 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
34 * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
35 * POSSIBILITY OF SUCH DAMAGE.
36 *
37 */
38
39#ifndef PCL_KDTREE_KDTREE_IMPL_FLANN_H_
40#define PCL_KDTREE_KDTREE_IMPL_FLANN_H_
41
42#include <flann/flann.hpp>
43
44#include <pcl/kdtree/kdtree_flann.h>
45#include <pcl/console/print.h>
46
47///////////////////////////////////////////////////////////////////////////////////////////
48template <typename PointT, typename Dist>
50 : pcl::KdTree<PointT> (sorted)
51 , flann_index_ ()
52 , identity_mapping_ (false)
53 , dim_ (0), total_nr_points_ (0)
54 , param_k_ (::flann::SearchParams (-1 , epsilon_))
55 , param_radius_ (::flann::SearchParams (-1, epsilon_, sorted))
56{
57 if (!std::is_same<std::size_t, pcl::index_t>::value) {
58 const auto message = "FLANN is not optimized for current index type. Will incur "
59 "extra allocations and copy\n";
60 if (std::is_same<int, pcl::index_t>::value) {
61 PCL_DEBUG(message); // since this has been the default behavior till PCL 1.12
62 }
63 else {
64 PCL_WARN(message);
65 }
66 }
67}
68
69///////////////////////////////////////////////////////////////////////////////////////////
70template <typename PointT, typename Dist>
72 : pcl::KdTree<PointT> (false)
73 , flann_index_ ()
74 , identity_mapping_ (false)
75 , dim_ (0), total_nr_points_ (0)
76 , param_k_ (::flann::SearchParams (-1 , epsilon_))
77 , param_radius_ (::flann::SearchParams (-1, epsilon_, false))
78{
79 *this = k;
80}
81
82///////////////////////////////////////////////////////////////////////////////////////////
83template <typename PointT, typename Dist> void
85{
86 epsilon_ = eps;
87 param_k_ = ::flann::SearchParams (-1 , epsilon_);
88 param_radius_ = ::flann::SearchParams (-1 , epsilon_, sorted_);
89}
90
91///////////////////////////////////////////////////////////////////////////////////////////
92template <typename PointT, typename Dist> void
94{
95 sorted_ = sorted;
96 param_k_ = ::flann::SearchParams (-1, epsilon_);
97 param_radius_ = ::flann::SearchParams (-1, epsilon_, sorted_);
98}
99
100///////////////////////////////////////////////////////////////////////////////////////////
101template <typename PointT, typename Dist> void
103{
104 cleanup (); // Perform an automatic cleanup of structures
105
106 epsilon_ = 0.0f; // default error bound value
107 dim_ = point_representation_->getNumberOfDimensions (); // Number of dimensions - default is 3 = xyz
108
109 input_ = cloud;
110 indices_ = indices;
111
112 // Allocate enough data
113 if (!input_)
114 {
115 PCL_ERROR ("[pcl::KdTreeFLANN::setInputCloud] Invalid input!\n");
116 return;
117 }
118 if (indices != nullptr)
119 {
120 convertCloudToArray (*input_, *indices_);
121 }
122 else
123 {
124 convertCloudToArray (*input_);
125 }
126 total_nr_points_ = static_cast<uindex_t> (index_mapping_.size ());
127 if (total_nr_points_ == 0)
128 {
129 PCL_ERROR ("[pcl::KdTreeFLANN::setInputCloud] Cannot create a KDTree with an empty input cloud!\n");
130 return;
131 }
132
133 flann_index_.reset (new FLANNIndex (::flann::Matrix<float> (cloud_.get (),
134 index_mapping_.size (),
135 dim_),
136 ::flann::KDTreeSingleIndexParams (15))); // max 15 points/leaf
137 flann_index_->buildIndex ();
138}
139
140///////////////////////////////////////////////////////////////////////////////////////////
141namespace pcl {
142namespace detail {
143// Replace using constexpr in C++17
144template <class IndexT,
145 class A,
146 class B,
147 class C,
148 class D,
149 class F,
150 CompatWithFlann<IndexT> = true>
151int
152knn_search(A& index, B& query, C& k_indices, D& dists, unsigned int k, F& params)
153{
154 // Wrap k_indices vector (no data allocation)
155 ::flann::Matrix<index_t> k_indices_mat(&k_indices[0], 1, k);
156 return index.knnSearch(query, k_indices_mat, dists, k, params);
157}
158
159template <class IndexT,
160 class A,
161 class B,
162 class C,
163 class D,
164 class F,
165 NotCompatWithFlann<IndexT> = true>
166int
167knn_search(A& index, B& query, C& k_indices, D& dists, unsigned int k, F& params)
168{
169 std::vector<std::size_t> indices(k);
170 k_indices.resize(k);
171 // Wrap indices vector (no data allocation)
172 ::flann::Matrix<std::size_t> indices_mat(&indices[0], 1, k);
173 auto ret = index.knnSearch(query, indices_mat, dists, k, params);
174 std::copy(indices.cbegin(), indices.cend(), k_indices.begin());
175 return ret;
176}
177
178template <class IndexT, class A, class B, class F, CompatWithFlann<IndexT> = true>
179int
180knn_search(A& index,
181 B& query,
182 std::vector<Indices>& k_indices,
183 std::vector<std::vector<float>>& dists,
184 unsigned int k,
185 F& params)
186{
187 return index.knnSearch(query, k_indices, dists, k, params);
188}
190template <class IndexT, class A, class B, class F, NotCompatWithFlann<IndexT> = true>
191int
192knn_search(A& index,
193 B& query,
194 std::vector<Indices>& k_indices,
195 std::vector<std::vector<float>>& dists,
196 unsigned int k,
197 F& params)
198{
199 std::vector<std::vector<std::size_t>> indices;
200 // flann will resize accordingly
201 auto ret = index.knnSearch(query, indices, dists, k, params);
202
203 k_indices.resize(indices.size());
204 {
205 auto it = indices.cbegin();
206 auto jt = k_indices.begin();
207 for (; it != indices.cend(); ++it, ++jt) {
208 jt->resize(it->size());
209 std::copy(it->cbegin(), it->cend(), jt->begin());
210 }
212 return ret;
213}
214} // namespace detail
215template <class FlannIndex,
216 class Query,
217 class Indices,
218 class Distances,
219 class SearchParams>
220int
221knn_search(const FlannIndex& index,
222 const Query& query,
223 Indices& indices,
224 Distances& dists,
225 unsigned int k,
226 const SearchParams& params)
227{
228 return detail::knn_search<pcl::index_t>(index, query, indices, dists, k, params);
229}
230} // namespace pcl
231
232template <typename PointT, typename Dist> int
234 Indices &k_indices,
235 std::vector<float> &k_distances) const
236{
237 assert (point_representation_->isValid (point) && "Invalid (NaN, Inf) point coordinates given to nearestKSearch!");
238
239 if (k > total_nr_points_)
240 k = total_nr_points_;
241
242 k_indices.resize (k);
243 k_distances.resize (k);
244
245 if (k==0)
246 return 0;
247
248 std::vector<float> query (dim_);
249 point_representation_->vectorize (static_cast<PointT> (point), query);
250
251 // Wrap the k_distances vector (no data copy)
252 ::flann::Matrix<float> k_distances_mat (&k_distances[0], 1, k);
254 knn_search(*flann_index_,
255 ::flann::Matrix<float>(&query[0], 1, dim_),
256 k_indices,
257 k_distances_mat,
258 k,
259 param_k_);
260
261 // Do mapping to original point cloud
262 if (!identity_mapping_)
263 {
264 for (std::size_t i = 0; i < static_cast<std::size_t> (k); ++i)
265 {
266 auto& neighbor_index = k_indices[i];
267 neighbor_index = index_mapping_[neighbor_index];
268 }
269 }
270
271 return (k);
272}
273
274///////////////////////////////////////////////////////////////////////////////////////////
275namespace pcl {
276namespace detail {
277// Replace using constexpr in C++17
278template <class IndexT,
279 class A,
280 class B,
281 class C,
282 class D,
283 class F,
284 CompatWithFlann<IndexT> = true>
285int
286radius_search(A& index, B& query, C& k_indices, D& dists, float radius, F& params)
287{
288 std::vector<pcl::Indices> indices(1);
289 int neighbors_in_radius = index.radiusSearch(query, indices, dists, radius, params);
290 k_indices = std::move(indices[0]);
291 return neighbors_in_radius;
292}
293
294template <class IndexT,
295 class A,
296 class B,
297 class C,
298 class D,
299 class F,
300 NotCompatWithFlann<IndexT> = true>
301int
302radius_search(A& index, B& query, C& k_indices, D& dists, float radius, F& params)
303{
304 std::vector<std::vector<std::size_t>> indices(1);
305 int neighbors_in_radius = index.radiusSearch(query, indices, dists, radius, params);
306 k_indices.resize(indices[0].size());
307 std::copy(indices[0].cbegin(), indices[0].cend(), k_indices.begin());
308 return neighbors_in_radius;
309}
310
311template <class IndexT, class A, class B, class F, CompatWithFlann<IndexT> = true>
312int
314 B& query,
315 std::vector<Indices>& k_indices,
316 std::vector<std::vector<float>>& dists,
317 float radius,
318 F& params)
319{
320 return index.radiusSearch(query, k_indices, dists, radius, params);
321}
322
323template <class IndexT, class A, class B, class F, NotCompatWithFlann<IndexT> = true>
324int
325radius_search(A& index,
326 B& query,
327 std::vector<Indices>& k_indices,
328 std::vector<std::vector<float>>& dists,
329 float radius,
330 F& params)
331{
332 std::vector<std::vector<std::size_t>> indices;
333 // flann will resize accordingly
334 auto ret = index.radiusSearch(query, indices, dists, radius, params);
335
336 k_indices.resize(indices.size());
337 {
338 auto it = indices.cbegin();
339 auto jt = k_indices.begin();
340 for (; it != indices.cend(); ++it, ++jt) {
341 jt->resize(it->size());
342 std::copy(it->cbegin(), it->cend(), jt->begin());
343 }
344 }
345 return ret;
346}
347} // namespace detail
348template <class FlannIndex,
349 class Query,
350 class Indices,
351 class Distances,
352 class SearchParams>
353int
354radius_search(const FlannIndex& index,
355 const Query& query,
356 Indices& indices,
357 Distances& dists,
358 float radius,
359 const SearchParams& params)
360{
361 return detail::radius_search<pcl::index_t>(
362 index, query, indices, dists, radius, params);
363}
364} // namespace pcl
365
366template <typename PointT, typename Dist> int
367pcl::KdTreeFLANN<PointT, Dist>::radiusSearch (const PointT &point, double radius, Indices &k_indices,
368 std::vector<float> &k_sqr_dists, unsigned int max_nn) const
369{
370 assert (point_representation_->isValid (point) && "Invalid (NaN, Inf) point coordinates given to radiusSearch!");
371
372 std::vector<float> query (dim_);
373 point_representation_->vectorize (static_cast<PointT> (point), query);
374
375 // Has max_nn been set properly?
376 if (max_nn == 0 || max_nn > total_nr_points_)
377 max_nn = total_nr_points_;
378
379 std::vector<std::vector<float> > dists(1);
380
381 ::flann::SearchParams params (param_radius_);
382 if (max_nn == total_nr_points_)
383 params.max_neighbors = -1; // return all neighbors in radius
384 else
385 params.max_neighbors = max_nn;
386
387 auto query_mat = ::flann::Matrix<float>(&query[0], 1, dim_);
388 int neighbors_in_radius = radius_search(*flann_index_,
389 query_mat,
390 k_indices,
391 dists,
392 static_cast<float>(radius * radius),
393 params);
394
395 k_sqr_dists = dists[0];
396
397 // Do mapping to original point cloud
398 if (!identity_mapping_)
399 {
400 for (int i = 0; i < neighbors_in_radius; ++i)
401 {
402 auto& neighbor_index = k_indices[i];
403 neighbor_index = index_mapping_[neighbor_index];
404 }
405 }
406
407 return (neighbors_in_radius);
408}
409
410///////////////////////////////////////////////////////////////////////////////////////////
411template <typename PointT, typename Dist> void
413{
414 // Data array cleanup
415 index_mapping_.clear ();
416
417 if (indices_)
418 indices_.reset ();
419}
420
421///////////////////////////////////////////////////////////////////////////////////////////
422template <typename PointT, typename Dist> void
424{
425 // No point in doing anything if the array is empty
426 if (cloud.empty ())
427 {
428 cloud_.reset ();
429 return;
430 }
431
432 const auto original_no_of_points = cloud.size ();
433
434 cloud_.reset (new float[original_no_of_points * dim_], std::default_delete<float[]> ());
435 float* cloud_ptr = cloud_.get ();
436 index_mapping_.reserve (original_no_of_points);
437 identity_mapping_ = true;
438
439 for (std::size_t cloud_index = 0; cloud_index < original_no_of_points; ++cloud_index)
440 {
441 // Check if the point is invalid
442 if (!point_representation_->isValid (cloud[cloud_index]))
443 {
444 identity_mapping_ = false;
445 continue;
446 }
447
448 index_mapping_.push_back (cloud_index);
449
450 point_representation_->vectorize (cloud[cloud_index], cloud_ptr);
451 cloud_ptr += dim_;
452 }
453}
454
455///////////////////////////////////////////////////////////////////////////////////////////
456template <typename PointT, typename Dist> void
458{
459 // No point in doing anything if the array is empty
460 if (cloud.empty ())
461 {
462 cloud_.reset ();
463 return;
464 }
465
466 int original_no_of_points = static_cast<int> (indices.size ());
467
468 cloud_.reset (new float[original_no_of_points * dim_], std::default_delete<float[]> ());
469 float* cloud_ptr = cloud_.get ();
470 index_mapping_.reserve (original_no_of_points);
471 // its a subcloud -> false
472 // true only identity:
473 // - indices size equals cloud size
474 // - indices only contain values between 0 and cloud.size - 1
475 // - no index is multiple times in the list
476 // => index is complete
477 // But we can not guarantee that => identity_mapping_ = false
478 identity_mapping_ = false;
479
480 for (const auto &index : indices)
481 {
482 // Check if the point is invalid
483 if (!point_representation_->isValid (cloud[index]))
484 continue;
485
486 // map from 0 - N -> indices [0] - indices [N]
487 index_mapping_.push_back (index); // If the returned index should be for the indices vector
488
489 point_representation_->vectorize (cloud[index], cloud_ptr);
490 cloud_ptr += dim_;
491 }
492}
493
494#define PCL_INSTANTIATE_KdTreeFLANN(T) template class PCL_EXPORTS pcl::KdTreeFLANN<T>;
495
496#endif //#ifndef _PCL_KDTREE_KDTREE_IMPL_FLANN_H_
497
KdTreeFLANN is a generic type of 3D spatial locator using kD-tree structures.
void setEpsilon(float eps) override
Set the search epsilon precision (error bound) for nearest neighbors searches.
shared_ptr< const Indices > IndicesConstPtr
int nearestKSearch(const PointT &point, unsigned int k, Indices &k_indices, std::vector< float > &k_sqr_distances) const override
Search for k-nearest neighbors for the given query point.
int radiusSearch(const PointT &point, double radius, Indices &k_indices, std::vector< float > &k_sqr_distances, unsigned int max_nn=0) const override
Search for all the nearest neighbors of the query point in a given radius.
typename KdTree< PointT >::PointCloudConstPtr PointCloudConstPtr
void setSortedResults(bool sorted)
KdTreeFLANN(bool sorted=true)
Default Constructor for KdTreeFLANN.
void setInputCloud(const PointCloudConstPtr &cloud, const IndicesConstPtr &indices=IndicesConstPtr()) override
Provide a pointer to the input dataset.
KdTree represents the base spatial locator class for kd-tree implementations.
Definition kdtree.h:55
PointCloud represents the base class in PCL for storing collections of 3D points.
bool empty() const
std::size_t size() const
@ B
Definition norms.h:54
int knn_search(A &index, B &query, C &k_indices, D &dists, unsigned int k, F &params)
int radius_search(A &index, B &query, C &k_indices, D &dists, float radius, F &params)
detail::int_type_t< detail::index_type_size, false > uindex_t
Type used for an unsigned index in PCL.
Definition types.h:120
IndicesAllocator<> Indices
Type used for indices in PCL.
Definition types.h:133
int radius_search(const FlannIndex &index, const Query &query, Indices &indices, Distances &dists, float radius, const SearchParams &params)
Comaptibility template function to allow use of various types of indices with FLANN.
int knn_search(const FlannIndex &index, const Query &query, Indices &indices, Distances &dists, unsigned int k, const SearchParams &params)
Comaptibility template function to allow use of various types of indices with FLANN.
A point structure representing Euclidean xyz coordinates, and the RGB color.