Point Cloud Library (PCL) 1.15.0
Loading...
Searching...
No Matches
flare.hpp
1/*
2* Software License Agreement (BSD License)
3*
4* Point Cloud Library (PCL) - www.pointclouds.org
5* Copyright (c) 2016-, Open Perception, 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 the copyright holder(s) 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*
37*/
38
39#ifndef PCL_FEATURES_IMPL_FLARE_H_
40#define PCL_FEATURES_IMPL_FLARE_H_
41
42#include <pcl/features/flare.h>
43#include <pcl/common/geometry.h>
44#include <pcl/search/kdtree.h> // for KdTree
45#include <pcl/search/organized.h> // for OrganizedNeighbor
46
47//////////////////////////////////////////////////////////////////////////////////////////////
48template<typename PointInT, typename PointNT, typename PointOutT, typename SignedDistanceT> bool
50{
52 {
53 PCL_ERROR ("[pcl::%s::initCompute] Init failed.\n", getClassName ().c_str ());
54 return (false);
55 }
56
57 if (tangent_radius_ == 0.0f)
58 {
59 PCL_ERROR ("[pcl::%s::initCompute] tangent_radius_ not set.\n", getClassName ().c_str ());
60 return (false);
61 }
62
63 // If no search sampled_surface_ has been defined, use the surface_ dataset as the search sampled_surface_ itself
64 if (!sampled_surface_)
65 {
66 fake_sampled_surface_ = true;
67 sampled_surface_ = surface_;
68
69 if (sampled_tree_)
70 {
71 PCL_WARN ("[pcl::%s::initCompute] sampled_surface_ is not set even if sampled_tree_ is already set.", getClassName ().c_str ());
72 PCL_WARN ("sampled_tree_ will be rebuilt from surface_. Use sampled_surface_.\n");
73 }
74 }
75
76 // Check if a space search locator was given for sampled_surface_
77 if (!sampled_tree_)
78 {
79 if (sampled_surface_->isOrganized () && surface_->isOrganized () && input_->isOrganized ())
80 sampled_tree_.reset (new pcl::search::OrganizedNeighbor<PointInT> ());
81 else
82 sampled_tree_.reset (new pcl::search::KdTree<PointInT> (false));
83 }
84
85 if (sampled_tree_->getInputCloud () != sampled_surface_) // Make sure the tree searches the sampled surface
86 sampled_tree_->setInputCloud (sampled_surface_);
87
88 return (true);
89}
90
91//////////////////////////////////////////////////////////////////////////////////////////////
92template<typename PointInT, typename PointNT, typename PointOutT, typename SignedDistanceT> bool
94{
95 // Reset the surface
96 if (fake_surface_)
97 {
98 surface_.reset ();
99 fake_surface_ = false;
100 }
101 // Reset the sampled surface
102 if (fake_sampled_surface_)
103 {
104 sampled_surface_.reset ();
105 fake_sampled_surface_ = false;
106 }
107 return (true);
108}
109
110//////////////////////////////////////////////////////////////////////////////////////////////
111template<typename PointInT, typename PointNT, typename PointOutT, typename SignedDistanceT> SignedDistanceT
113 Eigen::Matrix3f &lrf)
114{
115 Eigen::Vector3f x_axis, y_axis;
116 Eigen::Vector3f fitted_normal; //z_axis
117
118 //find Z axis
119
120 //extract support points for the computation of Z axis
121 pcl::Indices neighbours_indices;
122 std::vector<float> neighbours_distances;
123
124 const std::size_t n_normal_neighbours =
125 this->searchForNeighbors (index, search_parameter_, neighbours_indices, neighbours_distances);
126 if (n_normal_neighbours < static_cast<std::size_t>(min_neighbors_for_normal_axis_))
127 {
128 if (!pcl::isFinite ((*normals_)[index]))
129 {
130 //normal is invalid
131 //setting lrf to NaN
132 lrf.setConstant (std::numeric_limits<float>::quiet_NaN ());
133 return (std::numeric_limits<SignedDistanceT>::max ());
134 }
135
136 //set z_axis as the normal of index point
137 fitted_normal = (*normals_)[index].getNormalVector3fMap ();
138 }
139 else
140 {
141 float plane_curvature;
142 normal_estimation_.computePointNormal (*surface_, neighbours_indices, fitted_normal (0), fitted_normal (1), fitted_normal (2), plane_curvature);
143
144 //disambiguate Z axis with normal mean
145 if (!pcl::flipNormalTowardsNormalsMean<PointNT> (*normals_, neighbours_indices, fitted_normal))
146 {
147 //all normals in the neighbourhood are invalid
148 //setting lrf to NaN
149 lrf.setConstant (std::numeric_limits<float>::quiet_NaN ());
150 return (std::numeric_limits<SignedDistanceT>::max ());
151 }
152 }
153
154 //setting LRF Z axis
155 lrf.row (2).matrix () = fitted_normal;
156
157 //find X axis
158
159 //extract support points for Rx radius
160 const std::size_t n_tangent_neighbours =
161 sampled_tree_->radiusSearch ((*input_)[index], tangent_radius_, neighbours_indices, neighbours_distances);
162
163 if (n_tangent_neighbours < static_cast<std::size_t>(min_neighbors_for_tangent_axis_))
164 {
165 //set X axis as a random axis
166 x_axis = pcl::geometry::randomOrthogonalAxis (fitted_normal);
167 y_axis = fitted_normal.cross (x_axis);
168
169 lrf.row (0).matrix () = x_axis;
170 lrf.row (1).matrix () = y_axis;
171
172 return (std::numeric_limits<SignedDistanceT>::max ());
173 }
174
175 //find point with the largest signed distance from tangent plane
176
177 SignedDistanceT shape_score;
178 SignedDistanceT best_shape_score = -std::numeric_limits<SignedDistanceT>::max ();
179 int best_shape_index = -1;
180
181 Eigen::Vector3f best_margin_point;
182
183 const float radius2 = tangent_radius_ * tangent_radius_;
184 const float margin_distance2 = margin_thresh_ * margin_thresh_ * radius2;
185
186 Vector3fMapConst feature_point = (*input_)[index].getVector3fMap ();
187
188 for (std::size_t curr_neigh = 0; curr_neigh < n_tangent_neighbours; ++curr_neigh)
189 {
190 const int& curr_neigh_idx = neighbours_indices[curr_neigh];
191 const float& neigh_distance_sqr = neighbours_distances[curr_neigh];
192
193 if (neigh_distance_sqr <= margin_distance2)
194 {
195 continue;
196 }
197
198 //point curr_neigh_idx is inside the ring between marginThresh and Radius
199
200 shape_score = fitted_normal.dot ((*sampled_surface_)[curr_neigh_idx].getVector3fMap ());
201
202 if (shape_score > best_shape_score)
203 {
204 best_shape_index = curr_neigh_idx;
205 best_shape_score = shape_score;
206 }
207 } //for each neighbor
208
209 if (best_shape_index == -1)
210 {
211 x_axis = pcl::geometry::randomOrthogonalAxis (fitted_normal);
212 y_axis = fitted_normal.cross (x_axis);
213
214 lrf.row (0).matrix () = x_axis;
215 lrf.row (1).matrix () = y_axis;
216
217 return (std::numeric_limits<SignedDistanceT>::max ());
218 }
219
220 //find orthogonal axis directed to best_shape_index point projection on plane with fittedNormal as axis
221 x_axis = pcl::geometry::projectedAsUnitVector (sampled_surface_->at (best_shape_index).getVector3fMap (), feature_point, fitted_normal);
222
223 y_axis = fitted_normal.cross (x_axis);
224
225 lrf.row (0).matrix () = x_axis;
226 lrf.row (1).matrix () = y_axis;
227 //z axis already set
228
229 best_shape_score -= fitted_normal.dot (feature_point);
230 return (best_shape_score);
231}
232
233//////////////////////////////////////////////////////////////////////////////////////////////
234template<typename PointInT, typename PointNT, typename PointOutT, typename SignedDistanceT> void
236{
237 //check whether used with search radius or search k-neighbors
238 if (this->getKSearch () != 0)
239 {
240 PCL_ERROR (
241 "[pcl::%s::computeFeature] Error! Search method set to k-neighborhood. Call setKSearch (0) and setRadiusSearch (radius) to use this class.\n",
242 getClassName ().c_str ());
243 return;
244 }
245
246 signed_distances_from_highest_points_.resize (indices_->size ());
247
248 for (std::size_t point_idx = 0; point_idx < indices_->size (); ++point_idx)
249 {
250 Eigen::Matrix3f currentLrf;
251 PointOutT &rf = output[point_idx];
252
253 signed_distances_from_highest_points_[point_idx] = computePointLRF ((*indices_)[point_idx], currentLrf);
254 if (signed_distances_from_highest_points_[point_idx] == std::numeric_limits<SignedDistanceT>::max ())
255 {
256 output.is_dense = false;
257 }
258
259 rf.getXAxisVector3fMap () = currentLrf.row (0);
260 rf.getYAxisVector3fMap () = currentLrf.row (1);
261 rf.getZAxisVector3fMap () = currentLrf.row (2);
262 }
263}
264
265#define PCL_INSTANTIATE_FLARELocalReferenceFrameEstimation(T,NT,OutT,SdT) template class PCL_EXPORTS pcl::FLARELocalReferenceFrameEstimation<T,NT,OutT,SdT>;
266
267#endif // PCL_FEATURES_IMPL_FLARE_H_
bool deinitCompute() override
This method should get called after the actual computation is ended.
Definition flare.hpp:93
void computeFeature(PointCloudOut &output) override
Abstract feature estimation method.
Definition flare.hpp:235
bool initCompute() override
This method should get called before starting the actual computation.
Definition flare.hpp:49
SignedDistanceT computePointLRF(const int index, Eigen::Matrix3f &lrf)
Estimate the LRF descriptor for a given point based on its spatial neighborhood of 3D points with nor...
Definition flare.hpp:112
search::KdTree is a wrapper class which inherits the pcl::KdTree class for performing search function...
Definition kdtree.h:62
OrganizedNeighbor is a class for optimized nearest neighbor search in organized projectable point clo...
Definition organized.h:66
Defines some geometrical functions and utility functions.
Eigen::Vector3f projectedAsUnitVector(Eigen::Vector3f const &point, Eigen::Vector3f const &plane_origin, Eigen::Vector3f const &plane_normal)
Given a plane defined by plane_origin and plane_normal, find the unit vector pointing from plane_orig...
Definition geometry.h:115
Eigen::Vector3f randomOrthogonalAxis(Eigen::Vector3f const &axis)
Define a random unit vector orthogonal to axis.
Definition geometry.h:134
bool isFinite(const PointT &pt)
Tests if the 3D components of a point are all finite param[in] pt point to be tested return true if f...
Definition point_tests.h:55
const Eigen::Map< const Eigen::Vector3f > Vector3fMapConst
IndicesAllocator<> Indices
Type used for indices in PCL.
Definition types.h:133