Point Cloud Library (PCL)  1.10.1
lmeds.hpp
1 /*
2  * Software License Agreement (BSD License)
3  *
4  * Point Cloud Library (PCL) - www.pointclouds.org
5  * Copyright (c) 2009, 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  * $Id$
38  *
39  */
40 
41 #ifndef PCL_SAMPLE_CONSENSUS_IMPL_LMEDS_H_
42 #define PCL_SAMPLE_CONSENSUS_IMPL_LMEDS_H_
43 
44 #include <pcl/sample_consensus/lmeds.h>
45 
46 //////////////////////////////////////////////////////////////////////////
47 template <typename PointT> bool
49 {
50  // Warn and exit if no threshold was set
51  if (threshold_ == std::numeric_limits<double>::max())
52  {
53  PCL_ERROR ("[pcl::LeastMedianSquares::computeModel] No threshold set!\n");
54  return (false);
55  }
56 
57  iterations_ = 0;
58  double d_best_penalty = std::numeric_limits<double>::max();
59 
60  std::vector<int> selection;
61  Eigen::VectorXf model_coefficients;
62  std::vector<double> distances;
63 
64  int n_inliers_count = 0;
65 
66  unsigned skipped_count = 0;
67  // suppress infinite loops by just allowing 10 x maximum allowed iterations for invalid model parameters!
68  const unsigned max_skip = max_iterations_ * 10;
69 
70  // Iterate
71  while (iterations_ < max_iterations_ && skipped_count < max_skip)
72  {
73  // Get X samples which satisfy the model criteria
74  sac_model_->getSamples (iterations_, selection);
75 
76  if (selection.empty ()) break;
77 
78  // Search for inliers in the point cloud for the current plane model M
79  if (!sac_model_->computeModelCoefficients (selection, model_coefficients))
80  {
81  //iterations_++;
82  ++skipped_count;
83  continue;
84  }
85 
86  double d_cur_penalty = 0;
87  // d_cur_penalty = sum (min (dist, threshold))
88 
89  // Iterate through the 3d points and calculate the distances from them to the model
90  sac_model_->getDistancesToModel (model_coefficients, distances);
91 
92  // No distances? The model must not respect the user given constraints
93  if (distances.empty ())
94  {
95  //iterations_++;
96  ++skipped_count;
97  continue;
98  }
99 
100  std::sort (distances.begin (), distances.end ());
101  // d_cur_penalty = median (distances)
102  std::size_t mid = sac_model_->getIndices ()->size () / 2;
103  if (mid >= distances.size ())
104  {
105  //iterations_++;
106  ++skipped_count;
107  continue;
108  }
109 
110  // Do we have a "middle" point or should we "estimate" one ?
111  if (sac_model_->getIndices ()->size () % 2 == 0)
112  d_cur_penalty = (sqrt (distances[mid-1]) + sqrt (distances[mid])) / 2;
113  else
114  d_cur_penalty = sqrt (distances[mid]);
115 
116  // Better match ?
117  if (d_cur_penalty < d_best_penalty)
118  {
119  d_best_penalty = d_cur_penalty;
120 
121  // Save the current model/coefficients selection as being the best so far
122  model_ = selection;
123  model_coefficients_ = model_coefficients;
124  }
125 
126  ++iterations_;
127  if (debug_verbosity_level > 1)
128  PCL_DEBUG ("[pcl::LeastMedianSquares::computeModel] Trial %d out of %d. Best penalty is %f.\n", iterations_, max_iterations_, d_best_penalty);
129  }
130 
131  if (model_.empty ())
132  {
133  if (debug_verbosity_level > 0)
134  PCL_DEBUG ("[pcl::LeastMedianSquares::computeModel] Unable to find a solution!\n");
135  return (false);
136  }
137 
138  // Classify the data points into inliers and outliers
139  // Sigma = 1.4826 * (1 + 5 / (n-d)) * sqrt (M)
140  // @note: See "Robust Regression Methods for Computer Vision: A Review"
141  //double sigma = 1.4826 * (1 + 5 / (sac_model_->getIndices ()->size () - best_model.size ())) * sqrt (d_best_penalty);
142  //double threshold = 2.5 * sigma;
143 
144  // Iterate through the 3d points and calculate the distances from them to the model again
145  sac_model_->getDistancesToModel (model_coefficients_, distances);
146  // No distances? The model must not respect the user given constraints
147  if (distances.empty ())
148  {
149  PCL_ERROR ("[pcl::LeastMedianSquares::computeModel] The model found failed to verify against the given constraints!\n");
150  return (false);
151  }
152 
153  std::vector<int> &indices = *sac_model_->getIndices ();
154 
155  if (distances.size () != indices.size ())
156  {
157  PCL_ERROR ("[pcl::LeastMedianSquares::computeModel] Estimated distances (%lu) differs than the normal of indices (%lu).\n", distances.size (), indices.size ());
158  return (false);
159  }
160 
161  inliers_.resize (distances.size ());
162  // Get the inliers for the best model found
163  n_inliers_count = 0;
164  for (std::size_t i = 0; i < distances.size (); ++i)
165  if (distances[i] <= threshold_)
166  inliers_[n_inliers_count++] = indices[i];
167 
168  // Resize the inliers vector
169  inliers_.resize (n_inliers_count);
170 
171  if (debug_verbosity_level > 0)
172  PCL_DEBUG ("[pcl::LeastMedianSquares::computeModel] Model: %lu size, %d inliers.\n", model_.size (), n_inliers_count);
173 
174  return (true);
175 }
176 
177 #define PCL_INSTANTIATE_LeastMedianSquares(T) template class PCL_EXPORTS pcl::LeastMedianSquares<T>;
178 
179 #endif // PCL_SAMPLE_CONSENSUS_IMPL_LMEDS_H_
180 
pcl::LeastMedianSquares::computeModel
bool computeModel(int debug_verbosity_level=0) override
Compute the actual model and find the inliers.
Definition: lmeds.hpp:48