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_RMSAC_H_
42 #define PCL_SAMPLE_CONSENSUS_IMPL_RMSAC_H_
43
44 #include <pcl/sample_consensus/rmsac.h>
45
46 //////////////////////////////////////////////////////////////////////////
47 template <typename PointT> bool
computeModel(int debug_verbosity_level)48 pcl::RandomizedMEstimatorSampleConsensus<PointT>::computeModel (int debug_verbosity_level)
49 {
50 // Warn and exit if no threshold was set
51 if (threshold_ == std::numeric_limits<double>::max())
52 {
53 PCL_ERROR ("[pcl::RandomizedMEstimatorSampleConsensus::computeModel] No threshold set!\n");
54 return (false);
55 }
56
57 iterations_ = 0;
58 double d_best_penalty = std::numeric_limits<double>::max();
59 double k = 1.0;
60
61 Indices selection;
62 Eigen::VectorXf model_coefficients (sac_model_->getModelSize ());
63 std::vector<double> distances;
64 std::set<index_t> indices_subset;
65
66 int n_inliers_count = 0;
67 unsigned skipped_count = 0;
68 // suppress infinite loops by just allowing 10 x maximum allowed iterations for invalid model parameters!
69 const unsigned max_skip = max_iterations_ * 10;
70
71 // Number of samples to try randomly
72 std::size_t fraction_nr_points = pcl_lrint (static_cast<double>(sac_model_->getIndices ()->size ()) * fraction_nr_pretest_ / 100.0);
73
74 // Iterate
75 while (iterations_ < k && skipped_count < max_skip)
76 {
77 // Get X samples which satisfy the model criteria
78 sac_model_->getSamples (iterations_, selection);
79
80 if (selection.empty ()) break;
81
82 // Search for inliers in the point cloud for the current plane model M
83 if (!sac_model_->computeModelCoefficients (selection, model_coefficients))
84 {
85 //iterations_++;
86 ++ skipped_count;
87 continue;
88 }
89
90 // RMSAC addon: verify a random fraction of the data
91 // Get X random samples which satisfy the model criterion
92 this->getRandomSamples (sac_model_->getIndices (), fraction_nr_points, indices_subset);
93
94 if (!sac_model_->doSamplesVerifyModel (indices_subset, model_coefficients, threshold_))
95 {
96 // Unfortunately we cannot "continue" after the first iteration, because k might not be set, while iterations gets incremented
97 if (k != 1.0)
98 {
99 ++iterations_;
100 continue;
101 }
102 }
103
104 double d_cur_penalty = 0;
105 // Iterate through the 3d points and calculate the distances from them to the model
106 sac_model_->getDistancesToModel (model_coefficients, distances);
107
108 if (distances.empty () && k > 1.0)
109 continue;
110
111 for (const double &distance : distances)
112 d_cur_penalty += std::min (distance, threshold_);
113
114 // Better match ?
115 if (d_cur_penalty < d_best_penalty)
116 {
117 d_best_penalty = d_cur_penalty;
118
119 // Save the current model/coefficients selection as being the best so far
120 model_ = selection;
121 model_coefficients_ = model_coefficients;
122
123 n_inliers_count = 0;
124 // Need to compute the number of inliers for this model to adapt k
125 for (const double &distance : distances)
126 if (distance <= threshold_)
127 n_inliers_count++;
128
129 // Compute the k parameter (k=std::log(z)/std::log(1-w^n))
130 double w = static_cast<double> (n_inliers_count) / static_cast<double>(sac_model_->getIndices ()->size ());
131 double p_no_outliers = 1 - std::pow (w, static_cast<double> (selection.size ()));
132 p_no_outliers = (std::max) (std::numeric_limits<double>::epsilon (), p_no_outliers); // Avoid division by -Inf
133 p_no_outliers = (std::min) (1 - std::numeric_limits<double>::epsilon (), p_no_outliers); // Avoid division by 0.
134 k = std::log (1 - probability_) / std::log (p_no_outliers);
135 }
136
137 ++iterations_;
138 if (debug_verbosity_level > 1)
139 PCL_DEBUG ("[pcl::RandomizedMEstimatorSampleConsensus::computeModel] Trial %d out of %d. Best penalty is %f.\n", iterations_, static_cast<int> (std::ceil (k)), d_best_penalty);
140 if (iterations_ > max_iterations_)
141 {
142 if (debug_verbosity_level > 0)
143 PCL_DEBUG ("[pcl::RandomizedMEstimatorSampleConsensus::computeModel] MSAC reached the maximum number of trials.\n");
144 break;
145 }
146 }
147
148 if (model_.empty ())
149 {
150 if (debug_verbosity_level > 0)
151 PCL_DEBUG ("[pcl::RandomizedMEstimatorSampleConsensus::computeModel] Unable to find a solution!\n");
152 return (false);
153 }
154
155 // Iterate through the 3d points and calculate the distances from them to the model again
156 sac_model_->getDistancesToModel (model_coefficients_, distances);
157 Indices &indices = *sac_model_->getIndices ();
158 if (distances.size () != indices.size ())
159 {
160 PCL_ERROR ("[pcl::RandomizedMEstimatorSampleConsensus::computeModel] Estimated distances (%lu) differs than the normal of indices (%lu).\n", distances.size (), indices.size ());
161 return (false);
162 }
163
164 inliers_.resize (distances.size ());
165 // Get the inliers for the best model found
166 n_inliers_count = 0;
167 for (std::size_t i = 0; i < distances.size (); ++i)
168 if (distances[i] <= threshold_)
169 inliers_[n_inliers_count++] = indices[i];
170
171 // Resize the inliers vector
172 inliers_.resize (n_inliers_count);
173
174 if (debug_verbosity_level > 0)
175 PCL_DEBUG ("[pcl::RandomizedMEstimatorSampleConsensus::computeModel] Model: %lu size, %d inliers.\n", model_.size (), n_inliers_count);
176
177 return (true);
178 }
179
180 #define PCL_INSTANTIATE_RandomizedMEstimatorSampleConsensus(T) template class PCL_EXPORTS pcl::RandomizedMEstimatorSampleConsensus<T>;
181
182 #endif // PCL_SAMPLE_CONSENSUS_IMPL_RMSAC_H_
183
184