Point Cloud Library (PCL)
1.3.1
|
00001 /* 00002 * Software License Agreement (BSD License) 00003 * 00004 * Copyright (c) 2009, Willow Garage, Inc. 00005 * All rights reserved. 00006 * 00007 * Redistribution and use in source and binary forms, with or without 00008 * modification, are permitted provided that the following conditions 00009 * are met: 00010 * 00011 * * Redistributions of source code must retain the above copyright 00012 * notice, this list of conditions and the following disclaimer. 00013 * * Redistributions in binary form must reproduce the above 00014 * copyright notice, this list of conditions and the following 00015 * disclaimer in the documentation and/or other materials provided 00016 * with the distribution. 00017 * * Neither the name of Willow Garage, Inc. nor the names of its 00018 * contributors may be used to endorse or promote products derived 00019 * from this software without specific prior written permission. 00020 * 00021 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS 00022 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT 00023 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS 00024 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE 00025 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, 00026 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, 00027 * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; 00028 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER 00029 * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 00030 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN 00031 * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 00032 * POSSIBILITY OF SUCH DAMAGE. 00033 * 00034 * $Id: lmeds.hpp 3280 2011-11-30 18:31:35Z gedikli $ 00035 * 00036 */ 00037 00038 #ifndef PCL_SAMPLE_CONSENSUS_IMPL_LMEDS_H_ 00039 #define PCL_SAMPLE_CONSENSUS_IMPL_LMEDS_H_ 00040 00041 #include "pcl/sample_consensus/lmeds.h" 00042 00044 template <typename PointT> bool 00045 pcl::LeastMedianSquares<PointT>::computeModel (int debug_verbosity_level) 00046 { 00047 // Warn and exit if no threshold was set 00048 if (threshold_ == DBL_MAX) 00049 { 00050 PCL_ERROR ("[pcl::LeastMedianSquares::computeModel] No threshold set!\n"); 00051 return (false); 00052 } 00053 00054 iterations_ = 0; 00055 double d_best_penalty = DBL_MAX; 00056 00057 std::vector<int> best_model; 00058 std::vector<int> selection; 00059 Eigen::VectorXf model_coefficients; 00060 std::vector<double> distances; 00061 00062 int n_inliers_count = 0; 00063 00064 unsigned skipped_count = 0; 00065 // supress infinite loops by just allowing 10 x maximum allowed iterations for invalid model parameters! 00066 const unsigned max_skip = max_iterations_ * 10; 00067 00068 // Iterate 00069 while (iterations_ < max_iterations_ && skipped_count < max_skip) 00070 { 00071 // Get X samples which satisfy the model criteria 00072 sac_model_->getSamples (iterations_, selection); 00073 00074 if (selection.empty ()) break; 00075 00076 // Search for inliers in the point cloud for the current plane model M 00077 if (!sac_model_->computeModelCoefficients (selection, model_coefficients)) 00078 { 00079 //iterations_++; 00080 ++skipped_count; 00081 continue; 00082 } 00083 00084 double d_cur_penalty = 0; 00085 // d_cur_penalty = sum (min (dist, threshold)) 00086 00087 // Iterate through the 3d points and calculate the distances from them to the model 00088 sac_model_->getDistancesToModel (model_coefficients, distances); 00089 00090 // No distances? The model must not respect the user given constraints 00091 if (distances.empty ()) 00092 { 00093 //iterations_++; 00094 ++skipped_count; 00095 continue; 00096 } 00097 00098 std::sort (distances.begin (), distances.end ()); 00099 // d_cur_penalty = median (distances) 00100 int mid = sac_model_->getIndices ()->size () / 2; 00101 if (mid >= (int)distances.size ()) 00102 { 00103 //iterations_++; 00104 ++skipped_count; 00105 continue; 00106 } 00107 00108 // Do we have a "middle" point or should we "estimate" one ? 00109 if (sac_model_->getIndices ()->size () % 2 == 0) 00110 d_cur_penalty = (sqrt (distances[mid-1]) + sqrt (distances[mid])) / 2; 00111 else 00112 d_cur_penalty = sqrt (distances[mid]); 00113 00114 // Better match ? 00115 if (d_cur_penalty < d_best_penalty) 00116 { 00117 d_best_penalty = d_cur_penalty; 00118 00119 // Save the current model/coefficients selection as being the best so far 00120 model_ = selection; 00121 model_coefficients_ = model_coefficients; 00122 } 00123 00124 ++iterations_; 00125 if (debug_verbosity_level > 1) 00126 PCL_DEBUG ("[pcl::LeastMedianSquares::computeModel] Trial %d out of %d. Best penalty is %f.\n", iterations_, max_iterations_, d_best_penalty); 00127 } 00128 00129 if (model_.empty ()) 00130 { 00131 if (debug_verbosity_level > 0) 00132 PCL_DEBUG ("[pcl::LeastMedianSquares::computeModel] Unable to find a solution!\n"); 00133 return (false); 00134 } 00135 00136 // Classify the data points into inliers and outliers 00137 // Sigma = 1.4826 * (1 + 5 / (n-d)) * sqrt (M) 00138 // @note: See "Robust Regression Methods for Computer Vision: A Review" 00139 //double sigma = 1.4826 * (1 + 5 / (sac_model_->getIndices ()->size () - best_model.size ())) * sqrt (d_best_penalty); 00140 //double threshold = 2.5 * sigma; 00141 00142 // Iterate through the 3d points and calculate the distances from them to the model again 00143 sac_model_->getDistancesToModel (model_coefficients_, distances); 00144 // No distances? The model must not respect the user given constraints 00145 if (distances.empty ()) 00146 { 00147 PCL_ERROR ("[pcl::LeastMedianSquares::computeModel] The model found failed to verify against the given constraints!\n"); 00148 return (false); 00149 } 00150 00151 std::vector<int> &indices = *sac_model_->getIndices (); 00152 00153 if (distances.size () != indices.size ()) 00154 { 00155 PCL_ERROR ("[pcl::LeastMedianSquares::computeModel] Estimated distances (%lu) differs than the normal of indices (%lu).\n", (unsigned long)distances.size (), (unsigned long)indices.size ()); 00156 return (false); 00157 } 00158 00159 inliers_.resize (distances.size ()); 00160 // Get the inliers for the best model found 00161 n_inliers_count = 0; 00162 for (size_t i = 0; i < distances.size (); ++i) 00163 if (distances[i] <= threshold_) 00164 inliers_[n_inliers_count++] = indices[i]; 00165 00166 // Resize the inliers vector 00167 inliers_.resize (n_inliers_count); 00168 00169 if (debug_verbosity_level > 0) 00170 PCL_DEBUG ("[pcl::LeastMedianSquares::computeModel] Model: %lu size, %d inliers.\n", (unsigned long)model_.size (), n_inliers_count); 00171 00172 return (true); 00173 } 00174 00175 #define PCL_INSTANTIATE_LeastMedianSquares(T) template class PCL_EXPORTS pcl::LeastMedianSquares<T>; 00176 00177 #endif // PCL_SAMPLE_CONSENSUS_IMPL_LMEDS_H_ 00178