Point Cloud Library (PCL)  1.11.1
statistical_outlier_removal.hpp
1 /*
2  * Software License Agreement (BSD License)
3  *
4  * Point Cloud Library (PCL) - www.pointclouds.org
5  * Copyright (c) 2010-2012, Willow Garage, 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  * $Id$
37  *
38  */
39 
40 #ifndef PCL_FILTERS_IMPL_STATISTICAL_OUTLIER_REMOVAL_H_
41 #define PCL_FILTERS_IMPL_STATISTICAL_OUTLIER_REMOVAL_H_
42 
43 #include <pcl/filters/statistical_outlier_removal.h>
44 
45 ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
46 template <typename PointT> void
48 {
49  // Initialize the search class
50  if (!searcher_)
51  {
52  if (input_->isOrganized ())
53  searcher_.reset (new pcl::search::OrganizedNeighbor<PointT> ());
54  else
55  searcher_.reset (new pcl::search::KdTree<PointT> (false));
56  }
57  searcher_->setInputCloud (input_);
58 
59  // The arrays to be used
60  std::vector<int> nn_indices (mean_k_);
61  std::vector<float> nn_dists (mean_k_);
62  std::vector<float> distances (indices_->size ());
63  indices.resize (indices_->size ());
64  removed_indices_->resize (indices_->size ());
65  int oii = 0, rii = 0; // oii = output indices iterator, rii = removed indices iterator
66 
67  // First pass: Compute the mean distances for all points with respect to their k nearest neighbors
68  int valid_distances = 0;
69  for (int iii = 0; iii < static_cast<int> (indices_->size ()); ++iii) // iii = input indices iterator
70  {
71  if (!std::isfinite ((*input_)[(*indices_)[iii]].x) ||
72  !std::isfinite ((*input_)[(*indices_)[iii]].y) ||
73  !std::isfinite ((*input_)[(*indices_)[iii]].z))
74  {
75  distances[iii] = 0.0;
76  continue;
77  }
78 
79  // Perform the nearest k search
80  if (searcher_->nearestKSearch ((*indices_)[iii], mean_k_ + 1, nn_indices, nn_dists) == 0)
81  {
82  distances[iii] = 0.0;
83  PCL_WARN ("[pcl::%s::applyFilter] Searching for the closest %d neighbors failed.\n", getClassName ().c_str (), mean_k_);
84  continue;
85  }
86 
87  // Calculate the mean distance to its neighbors
88  double dist_sum = 0.0;
89  for (int k = 1; k < mean_k_ + 1; ++k) // k = 0 is the query point
90  dist_sum += sqrt (nn_dists[k]);
91  distances[iii] = static_cast<float> (dist_sum / mean_k_);
92  valid_distances++;
93  }
94 
95  // Estimate the mean and the standard deviation of the distance vector
96  double sum = 0, sq_sum = 0;
97  for (const float &distance : distances)
98  {
99  sum += distance;
100  sq_sum += distance * distance;
101  }
102  double mean = sum / static_cast<double>(valid_distances);
103  double variance = (sq_sum - sum * sum / static_cast<double>(valid_distances)) / (static_cast<double>(valid_distances) - 1);
104  double stddev = sqrt (variance);
105  //getMeanStd (distances, mean, stddev);
106 
107  double distance_threshold = mean + std_mul_ * stddev;
108 
109  // Second pass: Classify the points on the computed distance threshold
110  for (int iii = 0; iii < static_cast<int> (indices_->size ()); ++iii) // iii = input indices iterator
111  {
112  // Points having a too high average distance are outliers and are passed to removed indices
113  // Unless negative was set, then it's the opposite condition
114  if ((!negative_ && distances[iii] > distance_threshold) || (negative_ && distances[iii] <= distance_threshold))
115  {
116  if (extract_removed_indices_)
117  (*removed_indices_)[rii++] = (*indices_)[iii];
118  continue;
119  }
120 
121  // Otherwise it was a normal point for output (inlier)
122  indices[oii++] = (*indices_)[iii];
123  }
124 
125  // Resize the output arrays
126  indices.resize (oii);
127  removed_indices_->resize (rii);
128 }
129 
130 #define PCL_INSTANTIATE_StatisticalOutlierRemoval(T) template class PCL_EXPORTS pcl::StatisticalOutlierRemoval<T>;
131 
132 #endif // PCL_FILTERS_IMPL_STATISTICAL_OUTLIER_REMOVAL_H_
133 
pcl::geometry::distance
float distance(const PointT &p1, const PointT &p2)
Definition: geometry.h:60
pcl::StatisticalOutlierRemoval::applyFilterIndices
void applyFilterIndices(std::vector< int > &indices)
Filtered results are indexed by an indices array.
Definition: statistical_outlier_removal.hpp:47
pcl::search::KdTree< PointT >
pcl::search::OrganizedNeighbor
OrganizedNeighbor is a class for optimized nearest neigbhor search in organized point clouds.
Definition: organized.h:64