Point Cloud Library (PCL)  1.12.1-dev
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 #include <pcl/search/organized.h> // for OrganizedNeighbor
45 #include <pcl/search/kdtree.h> // for KdTree
46 
47 ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
48 template <typename PointT> void
50 {
51  // Initialize the search class
52  if (!searcher_)
53  {
54  if (input_->isOrganized ())
55  searcher_.reset (new pcl::search::OrganizedNeighbor<PointT> ());
56  else
57  searcher_.reset (new pcl::search::KdTree<PointT> (false));
58  }
59  searcher_->setInputCloud (input_);
60 
61  // The arrays to be used
62  Indices nn_indices (mean_k_);
63  std::vector<float> nn_dists (mean_k_);
64  std::vector<float> distances (indices_->size ());
65  indices.resize (indices_->size ());
66  removed_indices_->resize (indices_->size ());
67  int oii = 0, rii = 0; // oii = output indices iterator, rii = removed indices iterator
68 
69  // First pass: Compute the mean distances for all points with respect to their k nearest neighbors
70  int valid_distances = 0;
71  for (int iii = 0; iii < static_cast<int> (indices_->size ()); ++iii) // iii = input indices iterator
72  {
73  if (!std::isfinite ((*input_)[(*indices_)[iii]].x) ||
74  !std::isfinite ((*input_)[(*indices_)[iii]].y) ||
75  !std::isfinite ((*input_)[(*indices_)[iii]].z))
76  {
77  distances[iii] = 0.0;
78  continue;
79  }
80 
81  // Perform the nearest k search
82  if (searcher_->nearestKSearch ((*indices_)[iii], mean_k_ + 1, nn_indices, nn_dists) == 0)
83  {
84  distances[iii] = 0.0;
85  PCL_WARN ("[pcl::%s::applyFilter] Searching for the closest %d neighbors failed.\n", getClassName ().c_str (), mean_k_);
86  continue;
87  }
88 
89  // Calculate the mean distance to its neighbors
90  double dist_sum = 0.0;
91  for (int k = 1; k < mean_k_ + 1; ++k) // k = 0 is the query point
92  dist_sum += sqrt (nn_dists[k]);
93  distances[iii] = static_cast<float> (dist_sum / mean_k_);
94  valid_distances++;
95  }
96 
97  // Estimate the mean and the standard deviation of the distance vector
98  double sum = 0, sq_sum = 0;
99  for (const float &distance : distances)
100  {
101  sum += distance;
102  sq_sum += distance * distance;
103  }
104  double mean = sum / static_cast<double>(valid_distances);
105  double variance = (sq_sum - sum * sum / static_cast<double>(valid_distances)) / (static_cast<double>(valid_distances) - 1);
106  double stddev = sqrt (variance);
107  //getMeanStd (distances, mean, stddev);
108 
109  double distance_threshold = mean + std_mul_ * stddev;
110 
111  // Second pass: Classify the points on the computed distance threshold
112  for (int iii = 0; iii < static_cast<int> (indices_->size ()); ++iii) // iii = input indices iterator
113  {
114  // Points having a too high average distance are outliers and are passed to removed indices
115  // Unless negative was set, then it's the opposite condition
116  if ((!negative_ && distances[iii] > distance_threshold) || (negative_ && distances[iii] <= distance_threshold))
117  {
118  if (extract_removed_indices_)
119  (*removed_indices_)[rii++] = (*indices_)[iii];
120  continue;
121  }
122 
123  // Otherwise it was a normal point for output (inlier)
124  indices[oii++] = (*indices_)[iii];
125  }
126 
127  // Resize the output arrays
128  indices.resize (oii);
129  removed_indices_->resize (rii);
130 }
131 
132 #define PCL_INSTANTIATE_StatisticalOutlierRemoval(T) template class PCL_EXPORTS pcl::StatisticalOutlierRemoval<T>;
133 
134 #endif // PCL_FILTERS_IMPL_STATISTICAL_OUTLIER_REMOVAL_H_
135 
void applyFilterIndices(Indices &indices)
Filtered results are indexed by an indices array.
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 neigbhor search in organized point clouds.
Definition: organized.h:61
float distance(const PointT &p1, const PointT &p2)
Definition: geometry.h:60
IndicesAllocator<> Indices
Type used for indices in PCL.
Definition: types.h:133