Point Cloud Library (PCL)  1.10.1-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/common/io.h>
45 
46 ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
47 template <typename PointT> void
49 {
50  // Initialize the search class
51  if (!searcher_)
52  {
53  if (input_->isOrganized ())
54  searcher_.reset (new pcl::search::OrganizedNeighbor<PointT> ());
55  else
56  searcher_.reset (new pcl::search::KdTree<PointT> (false));
57  }
58  searcher_->setInputCloud (input_);
59 
60  // The arrays to be used
61  std::vector<int> nn_indices (mean_k_);
62  std::vector<float> nn_dists (mean_k_);
63  std::vector<float> distances (indices_->size ());
64  indices.resize (indices_->size ());
65  removed_indices_->resize (indices_->size ());
66  int oii = 0, rii = 0; // oii = output indices iterator, rii = removed indices iterator
67 
68  // First pass: Compute the mean distances for all points with respect to their k nearest neighbors
69  int valid_distances = 0;
70  for (int iii = 0; iii < static_cast<int> (indices_->size ()); ++iii) // iii = input indices iterator
71  {
72  if (!std::isfinite (input_->points[(*indices_)[iii]].x) ||
73  !std::isfinite (input_->points[(*indices_)[iii]].y) ||
74  !std::isfinite (input_->points[(*indices_)[iii]].z))
75  {
76  distances[iii] = 0.0;
77  continue;
78  }
79 
80  // Perform the nearest k search
81  if (searcher_->nearestKSearch ((*indices_)[iii], mean_k_ + 1, nn_indices, nn_dists) == 0)
82  {
83  distances[iii] = 0.0;
84  PCL_WARN ("[pcl::%s::applyFilter] Searching for the closest %d neighbors failed.\n", getClassName ().c_str (), mean_k_);
85  continue;
86  }
87 
88  // Calculate the mean distance to its neighbors
89  double dist_sum = 0.0;
90  for (int k = 1; k < mean_k_ + 1; ++k) // k = 0 is the query point
91  dist_sum += sqrt (nn_dists[k]);
92  distances[iii] = static_cast<float> (dist_sum / mean_k_);
93  valid_distances++;
94  }
95 
96  // Estimate the mean and the standard deviation of the distance vector
97  double sum = 0, sq_sum = 0;
98  for (const float &distance : distances)
99  {
100  sum += distance;
101  sq_sum += distance * distance;
102  }
103  double mean = sum / static_cast<double>(valid_distances);
104  double variance = (sq_sum - sum * sum / static_cast<double>(valid_distances)) / (static_cast<double>(valid_distances) - 1);
105  double stddev = sqrt (variance);
106  //getMeanStd (distances, mean, stddev);
107 
108  double distance_threshold = mean + std_mul_ * stddev;
109 
110  // Second pass: Classify the points on the computed distance threshold
111  for (int iii = 0; iii < static_cast<int> (indices_->size ()); ++iii) // iii = input indices iterator
112  {
113  // Points having a too high average distance are outliers and are passed to removed indices
114  // Unless negative was set, then it's the opposite condition
115  if ((!negative_ && distances[iii] > distance_threshold) || (negative_ && distances[iii] <= distance_threshold))
116  {
117  if (extract_removed_indices_)
118  (*removed_indices_)[rii++] = (*indices_)[iii];
119  continue;
120  }
121 
122  // Otherwise it was a normal point for output (inlier)
123  indices[oii++] = (*indices_)[iii];
124  }
125 
126  // Resize the output arrays
127  indices.resize (oii);
128  removed_indices_->resize (rii);
129 }
130 
131 #define PCL_INSTANTIATE_StatisticalOutlierRemoval(T) template class PCL_EXPORTS pcl::StatisticalOutlierRemoval<T>;
132 
133 #endif // PCL_FILTERS_IMPL_STATISTICAL_OUTLIER_REMOVAL_H_
134 
void applyFilterIndices(std::vector< int > &indices)
Filtered results are indexed by an indices array.
float distance(const PointT &p1, const PointT &p2)
Definition: geometry.h:60
OrganizedNeighbor is a class for optimized nearest neigbhor search in organized point clouds...
Definition: organized.h:63