StatisticalOutlierRemoval uses point neighborhood statistics to filter outlier data. More...

#include <statistical_outlier_removal.h>

Inheritance diagram for pcl::StatisticalOutlierRemoval< PointT >:

Public Member Functions
int	getMeanK ()
	Get the number of nearest neighbors to use for mean distance estimation.
double	getStddevMulThresh ()
	Get the standard deviation multiplier for the distance threshold calculation.
void	setMeanK (int nr_k)
	Set the number of nearest neighbors to use for mean distance estimation.
void	setStddevMulThresh (double stddev_mult)
	Set the standard deviation multiplier for the distance threshold calculation.
	StatisticalOutlierRemoval (bool extract_removed_indices=false)
	Constructor.
Protected Types
typedef FilterIndices< PointT > ::PointCloud	PointCloud
typedef PointCloud::ConstPtr	PointCloudConstPtr
typedef PointCloud::Ptr	PointCloudPtr
typedef pcl::search::Search < PointT >::Ptr	SearcherPtr
Protected Member Functions
void	applyFilter (PointCloud &output)
	Filtered results are stored in a separate point cloud.
void	applyFilter (std::vector< int > &indices)
	Filtered results are indexed by an indices array.
void	applyFilterIndices (std::vector< int > &indices)
	Filtered results are indexed by an indices array.
Private Attributes
int	mean_k_
	The number of points to use for mean distance estimation.
SearcherPtr	searcher_
	A pointer to the spatial search object.
double	std_mul_
	Standard deviations threshold (i.e., points outside of $\mu \pm \sigma \cdot std\_mul$ will be marked as outliers).

Detailed Description

template<typename PointT>
class pcl::StatisticalOutlierRemoval< PointT >

StatisticalOutlierRemoval uses point neighborhood statistics to filter outlier data.

The algorithm iterates through the entire input twice: During the first iteration it will compute the average distance that each point has to its nearest k neighbors. The value of k can be set using setMeanK(). Next, the mean and standard deviation of all these distances are computed in order to determine a distance threshold. The distance threshold will be equal to: mean + stddev_mult * stddev. The multiplier for the standard deviation can be set using setStddevMulThresh(). During the next iteration the points will be classified as inlier or outlier if their average neighbor distance is below or above this threshold respectively.
The neighbors found for each query point will be found amongst ALL points of setInputCloud(), not just those indexed by setIndices(). The setIndices() method only indexes the points that will be iterated through as search query points.

For more information:

R. B. Rusu, Z. C. Marton, N. Blodow, M. Dolha, and M. Beetz. Towards 3D Point Cloud Based Object Maps for Household Environments Robotics and Autonomous Systems Journal (Special Issue on Semantic Knowledge), 2008.

Usage example:

 pcl::StatisticalOutlierRemoval<PointType> sorfilter (true); // Initializing with true will allow us to extract the removed indices
 sorfilter.setInputCloud (cloud_in);
 sorfilter.setMeanK (8);
 sorfilter.setStddevMulThresh (1.0);
 sorfilter.filter (*cloud_out);
 // The resulting cloud_out contains all points of cloud_in that have an average distance to their 8 nearest neighbors that is below the computed threshold
 // Using a standard deviation multiplier of 1.0 and assuming the average distances are normally distributed there is a 84.1% chance that a point will be an inlier
 indices_rem = sorfilter.getRemovedIndices ();
 // The indices_rem array indexes all points of cloud_in that are outliers

Author:: Radu Bogdan Rusu

Definition at line 81 of file statistical_outlier_removal.h.

Member Typedef Documentation

template<typename PointT>

typedef FilterIndices<PointT>::PointCloud pcl::StatisticalOutlierRemoval< PointT >::PointCloud [protected]

Reimplemented from pcl::FilterIndices< PointT >.

Definition at line 84 of file statistical_outlier_removal.h.

template<typename PointT>

typedef PointCloud::ConstPtr pcl::StatisticalOutlierRemoval< PointT >::PointCloudConstPtr [protected]

Reimplemented from pcl::Filter< PointT >.

Definition at line 86 of file statistical_outlier_removal.h.

template<typename PointT>

typedef PointCloud::Ptr pcl::StatisticalOutlierRemoval< PointT >::PointCloudPtr [protected]

Reimplemented from pcl::Filter< PointT >.

Definition at line 85 of file statistical_outlier_removal.h.

template<typename PointT>

typedef pcl::search::Search<PointT>::Ptr pcl::StatisticalOutlierRemoval< PointT >::SearcherPtr [protected]

Definition at line 87 of file statistical_outlier_removal.h.

Constructor & Destructor Documentation

template<typename PointT>

pcl::StatisticalOutlierRemoval< PointT >::StatisticalOutlierRemoval ( bool extract_removed_indices = false ) [inline]

Constructor.

Parameters:

[in] extract_removed_indices Set to true if you want to be able to extract the indices of points being removed (default = false).

Definition at line 93 of file statistical_outlier_removal.h.

Member Function Documentation

template<typename PointT >

void pcl::StatisticalOutlierRemoval< PointT >::applyFilter ( PointCloud & output ) [protected, virtual]

Filtered results are stored in a separate point cloud.

Parameters:

[out] output The resultant point cloud.

Implements pcl::Filter< PointT >.

Definition at line 48 of file statistical_outlier_removal.hpp.

template<typename PointT>

void pcl::StatisticalOutlierRemoval< PointT >::applyFilter ( std::vector< int > & indices ) [inline, protected, virtual]

Filtered results are indexed by an indices array.

Parameters:

[out] indices The resultant indices.

Implements pcl::FilterIndices< PointT >.

Definition at line 163 of file statistical_outlier_removal.h.

template<typename PointT >

void pcl::StatisticalOutlierRemoval< PointT >::applyFilterIndices ( std::vector< int > & indices ) [protected]

Filtered results are indexed by an indices array.

Parameters:

[out] indices The resultant indices.

Definition at line 73 of file statistical_outlier_removal.hpp.

template<typename PointT>

int pcl::StatisticalOutlierRemoval< PointT >::getMeanK ( ) [inline]

Get the number of nearest neighbors to use for mean distance estimation.

Returns:: The number of points to use for mean distance estimation.

Definition at line 115 of file statistical_outlier_removal.h.

template<typename PointT>

double pcl::StatisticalOutlierRemoval< PointT >::getStddevMulThresh ( ) [inline]

Get the standard deviation multiplier for the distance threshold calculation.

The distance threshold will be equal to: mean + stddev_mult * stddev. Points will be classified as inlier or outlier if their average neighbor distance is below or above this threshold respectively.

Parameters:

[in] stddev_mult The standard deviation multiplier.

Definition at line 137 of file statistical_outlier_removal.h.

template<typename PointT>

void pcl::StatisticalOutlierRemoval< PointT >::setMeanK ( int nr_k ) [inline]

Set the number of nearest neighbors to use for mean distance estimation.

Parameters:

[in] nr_k The number of points to use for mean distance estimation.

Definition at line 106 of file statistical_outlier_removal.h.

template<typename PointT>

void pcl::StatisticalOutlierRemoval< PointT >::setStddevMulThresh ( double stddev_mult ) [inline]

Set the standard deviation multiplier for the distance threshold calculation.

The distance threshold will be equal to: mean + stddev_mult * stddev. Points will be classified as inlier or outlier if their average neighbor distance is below or above this threshold respectively.

Parameters:

[in] stddev_mult The standard deviation multiplier.

Definition at line 126 of file statistical_outlier_removal.h.