Public Types | Public Member Functions

pcl::FPFHEstimation< PointInT, PointNT, PointOutT > Class Template Reference
[Module features]

FPFHEstimation estimates the Fast Point Feature Histogram (FPFH) descriptor for a given point cloud dataset containing points and normals. More...

#include <pcl/features/fpfh.h>

Inheritance diagram for pcl::FPFHEstimation< PointInT, PointNT, PointOutT >:
Inheritance graph
[legend]
Collaboration diagram for pcl::FPFHEstimation< PointInT, PointNT, PointOutT >:
Collaboration graph
[legend]

List of all members.

Public Types

typedef Feature< PointInT,
PointOutT >::PointCloudOut 
PointCloudOut
typedef pcl::PointCloud< PointNT > PointCloudN
typedef PointCloudN::Ptr PointCloudNPtr
typedef PointCloudN::ConstPtr PointCloudNConstPtr
typedef boost::shared_ptr
< FeatureFromNormals< PointInT,
PointNT, PointOutT > > 
Ptr
typedef boost::shared_ptr
< const FeatureFromNormals
< PointInT, PointNT, PointOutT > > 
ConstPtr
typedef PCLBase< PointInT > BaseClass
typedef pcl::search::Search
< PointInT > 
KdTree
typedef pcl::search::Search
< PointInT >::Ptr 
KdTreePtr
typedef boost::function< int(size_t,
double, std::vector< int >
&, std::vector< float > &)> 
SearchMethod
typedef boost::function< int(const
PointCloudIn &cloud, size_t
index, double, std::vector
< int > &, std::vector< float > &)> 
SearchMethodSurface
typedef pcl::PointCloud< PointInT > PointCloud
typedef PointCloud::Ptr PointCloudPtr
typedef PointCloud::ConstPtr PointCloudConstPtr
typedef PointIndices::Ptr PointIndicesPtr
typedef PointIndices::ConstPtr PointIndicesConstPtr

Public Member Functions

 FPFHEstimation ()
 Empty constructor.
bool computePairFeatures (const pcl::PointCloud< PointInT > &cloud, const pcl::PointCloud< PointNT > &normals, int p_idx, int q_idx, float &f1, float &f2, float &f3, float &f4)
 Compute the 4-tuple representation containing the three angles and one distance between two points represented by Cartesian coordinates and normals.
void computePointSPFHSignature (const pcl::PointCloud< PointInT > &cloud, const pcl::PointCloud< PointNT > &normals, int p_idx, int row, const std::vector< int > &indices, Eigen::MatrixXf &hist_f1, Eigen::MatrixXf &hist_f2, Eigen::MatrixXf &hist_f3)
 Estimate the SPFH (Simple Point Feature Histograms) individual signatures of the three angular (f1, f2, f3) features for a given point based on its spatial neighborhood of 3D points with normals.
void weightPointSPFHSignature (const Eigen::MatrixXf &hist_f1, const Eigen::MatrixXf &hist_f2, const Eigen::MatrixXf &hist_f3, const std::vector< int > &indices, const std::vector< float > &dists, Eigen::VectorXf &fpfh_histogram)
 Weight the SPFH (Simple Point Feature Histograms) individual histograms to create the final FPFH (Fast Point Feature Histogram) for a given point based on its 3D spatial neighborhood.
void setNrSubdivisions (int nr_bins_f1, int nr_bins_f2, int nr_bins_f3)
 Set the number of subdivisions for each angular feature interval.
void getNrSubdivisions (int &nr_bins_f1, int &nr_bins_f2, int &nr_bins_f3)
 Get the number of subdivisions for each angular feature interval.
void setInputNormals (const PointCloudNConstPtr &normals)
 Provide a pointer to the input dataset that contains the point normals of the XYZ dataset.
PointCloudNConstPtr getInputNormals ()
 Get a pointer to the normals of the input XYZ point cloud dataset.
void setSearchSurface (const PointCloudInConstPtr &cloud)
 Provide a pointer to a dataset to add additional information to estimate the features for every point in the input dataset.
PointCloudInConstPtr getSearchSurface ()
 Get a pointer to the surface point cloud dataset.
void setSearchMethod (const KdTreePtr &tree)
 Provide a pointer to the search object.
KdTreePtr getSearchMethod ()
 Get a pointer to the search method used.
double getSearchParameter ()
 Get the internal search parameter.
void setKSearch (int k)
 Set the number of k nearest neighbors to use for the feature estimation.
int getKSearch ()
 get the number of k nearest neighbors used for the feature estimation.
void setRadiusSearch (double radius)
 Set the sphere radius that is to be used for determining the nearest neighbors used for the feature estimation.
double getRadiusSearch ()
 Get the sphere radius used for determining the neighbors.
void compute (PointCloudOut &output)
 Base method for feature estimation for all points given in <setInputCloud (), setIndices ()> using the surface in setSearchSurface () and the spatial locator in setSearchMethod ().
int searchForNeighbors (size_t index, double parameter, std::vector< int > &indices, std::vector< float > &distances) const
 Search for k-nearest neighbors using the spatial locator from setSearchmethod, and the given surface from setSearchSurface.
int searchForNeighbors (const PointCloudIn &cloud, size_t index, double parameter, std::vector< int > &indices, std::vector< float > &distances) const
 Search for k-nearest neighbors using the spatial locator from setSearchmethod, and the given surface from setSearchSurface.
virtual void setInputCloud (const PointCloudConstPtr &cloud)
 Provide a pointer to the input dataset.
PointCloudConstPtr const getInputCloud ()
 Get a pointer to the input point cloud dataset.
void setIndices (const IndicesPtr &indices)
 Provide a pointer to the vector of indices that represents the input data.
void setIndices (const PointIndicesConstPtr &indices)
 Provide a pointer to the vector of indices that represents the input data.
void setIndices (size_t row_start, size_t col_start, size_t nb_rows, size_t nb_cols)
 Set the indices for the points laying within an interest region of the point cloud.
IndicesPtr const getIndices ()
 Get a pointer to the vector of indices used.
const PointInT & operator[] (size_t pos)
 Override PointCloud operator[] to shorten code.

Detailed Description

template<typename PointInT, typename PointNT, typename PointOutT>
class pcl::FPFHEstimation< PointInT, PointNT, PointOutT >

FPFHEstimation estimates the Fast Point Feature Histogram (FPFH) descriptor for a given point cloud dataset containing points and normals.

Note:
If you use this code in any academic work, please cite:
Note:
The code is stateful as we do not expect this class to be multicore parallelized. Please look at FPFHEstimationOMP for examples on parallel implementations of the FPFH (Fast Point Feature Histogram).
Author:
Radu Bogdan Rusu

Definition at line 71 of file fpfh.h.


Member Typedef Documentation

template<typename PointInT, typename PointOutT>
typedef PCLBase<PointInT> pcl::Feature< PointInT, PointOutT >::BaseClass [inherited]

Reimplemented in pcl::NarfDescriptor, and pcl::RangeImageBorderExtractor.

Definition at line 103 of file feature.h.

template<typename PointInT, typename PointNT, typename PointOutT>
typedef boost::shared_ptr< const FeatureFromNormals<PointInT, PointNT, PointOutT> > pcl::FeatureFromNormals< PointInT, PointNT, PointOutT >::ConstPtr [inherited]

Reimplemented from pcl::Feature< PointInT, PointOutT >.

Definition at line 290 of file feature.h.

template<typename PointInT, typename PointOutT>
typedef pcl::search::Search<PointInT> pcl::Feature< PointInT, PointOutT >::KdTree [inherited]

Definition at line 108 of file feature.h.

template<typename PointInT, typename PointOutT>
typedef pcl::search::Search<PointInT>::Ptr pcl::Feature< PointInT, PointOutT >::KdTreePtr [inherited]

Reimplemented in pcl::CVFHEstimation< PointInT, PointNT, PointOutT >.

Definition at line 109 of file feature.h.

typedef pcl::PointCloud<PointInT > pcl::PCLBase< PointInT >::PointCloud [inherited]

Reimplemented in pcl::ConcaveHull< PointInT >, and pcl::ConvexHull< PointInT >.

Definition at line 74 of file pcl_base.h.

typedef PointCloud::ConstPtr pcl::PCLBase< PointInT >::PointCloudConstPtr [inherited]

Reimplemented in pcl::ConcaveHull< PointInT >, and pcl::ConvexHull< PointInT >.

Definition at line 76 of file pcl_base.h.

template<typename PointInT, typename PointNT, typename PointOutT>
typedef pcl::PointCloud<PointNT> pcl::FeatureFromNormals< PointInT, PointNT, PointOutT >::PointCloudN [inherited]

Reimplemented in pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >.

Definition at line 285 of file feature.h.

template<typename PointInT, typename PointNT, typename PointOutT>
typedef PointCloudN::ConstPtr pcl::FeatureFromNormals< PointInT, PointNT, PointOutT >::PointCloudNConstPtr [inherited]

Reimplemented in pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >.

Definition at line 287 of file feature.h.

template<typename PointInT, typename PointNT, typename PointOutT>
typedef PointCloudN::Ptr pcl::FeatureFromNormals< PointInT, PointNT, PointOutT >::PointCloudNPtr [inherited]

Reimplemented in pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >.

Definition at line 286 of file feature.h.

template<typename PointInT , typename PointNT , typename PointOutT >
typedef Feature<PointInT, PointOutT>::PointCloudOut pcl::FPFHEstimation< PointInT, PointNT, PointOutT >::PointCloudOut
typedef PointCloud::Ptr pcl::PCLBase< PointInT >::PointCloudPtr [inherited]
typedef PointIndices::ConstPtr pcl::PCLBase< PointInT >::PointIndicesConstPtr [inherited]

Definition at line 79 of file pcl_base.h.

typedef PointIndices::Ptr pcl::PCLBase< PointInT >::PointIndicesPtr [inherited]

Definition at line 78 of file pcl_base.h.

template<typename PointInT, typename PointNT, typename PointOutT>
typedef boost::shared_ptr< FeatureFromNormals<PointInT, PointNT, PointOutT> > pcl::FeatureFromNormals< PointInT, PointNT, PointOutT >::Ptr [inherited]

Reimplemented from pcl::Feature< PointInT, PointOutT >.

Definition at line 289 of file feature.h.

template<typename PointInT, typename PointOutT>
typedef boost::function<int (size_t, double, std::vector<int> &, std::vector<float> &)> pcl::Feature< PointInT, PointOutT >::SearchMethod [inherited]

Definition at line 117 of file feature.h.

template<typename PointInT, typename PointOutT>
typedef boost::function<int (const PointCloudIn &cloud, size_t index, double, std::vector<int> &, std::vector<float> &)> pcl::Feature< PointInT, PointOutT >::SearchMethodSurface [inherited]

Definition at line 118 of file feature.h.


Constructor & Destructor Documentation

template<typename PointInT , typename PointNT , typename PointOutT >
pcl::FPFHEstimation< PointInT, PointNT, PointOutT >::FPFHEstimation (  )  [inline]

Empty constructor.

Definition at line 86 of file fpfh.h.


Member Function Documentation

template<typename PointInT , typename PointOutT >
void pcl::Feature< PointInT, PointOutT >::compute ( PointCloudOut output  )  [inherited]

Base method for feature estimation for all points given in <setInputCloud (), setIndices ()> using the surface in setSearchSurface () and the spatial locator in setSearchMethod ().

Parameters:
output the resultant point cloud model dataset containing the estimated features

Reimplemented in pcl::NarfDescriptor, and pcl::RangeImageBorderExtractor.

Definition at line 249 of file feature.hpp.

template<typename PointInT , typename PointNT , typename PointOutT >
bool pcl::FPFHEstimation< PointInT, PointNT, PointOutT >::computePairFeatures ( const pcl::PointCloud< PointInT > &  cloud,
const pcl::PointCloud< PointNT > &  normals,
int  p_idx,
int  q_idx,
float &  f1,
float &  f2,
float &  f3,
float &  f4 
)

Compute the 4-tuple representation containing the three angles and one distance between two points represented by Cartesian coordinates and normals.

Note:
For explanations about the features, please see the literature mentioned above (the order of the features might be different).
Parameters:
cloud the dataset containing the XYZ Cartesian coordinates of the two points
normals the dataset containing the surface normals (assuming normalized vectors) at each point in cloud
p_idx the index of the first point (source)
q_idx the index of the second point (target)
f1 the first angular feature (angle between the projection of nq_idx and u)
f2 the second angular feature (angle between nq_idx and v)
f3 the third angular feature (angle between np_idx and |p_idx - q_idx|)
f4 the distance feature (p_idx - q_idx)

Definition at line 46 of file fpfh.hpp.

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::FPFHEstimation< PointInT, PointNT, PointOutT >::computePointSPFHSignature ( const pcl::PointCloud< PointInT > &  cloud,
const pcl::PointCloud< PointNT > &  normals,
int  p_idx,
int  row,
const std::vector< int > &  indices,
Eigen::MatrixXf &  hist_f1,
Eigen::MatrixXf &  hist_f2,
Eigen::MatrixXf &  hist_f3 
)

Estimate the SPFH (Simple Point Feature Histograms) individual signatures of the three angular (f1, f2, f3) features for a given point based on its spatial neighborhood of 3D points with normals.

Parameters:
cloud the dataset containing the XYZ Cartesian coordinates of the two points
normals the dataset containing the surface normals at each point in cloud
p_idx the index of the query point (source)
row the index row in feature histogramms
indices the k-neighborhood point indices in the dataset
hist_f1 the resultant SPFH histogram for feature f1
hist_f2 the resultant SPFH histogram for feature f2
hist_f3 the resultant SPFH histogram for feature f3

Definition at line 59 of file fpfh.hpp.

IndicesPtr const pcl::PCLBase< PointInT >::getIndices (  )  [inline, inherited]

Get a pointer to the vector of indices used.

Definition at line 171 of file pcl_base.h.

PointCloudConstPtr const pcl::PCLBase< PointInT >::getInputCloud (  )  [inline, inherited]

Get a pointer to the input point cloud dataset.

Definition at line 99 of file pcl_base.h.

template<typename PointInT, typename PointNT, typename PointOutT>
PointCloudNConstPtr pcl::FeatureFromNormals< PointInT, PointNT, PointOutT >::getInputNormals (  )  [inline, inherited]

Get a pointer to the normals of the input XYZ point cloud dataset.

Definition at line 312 of file feature.h.

template<typename PointInT, typename PointOutT>
int pcl::Feature< PointInT, PointOutT >::getKSearch (  )  [inline, inherited]

get the number of k nearest neighbors used for the feature estimation.

Definition at line 166 of file feature.h.

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::FPFHEstimation< PointInT, PointNT, PointOutT >::getNrSubdivisions ( int &  nr_bins_f1,
int &  nr_bins_f2,
int &  nr_bins_f3 
) [inline]

Get the number of subdivisions for each angular feature interval.

Definition at line 157 of file fpfh.h.

template<typename PointInT, typename PointOutT>
double pcl::Feature< PointInT, PointOutT >::getRadiusSearch (  )  [inline, inherited]

Get the sphere radius used for determining the neighbors.

Definition at line 177 of file feature.h.

template<typename PointInT, typename PointOutT>
KdTreePtr pcl::Feature< PointInT, PointOutT >::getSearchMethod (  )  [inline, inherited]

Get a pointer to the search method used.

Definition at line 152 of file feature.h.

template<typename PointInT, typename PointOutT>
double pcl::Feature< PointInT, PointOutT >::getSearchParameter (  )  [inline, inherited]

Get the internal search parameter.

Definition at line 156 of file feature.h.

template<typename PointInT, typename PointOutT>
PointCloudInConstPtr pcl::Feature< PointInT, PointOutT >::getSearchSurface (  )  [inline, inherited]

Get a pointer to the surface point cloud dataset.

Definition at line 142 of file feature.h.

const PointInT & pcl::PCLBase< PointInT >::operator[] ( size_t  pos  )  [inline, inherited]

Override PointCloud operator[] to shorten code.

Note:
this method can be called instead of (*input_)[(*indices_)[pos]] or input_->points[(*indices_)[pos]]
Parameters:
pos position in indices_ vector

Definition at line 178 of file pcl_base.h.

template<typename PointInT, typename PointOutT>
int pcl::Feature< PointInT, PointOutT >::searchForNeighbors ( size_t  index,
double  parameter,
std::vector< int > &  indices,
std::vector< float > &  distances 
) const [inline, inherited]

Search for k-nearest neighbors using the spatial locator from setSearchmethod, and the given surface from setSearchSurface.

Parameters:
index the index of the query point
parameter the search parameter (either k or radius)
indices the resultant vector of indices representing the k-nearest neighbors
distances the resultant vector of distances representing the distances from the query point to the k-nearest neighbors

Definition at line 196 of file feature.h.

template<typename PointInT, typename PointOutT>
int pcl::Feature< PointInT, PointOutT >::searchForNeighbors ( const PointCloudIn cloud,
size_t  index,
double  parameter,
std::vector< int > &  indices,
std::vector< float > &  distances 
) const [inline, inherited]

Search for k-nearest neighbors using the spatial locator from setSearchmethod, and the given surface from setSearchSurface.

Parameters:
cloud the query point cloud
index the index of the query point in cloud
parameter the search parameter (either k or radius)
indices the resultant vector of indices representing the k-nearest neighbors
distances the resultant vector of distances representing the distances from the query point to the k-nearest neighbors

Definition at line 215 of file feature.h.

void pcl::PCLBase< PointInT >::setIndices ( size_t  row_start,
size_t  col_start,
size_t  nb_rows,
size_t  nb_cols 
) [inline, inherited]

Set the indices for the points laying within an interest region of the point cloud.

Note:
you shouldn't call this method on unorganized point clouds!
Parameters:
row_start the offset on rows
col_start the offset on columns
nb_rows the number of rows to be considered row_start included
nb_cols the number of columns to be considered col_start included

Definition at line 132 of file pcl_base.h.

void pcl::PCLBase< PointInT >::setIndices ( const PointIndicesConstPtr indices  )  [inline, inherited]

Provide a pointer to the vector of indices that represents the input data.

Parameters:
indices a pointer to the vector of indices that represents the input data.

Definition at line 116 of file pcl_base.h.

void pcl::PCLBase< PointInT >::setIndices ( const IndicesPtr indices  )  [inline, inherited]

Provide a pointer to the vector of indices that represents the input data.

Parameters:
indices a pointer to the vector of indices that represents the input data.

Definition at line 105 of file pcl_base.h.

virtual void pcl::PCLBase< PointInT >::setInputCloud ( const PointCloudConstPtr cloud  )  [inline, virtual, inherited]

Provide a pointer to the input dataset.

Parameters:
cloud the const boost shared pointer to a PointCloud message

Definition at line 95 of file pcl_base.h.

template<typename PointInT, typename PointNT, typename PointOutT>
void pcl::FeatureFromNormals< PointInT, PointNT, PointOutT >::setInputNormals ( const PointCloudNConstPtr normals  )  [inline, inherited]

Provide a pointer to the input dataset that contains the point normals of the XYZ dataset.

In case of search surface is set to be different from the input cloud, normals should correspond to the search surface, not the input cloud!

Parameters:
normals the const boost shared pointer to a PointCloud of normals. By convention, L2 norm of each normal should be 1.

Definition at line 308 of file feature.h.

template<typename PointInT, typename PointOutT>
void pcl::Feature< PointInT, PointOutT >::setKSearch ( int  k  )  [inline, inherited]

Set the number of k nearest neighbors to use for the feature estimation.

Parameters:
k the number of k-nearest neighbors

Reimplemented in pcl::RSDEstimation< PointInT, PointNT, PointOutT >.

Definition at line 162 of file feature.h.

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::FPFHEstimation< PointInT, PointNT, PointOutT >::setNrSubdivisions ( int  nr_bins_f1,
int  nr_bins_f2,
int  nr_bins_f3 
) [inline]

Set the number of subdivisions for each angular feature interval.

Parameters:
nr_bins_f1 number of subdivisions for the first angular feature
nr_bins_f2 number of subdivisions for the second angular feature
nr_bins_f3 number of subdivisions for the third angular feature

Definition at line 148 of file fpfh.h.

template<typename PointInT, typename PointOutT>
void pcl::Feature< PointInT, PointOutT >::setRadiusSearch ( double  radius  )  [inline, inherited]

Set the sphere radius that is to be used for determining the nearest neighbors used for the feature estimation.

Parameters:
radius the sphere radius used as the maximum distance to consider a point a neighbor

Definition at line 173 of file feature.h.

template<typename PointInT, typename PointOutT>
void pcl::Feature< PointInT, PointOutT >::setSearchMethod ( const KdTreePtr tree  )  [inline, inherited]

Provide a pointer to the search object.

Parameters:
tree a pointer to the spatial search object.

Definition at line 148 of file feature.h.

template<typename PointInT, typename PointOutT>
void pcl::Feature< PointInT, PointOutT >::setSearchSurface ( const PointCloudInConstPtr cloud  )  [inline, inherited]

Provide a pointer to a dataset to add additional information to estimate the features for every point in the input dataset.

This is optional, if this is not set, it will only use the data in the input cloud to estimate the features. This is useful when you only need to compute the features for a downsampled cloud.

Parameters:
cloud a pointer to a PointCloud message

Definition at line 133 of file feature.h.

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::FPFHEstimation< PointInT, PointNT, PointOutT >::weightPointSPFHSignature ( const Eigen::MatrixXf &  hist_f1,
const Eigen::MatrixXf &  hist_f2,
const Eigen::MatrixXf &  hist_f3,
const std::vector< int > &  indices,
const std::vector< float > &  dists,
Eigen::VectorXf &  fpfh_histogram 
)

Weight the SPFH (Simple Point Feature Histograms) individual histograms to create the final FPFH (Fast Point Feature Histogram) for a given point based on its 3D spatial neighborhood.

Parameters:
hist_f1 the histogram feature vector of f1 values over the given patch
hist_f2 the histogram feature vector of f2 values over the given patch
hist_f3 the histogram feature vector of f3 values over the given patch
indices the point indices of p_idx's k-neighborhood in the point cloud
dists the distances from p_idx to all its k-neighbors
fpfh_histogram the resultant FPFH histogram representing the feature at the query point

Definition at line 103 of file fpfh.hpp.


The documentation for this class was generated from the following files: