Point Cloud Library (PCL)  1.14.0-dev
List of all members | Public Types | Public Member Functions | Protected Member Functions
pcl::SpinImageEstimation< PointInT, PointNT, PointOutT > Class Template Reference

Estimates spin-image descriptors in the given input points. More...

#include <pcl/features/spin_image.h>

+ Inheritance diagram for pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >:
+ Collaboration diagram for pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >:

Public Types

using Ptr = shared_ptr< SpinImageEstimation< PointInT, PointNT, PointOutT > >
 
using ConstPtr = shared_ptr< const SpinImageEstimation< PointInT, PointNT, PointOutT > >
 
using PointCloudOut = typename Feature< PointInT, PointOutT >::PointCloudOut
 
using PointCloudN = pcl::PointCloud< PointNT >
 
using PointCloudNPtr = typename PointCloudN::Ptr
 
using PointCloudNConstPtr = typename PointCloudN::ConstPtr
 
using PointCloudIn = pcl::PointCloud< PointInT >
 
using PointCloudInPtr = typename PointCloudIn::Ptr
 
using PointCloudInConstPtr = typename PointCloudIn::ConstPtr
 
- Public Types inherited from pcl::Feature< PointInT, PointOutT >
using BaseClass = PCLBase< PointInT >
 
using Ptr = shared_ptr< Feature< PointInT, PointOutT > >
 
using ConstPtr = shared_ptr< const Feature< PointInT, PointOutT > >
 
using KdTree = pcl::search::Search< PointInT >
 
using KdTreePtr = typename KdTree::Ptr
 
using PointCloudIn = pcl::PointCloud< PointInT >
 
using PointCloudInPtr = typename PointCloudIn::Ptr
 
using PointCloudInConstPtr = typename PointCloudIn::ConstPtr
 
using PointCloudOut = pcl::PointCloud< PointOutT >
 
using SearchMethod = std::function< int(std::size_t, double, pcl::Indices &, std::vector< float > &)>
 
using SearchMethodSurface = std::function< int(const PointCloudIn &cloud, std::size_t index, double, pcl::Indices &, std::vector< float > &)>
 
- Public Types inherited from pcl::PCLBase< PointInT >
using PointCloud = pcl::PointCloud< PointInT >
 
using PointCloudPtr = typename PointCloud::Ptr
 
using PointCloudConstPtr = typename PointCloud::ConstPtr
 
using PointIndicesPtr = PointIndices::Ptr
 
using PointIndicesConstPtr = PointIndices::ConstPtr
 

Public Member Functions

 SpinImageEstimation (unsigned int image_width=8, double support_angle_cos=0.0, unsigned int min_pts_neighb=0)
 Constructs empty spin image estimator. More...
 
 ~SpinImageEstimation () override=default
 Empty destructor. More...
 
void setImageWidth (unsigned int bin_count)
 Sets spin-image resolution. More...
 
void setSupportAngle (double support_angle_cos)
 Sets the maximum angle for the point normal to get to support region. More...
 
void setMinPointCountInNeighbourhood (unsigned int min_pts_neighb)
 Sets minimal points count for spin image computation. More...
 
void setInputNormals (const PointCloudNConstPtr &normals)
 Provide a pointer to the input dataset that contains the point normals of the input XYZ dataset given by setInputCloud. More...
 
void setRotationAxis (const PointNT &axis)
 Sets single vector a rotation axis for all input points. More...
 
void setInputRotationAxes (const PointCloudNConstPtr &axes)
 Sets array of vectors as rotation axes for input points. More...
 
void useNormalsAsRotationAxis ()
 Sets input normals as rotation axes (default setting). More...
 
void setAngularDomain (bool is_angular=true)
 Sets/unsets flag for angular spin-image domain. More...
 
void setRadialStructure (bool is_radial=true)
 Sets/unsets flag for radial spin-image structure. More...
 
- Public Member Functions inherited from pcl::Feature< PointInT, PointOutT >
 Feature ()
 Empty constructor. More...
 
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. More...
 
PointCloudInConstPtr getSearchSurface () const
 Get a pointer to the surface point cloud dataset. More...
 
void setSearchMethod (const KdTreePtr &tree)
 Provide a pointer to the search object. More...
 
KdTreePtr getSearchMethod () const
 Get a pointer to the search method used. More...
 
double getSearchParameter () const
 Get the internal search parameter. More...
 
void setKSearch (int k)
 Set the number of k nearest neighbors to use for the feature estimation. More...
 
int getKSearch () const
 get the number of k nearest neighbors used for the feature estimation. More...
 
void setRadiusSearch (double radius)
 Set the sphere radius that is to be used for determining the nearest neighbors used for the feature estimation. More...
 
double getRadiusSearch () const
 Get the sphere radius used for determining the neighbors. More...
 
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 () More...
 
- Public Member Functions inherited from pcl::PCLBase< PointInT >
 PCLBase ()
 Empty constructor. More...
 
 PCLBase (const PCLBase &base)
 Copy constructor. More...
 
virtual ~PCLBase ()=default
 Destructor. More...
 
virtual void setInputCloud (const PointCloudConstPtr &cloud)
 Provide a pointer to the input dataset. More...
 
PointCloudConstPtr const getInputCloud () const
 Get a pointer to the input point cloud dataset. More...
 
virtual void setIndices (const IndicesPtr &indices)
 Provide a pointer to the vector of indices that represents the input data. More...
 
virtual void setIndices (const IndicesConstPtr &indices)
 Provide a pointer to the vector of indices that represents the input data. More...
 
virtual void setIndices (const PointIndicesConstPtr &indices)
 Provide a pointer to the vector of indices that represents the input data. More...
 
virtual void setIndices (std::size_t row_start, std::size_t col_start, std::size_t nb_rows, std::size_t nb_cols)
 Set the indices for the points laying within an interest region of the point cloud. More...
 
IndicesPtr getIndices ()
 Get a pointer to the vector of indices used. More...
 
IndicesConstPtr const getIndices () const
 Get a pointer to the vector of indices used. More...
 
const PointInT & operator[] (std::size_t pos) const
 Override PointCloud operator[] to shorten code. More...
 

Protected Member Functions

void computeFeature (PointCloudOut &output) override
 Estimate the Spin Image descriptors at a set of points given by setInputWithNormals() using the surface in setSearchSurfaceWithNormals() and the spatial locator. More...
 
bool initCompute () override
 initializes computations specific to spin-image. More...
 
Eigen::ArrayXXd computeSiForPoint (int index) const
 Computes a spin-image for the point of the scan. More...
 
- Protected Member Functions inherited from pcl::Feature< PointInT, PointOutT >
const std::string & getClassName () const
 Get a string representation of the name of this class. More...
 
virtual bool deinitCompute ()
 This method should get called after ending the actual computation. More...
 
int searchForNeighbors (std::size_t index, double parameter, pcl::Indices &indices, std::vector< float > &distances) const
 Search for k-nearest neighbors using the spatial locator from setSearchmethod, and the given surface from setSearchSurface. More...
 
int searchForNeighbors (const PointCloudIn &cloud, std::size_t index, double parameter, pcl::Indices &indices, std::vector< float > &distances) const
 Search for k-nearest neighbors using the spatial locator from setSearchmethod, and the given surface from setSearchSurface. More...
 
- Protected Member Functions inherited from pcl::PCLBase< PointInT >
bool initCompute ()
 This method should get called before starting the actual computation. More...
 
bool deinitCompute ()
 This method should get called after finishing the actual computation. More...
 

Additional Inherited Members

- Protected Attributes inherited from pcl::Feature< PointInT, PointOutT >
std::string feature_name_
 The feature name. More...
 
SearchMethodSurface search_method_surface_
 The search method template for points. More...
 
PointCloudInConstPtr surface_
 An input point cloud describing the surface that is to be used for nearest neighbors estimation. More...
 
KdTreePtr tree_
 A pointer to the spatial search object. More...
 
double search_parameter_
 The actual search parameter (from either search_radius_ or k_). More...
 
double search_radius_
 The nearest neighbors search radius for each point. More...
 
int k_
 The number of K nearest neighbors to use for each point. More...
 
bool fake_surface_
 If no surface is given, we use the input PointCloud as the surface. More...
 
- Protected Attributes inherited from pcl::PCLBase< PointInT >
PointCloudConstPtr input_
 The input point cloud dataset. More...
 
IndicesPtr indices_
 A pointer to the vector of point indices to use. More...
 
bool use_indices_
 Set to true if point indices are used. More...
 
bool fake_indices_
 If no set of indices are given, we construct a set of fake indices that mimic the input PointCloud. More...
 

Detailed Description

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

Estimates spin-image descriptors in the given input points.

This class represents spin image descriptor. Spin image is a histogram of point locations summed along the bins of the image. A 2D accumulator indexed by a and b is created. Next, the coordinates (a, b) are computed for a vertex in the surface mesh that is within the support of the spin image (explained below). The bin indexed by (a, b) in the accumulator is then incremented; bilinear interpolation is used to smooth the contribution of the vertex. This procedure is repeated for all vertices within the support of the spin image. The resulting accumulator can be thought of as an image; dark areas in the image correspond to bins that contain many projected points. As long as the size of the bins in the accumulator is greater than the median distance between vertices in the mesh (the definition of mesh resolution), the position of individual vertices will be averaged out during spin image generation.

Attention
The input normals given by setInputNormals have to match the input point cloud given by setInputCloud. This behavior is different than feature estimation methods that extend FeatureFromNormals, which match the normals with the search surface.

With the default parameters, pcl::Histogram<153> is a good choice for PointOutT. Of course the dimension of this descriptor must change to match the number of bins set by the parameters. If you use SpinImageEstimation with something other than pcl::Histogram<153>, you may need to put #define PCL_NO_PRECOMPILE 1 before including pcl/features/spin_image.h.

For further information please see:

The class also implements radial spin images and spin-images in angular domain (or both).

Author
Roman Shapovalov, Alexander Velizhev

Definition at line 89 of file spin_image.h.

Member Typedef Documentation

◆ ConstPtr

template<typename PointInT , typename PointNT , typename PointOutT >
using pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::ConstPtr = shared_ptr<const SpinImageEstimation<PointInT, PointNT, PointOutT> >

Definition at line 93 of file spin_image.h.

◆ PointCloudIn

template<typename PointInT , typename PointNT , typename PointOutT >
using pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::PointCloudIn = pcl::PointCloud<PointInT>

Definition at line 109 of file spin_image.h.

◆ PointCloudInConstPtr

template<typename PointInT , typename PointNT , typename PointOutT >
using pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::PointCloudInConstPtr = typename PointCloudIn::ConstPtr

Definition at line 111 of file spin_image.h.

◆ PointCloudInPtr

template<typename PointInT , typename PointNT , typename PointOutT >
using pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::PointCloudInPtr = typename PointCloudIn::Ptr

Definition at line 110 of file spin_image.h.

◆ PointCloudN

template<typename PointInT , typename PointNT , typename PointOutT >
using pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::PointCloudN = pcl::PointCloud<PointNT>

Definition at line 105 of file spin_image.h.

◆ PointCloudNConstPtr

template<typename PointInT , typename PointNT , typename PointOutT >
using pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::PointCloudNConstPtr = typename PointCloudN::ConstPtr

Definition at line 107 of file spin_image.h.

◆ PointCloudNPtr

template<typename PointInT , typename PointNT , typename PointOutT >
using pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::PointCloudNPtr = typename PointCloudN::Ptr

Definition at line 106 of file spin_image.h.

◆ PointCloudOut

template<typename PointInT , typename PointNT , typename PointOutT >
using pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::PointCloudOut = typename Feature<PointInT, PointOutT>::PointCloudOut

Definition at line 103 of file spin_image.h.

◆ Ptr

template<typename PointInT , typename PointNT , typename PointOutT >
using pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::Ptr = shared_ptr<SpinImageEstimation<PointInT, PointNT, PointOutT> >

Definition at line 92 of file spin_image.h.

Constructor & Destructor Documentation

◆ SpinImageEstimation()

template<typename PointInT , typename PointNT , typename PointOutT >
pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::SpinImageEstimation ( unsigned int  image_width = 8,
double  support_angle_cos = 0.0,
unsigned int  min_pts_neighb = 0 
)

Constructs empty spin image estimator.

Parameters
[in]image_widthspin-image resolution, number of bins along one dimension
[in]support_angle_cosminimal allowed cosine of the angle between the normals of input point and search surface point for the point to be retained in the support
[in]min_pts_neighbmin number of points in the support to correctly estimate spin-image. If at some point the support contains less points, exception is thrown

Definition at line 52 of file spin_image.hpp.

References pcl::Feature< PointInT, PointOutT >::feature_name_, and pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::setImageWidth().

◆ ~SpinImageEstimation()

template<typename PointInT , typename PointNT , typename PointOutT >
pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::~SpinImageEstimation ( )
overridedefault

Empty destructor.

Member Function Documentation

◆ computeFeature()

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::computeFeature ( PointCloudOut output)
overrideprotectedvirtual

Estimate the Spin Image descriptors at a set of points given by setInputWithNormals() using the surface in setSearchSurfaceWithNormals() and the spatial locator.

Parameters
[out]outputthe resultant point cloud that contains the Spin Image feature estimates

Implements pcl::Feature< PointInT, PointOutT >.

Definition at line 324 of file spin_image.hpp.

◆ computeSiForPoint()

template<typename PointInT , typename PointNT , typename PointOutT >
Eigen::ArrayXXd pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::computeSiForPoint ( int  index) const
protected

Computes a spin-image for the point of the scan.

Parameters
[in]indexthe index of the reference point in the input cloud
Returns
estimated spin-image (or its variant) as a matrix

Definition at line 69 of file spin_image.hpp.

References M_PI.

◆ initCompute()

template<typename PointInT , typename PointNT , typename PointOutT >
bool pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::initCompute
overrideprotectedvirtual

initializes computations specific to spin-image.

Returns
true iff input data and initialization are correct

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

Definition at line 239 of file spin_image.hpp.

References pcl::Feature< PointInT, PointOutT >::deinitCompute().

◆ setAngularDomain()

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::setAngularDomain ( bool  is_angular = true)
inline

Sets/unsets flag for angular spin-image domain.

Angular spin-image differs from the vanilla one in the way that not the points are collected in the bins but the angles between their normals and the normal to the reference point. For further information please see Endres, F., Plagemann, C., Stachniss, C., & Burgard, W. (2009). Unsupervised Discovery of Object Classes from Range Data using Latent Dirichlet Allocation. In Robotics: Science and Systems. Seattle, USA.

Parameters
[in]is_angulartrue for angular domain, false for point domain

Definition at line 252 of file spin_image.h.

◆ setImageWidth()

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::setImageWidth ( unsigned int  bin_count)
inline

Sets spin-image resolution.

Parameters
[in]bin_countspin-image resolution, number of bins along one dimension

Definition at line 134 of file spin_image.h.

Referenced by pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::SpinImageEstimation().

◆ setInputNormals()

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::setInputNormals ( const PointCloudNConstPtr normals)
inline

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

Attention
The input normals given by setInputNormals have to match the input point cloud given by setInputCloud. This behavior is different than feature estimation methods that extend FeatureFromNormals, which match the normals with the search surface.
Parameters
[in]normalsthe const boost shared pointer to a PointCloud of normals. By convention, L2 norm of each normal should be 1.

Definition at line 199 of file spin_image.h.

◆ setInputRotationAxes()

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::setInputRotationAxes ( const PointCloudNConstPtr axes)
inline

Sets array of vectors as rotation axes for input points.

Useful e.g. when one wants to use tangents instead of normals as rotation axes

Parameters
[in]axesunit-length vectors that serves as rotation axes for the corresponding input points' reference frames

Definition at line 224 of file spin_image.h.

◆ setMinPointCountInNeighbourhood()

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::setMinPointCountInNeighbourhood ( unsigned int  min_pts_neighb)
inline

Sets minimal points count for spin image computation.

Parameters
[in]min_pts_neighbmin number of points in the support to correctly estimate spin-image. If at some point the support contains less points, exception is thrown

Definition at line 183 of file spin_image.h.

◆ setRadialStructure()

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::setRadialStructure ( bool  is_radial = true)
inline

Sets/unsets flag for radial spin-image structure.

Instead of rectangular coordinate system for reference frame polar coordinates are used. Binning is done depending on the distance and inclination angle from the reference point

Parameters
[in]is_radialtrue for radial spin-image structure, false for rectangular

Definition at line 262 of file spin_image.h.

◆ setRotationAxis()

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::setRotationAxis ( const PointNT &  axis)
inline

Sets single vector a rotation axis for all input points.

It could be useful e.g. when the vertical axis is known.

Parameters
[in]axisunit-length vector that serves as rotation axis for reference frame

Definition at line 210 of file spin_image.h.

◆ setSupportAngle()

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::setSupportAngle ( double  support_angle_cos)
inline

Sets the maximum angle for the point normal to get to support region.

Parameters
[in]support_angle_cosminimal allowed cosine of the angle between the normals of input point and search surface point for the point to be retained in the support

Definition at line 166 of file spin_image.h.

◆ useNormalsAsRotationAxis()

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::SpinImageEstimation< PointInT, PointNT, PointOutT >::useNormalsAsRotationAxis ( )
inline

Sets input normals as rotation axes (default setting).

Definition at line 234 of file spin_image.h.


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