Point Cloud Library (PCL)  1.7.0
Public Types | Public Member Functions
pcl::DifferenceOfNormalsEstimation< PointInT, PointNT, PointOutT > Class Template Reference

A Difference of Normals (DoN) scale filter implementation for point cloud data. More...

#include <pcl/features/don.h>

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

List of all members.

Public Types

typedef boost::shared_ptr
< DifferenceOfNormalsEstimation
< PointInT, PointNT, PointOutT > > 
Ptr
typedef boost::shared_ptr
< const
DifferenceOfNormalsEstimation
< PointInT, PointNT, PointOutT > > 
ConstPtr

Public Member Functions

 DifferenceOfNormalsEstimation ()
 Creates a new Difference of Normals filter.
virtual ~DifferenceOfNormalsEstimation ()
void setNormalScaleSmall (const PointCloudNConstPtr &normals)
 Set the normals calculated using a smaller search radius (scale) for the DoN operator.
void setNormalScaleLarge (const PointCloudNConstPtr &normals)
 Set the normals calculated using a larger search radius (scale) for the DoN operator.
virtual void computeFeature (PointCloudOut &output)
 Computes the DoN vector for each point in the input point cloud and outputs the vector cloud to the given output.
virtual bool initCompute ()
 Initialize for computation of features.

Detailed Description

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

A Difference of Normals (DoN) scale filter implementation for point cloud data.

For each point in the point cloud two normals estimated with a differing search radius (sigma_s, sigma_l) are subtracted, the difference of these normals provides a scale-based feature which can be further used to filter the point cloud, somewhat like the Difference of Guassians in image processing, but instead on surfaces. Best results are had when the two search radii are related as sigma_l=10*sigma_s, the octaves between the two search radii can be though of as a filter bandwidth. For appropriate values and thresholds it can be used for surface edge extraction.

Attention:
The input normals given by setInputNormalsSmall and setInputNormalsLarge 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.
Note:
For more information please see Yani Ioannou. Automatic Urban Modelling using Mobile Urban LIDAR Data. Thesis (Master, Computing), Queen's University, March, 2010.
Author:
Yani Ioannou.

Definition at line 68 of file don.h.


Member Typedef Documentation

template<typename PointInT , typename PointNT , typename PointOutT >
typedef boost::shared_ptr<const DifferenceOfNormalsEstimation<PointInT, PointNT, PointOutT> > pcl::DifferenceOfNormalsEstimation< PointInT, PointNT, PointOutT >::ConstPtr

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

Definition at line 79 of file don.h.

template<typename PointInT , typename PointNT , typename PointOutT >
typedef boost::shared_ptr<DifferenceOfNormalsEstimation<PointInT, PointNT, PointOutT> > pcl::DifferenceOfNormalsEstimation< PointInT, PointNT, PointOutT >::Ptr

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

Definition at line 78 of file don.h.


Constructor & Destructor Documentation

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

Creates a new Difference of Normals filter.

Definition at line 84 of file don.h.

References pcl::Feature< PointInT, PointOutT >::feature_name_.

template<typename PointInT , typename PointNT , typename PointOutT >
virtual pcl::DifferenceOfNormalsEstimation< PointInT, PointNT, PointOutT >::~DifferenceOfNormalsEstimation ( ) [inline, virtual]

Definition at line 89 of file don.h.


Member Function Documentation

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::DifferenceOfNormalsEstimation< PointInT, PointNT, PointOutT >::computeFeature ( PointCloudOut output) [virtual]

Computes the DoN vector for each point in the input point cloud and outputs the vector cloud to the given output.

Parameters:
outputthe cloud to output the DoN vector cloud to.

Implements pcl::Feature< PointInT, PointOutT >.

Definition at line 85 of file don.hpp.

References pcl::PointCloud< PointT >::points.

template<typename PointInT , typename PointNT , typename PointOutT >
bool pcl::DifferenceOfNormalsEstimation< PointInT, PointNT, PointOutT >::initCompute ( ) [virtual]

Initialize for computation of features.

Returns:
true if parameters (input normals, input) are sufficient to perform computation.

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

Definition at line 44 of file don.hpp.

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::DifferenceOfNormalsEstimation< PointInT, PointNT, PointOutT >::setNormalScaleLarge ( const PointCloudNConstPtr &  normals) [inline]

Set the normals calculated using a larger search radius (scale) for the DoN operator.

Parameters:
normalsthe larger radius (scale) of the DoN filter.

Definition at line 109 of file don.h.

template<typename PointInT , typename PointNT , typename PointOutT >
void pcl::DifferenceOfNormalsEstimation< PointInT, PointNT, PointOutT >::setNormalScaleSmall ( const PointCloudNConstPtr &  normals) [inline]

Set the normals calculated using a smaller search radius (scale) for the DoN operator.

Parameters:
normalsthe smaller radius (scale) of the DoN filter.

Definition at line 99 of file don.h.


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