Point Cloud Library (PCL)
1.14.1-dev
|
Build a Normal Distributions Transform of a 2D point cloud. More...
#include <pcl/registration/impl/ndt_2d.hpp>
Public Member Functions | |
NDT2D (PointCloudConstPtr cloud, const Eigen::Vector2f &about, const Eigen::Vector2f &extent, const Eigen::Vector2f &step) | |
ValueAndDerivatives< 3, double > | test (const PointT &transformed_pt, const double &cos_theta, const double &sin_theta) const |
Return the 'score' (denormalised likelihood) and derivatives of score of the point p given this distribution. More... | |
Protected Attributes | |
std::shared_ptr< SingleGrid > | single_grids_ [4] |
Build a Normal Distributions Transform of a 2D point cloud.
This consists of the sum of four overlapping models of the original points with normal distributions. The value and derivatives of the model at any point can be evaluated with the test (...) function.
Definition at line 311 of file ndt_2d.hpp.
|
inline |
[in] | cloud | the input point cloud |
[in] | about | Centre of the grid for normal distributions model |
[in] | extent | Extent of grid for normal distributions model |
[in] | step | Size of region that each normal distribution will model |
Definition at line 323 of file ndt_2d.hpp.
References pcl::ndt2d::NDT2D< PointT >::single_grids_.
|
inline |
Return the 'score' (denormalised likelihood) and derivatives of score of the point p given this distribution.
[in] | transformed_pt | Location to evaluate at. |
[in] | cos_theta | sin(theta) of the current rotation angle of rigid transformation: to avoid repeated evaluation |
[in] | sin_theta | cos(theta) of the current rotation angle of rigid transformation: to avoid repeated evaluation |
Definition at line 344 of file ndt_2d.hpp.
References pcl::ndt2d::NDT2D< PointT >::single_grids_, and pcl::ndt2d::ValueAndDerivatives< N, T >::Zero().
Referenced by pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::computeTransformation().
|
protected |
Definition at line 355 of file ndt_2d.hpp.
Referenced by pcl::ndt2d::NDT2D< PointT >::NDT2D(), and pcl::ndt2d::NDT2D< PointT >::test().