Point Cloud Library (PCL)
1.9.1
|
37 #ifndef PCL_HARRIS_KEYPOINT_6D_H_
38 #define PCL_HARRIS_KEYPOINT_6D_H_
40 #include <pcl/keypoints/keypoint.h>
49 template <
typename Po
intInT,
typename Po
intOutT,
typename NormalT = pcl::Normal>
53 typedef boost::shared_ptr<HarrisKeypoint6D<PointInT, PointOutT, NormalT> >
Ptr;
54 typedef boost::shared_ptr<const HarrisKeypoint6D<PointInT, PointOutT, NormalT> >
ConstPtr;
77 : threshold_ (threshold)
84 name_ =
"HarrisKeypoint6D";
135 unsigned int threads_;
136 boost::shared_ptr<pcl::PointCloud<NormalT> > normals_;
137 boost::shared_ptr<pcl::PointCloud<pcl::IntensityGradient> > intensity_gradients_;
141 #include <pcl/keypoints/impl/harris_6d.hpp>
143 #endif // #ifndef PCL_HARRIS_KEYPOINT_6D_H_
void setRadius(float radius)
set the radius for normal estimation and non maxima supression.
This file defines compatibility wrappers for low level I/O functions.
A point structure representing normal coordinates and the surface curvature estimate.
boost::shared_ptr< HarrisKeypoint6D< PointInT, PointOutT, NormalT > > Ptr
boost::shared_ptr< const HarrisKeypoint6D< PointInT, PointOutT, NormalT > > ConstPtr
void setNumberOfThreads(unsigned int nr_threads=0)
Initialize the scheduler and set the number of threads to use.
void calculateCombinedCovar(const std::vector< int > &neighbors, float *coefficients) const
void setRefine(bool do_refine)
whether the detected key points should be refined or not.
PointCloudIn::ConstPtr PointCloudInConstPtr
void responseTomasi(PointCloudOut &output) const
Keypoint< PointInT, PointOutT >::PointCloudIn PointCloudIn
virtual void setSearchSurface(const PointCloudInConstPtr &cloud)
virtual ~HarrisKeypoint6D()
Empty destructor.
Keypoint detector for detecting corners in 3D (XYZ), 2D (intensity) AND mixed versions of these.
Keypoint< PointInT, PointOutT >::KdTree KdTree
std::string name_
The key point detection method's name.
HarrisKeypoint6D(float radius=0.01, float threshold=0.0)
Constructor.
boost::shared_ptr< const PointCloud< PointInT > > ConstPtr
void refineCorners(PointCloudOut &corners) const
void setNonMaxSupression(bool=false)
whether non maxima suppression should be applied or the response for each point should be returned
double search_radius_
The nearest neighbors search radius for each point.
Keypoint represents the base class for key points.
Keypoint< PointInT, PointOutT >::PointCloudOut PointCloudOut
PointCloudInConstPtr surface_
An input point cloud describing the surface that is to be used for nearest neighbors estimation.
void detectKeypoints(PointCloudOut &output)
Abstract key point detection method.
void setThreshold(float threshold)
set the threshold value for detecting corners.
A point structure representing the intensity gradient of an XYZI point cloud.