but_plane_detector::Normals Class Reference

#include <normals.h>

Public Member Functions
cv::Vec4f	getNormal (int i, int j, int step, float depthThreshold)
cv::Vec4f	getNormalLSQ (int i, int j, int step, float depthThreshold)
cv::Vec4f	getNormalLSQAround (int i, int j, int step, float depthThreshold)
cv::Vec4f	getNormalLTS (int i, int j, int step, float depthThreshold, float outlierThreshold, int maxIter)
cv::Vec4f	getNormalLTSAround (int i, int j, int step, float depthThreshold, float outlierThreshold, int maxIter)
	Normals (pcl::PointCloud< pcl::PointXYZ > &pointcloud, int normalType=NormalType::PCL, int neighborhood=4, float threshold=0.2, float outlierThreshold=0.02, int iter=3)
	Normals (cv::Mat &points, const sensor_msgs::CameraInfoConstPtr &cam_info, int normalType=NormalType::PCL, int neighborhood=4, float threshold=0.2, float outlierThreshold=0.02, int iter=3)
Static Public Member Functions
static Plane< float >	LeastSquaresPlane (std::vector< cv::Vec3f > &points)
Public Attributes
sensor_msgs::CameraInfoConstPtr	m_cam_info
cv::Mat	m_planes
cv::Mat	m_points
int	m_quantbins
std::vector< cv::Vec3f >	m_quantVectors
Private Member Functions
unsigned int	getQuantVector (std::vector< cv::Vec3f > &vec, cv::Vec3f &vector)
void	initQuantVectors (int n_bins, std::vector< cv::Vec3f > &vec)
void	nextStep (int step, int &x, int &y, int &plusX, int &plusY)

Detailed Description

Class encapsulating normal computation from depth image

Definition at line 300 of file normals.h.

Constructor & Destructor Documentation

but_plane_detector::Normals::Normals	(	cv::Mat &	points,
		const sensor_msgs::CameraInfoConstPtr &	cam_info,
		int	normalType = `NormalType::PCL`,
		int	neighborhood = `4`,
		float	threshold = `0.2`,
		float	outlierThreshold = `0.02`,
		int	iter = `3`
	)

Constructor - computes real point positions (in scene coordinates) and normals and initiates all variables

Parameters:

	points	Input CV_16UC depth matrix (raw input from kinect)
	cam_info	Camera info message (ROS)
	normalType	Type of normal computation method (NormalType enum)

See also:: NormalType()

Parameters:

	neighborhood	Neighborhood from which normals are computed
	threshold	Threshold for depth difference outlier marking (if depth of neighbor is greater than this threshold, point is skipped)
	outlierThreshold	Outlier threshold for least trimmed squares regression (max error between point depth and proposed plane)
	iter	Maximum RANSAC iterations

Definition at line 71 of file normals.cpp.

but_plane_detector::Normals::Normals	(	pcl::PointCloud< pcl::PointXYZ > &	pointcloud,
		int	normalType = `NormalType::PCL`,
		int	neighborhood = `4`,
		float	threshold = `0.2`,
		float	outlierThreshold = `0.02`,
		int	iter = `3`
	)

Constructor - computes real point positions (in scene coordinates) and normals and initiates all variables

Parameters:

	pointcloud	Point cloud
	threshold	Threshold for depth difference outlier marking (if depth of neighbor is greater than this threshold, point is skipped)
	neighborhood	Neighborhood from which normals are computed

Definition at line 282 of file normals.cpp.

Member Function Documentation

cv::Vec4f but_plane_detector::Normals::getNormal	(	int	i,
		int	j,
		int	step,
		float	depthThreshold
	)

Function computes normal for point (i, j) using direct computation (mean of surrounding triangles)

Parameters:

	i	row index
	j	column index
	step	Neighborhood to compute with
	depthThreshold	Threshold for depth difference outlier marking (if depth of neighbor is greater than this threshold, point is skipped)

Definition at line 444 of file normals.cpp.

cv::Vec4f but_plane_detector::Normals::getNormalLSQ	(	int	i,
		int	j,
		int	step,
		float	depthThreshold
	)

Function computes normal for point (i, j) using least squares regression

Parameters:

	i	row index
	j	column index
	step	Neighborhood to compute with
	depthThreshold	Threshold for depth difference outlier marking (if depth of neighbor is greater than this threshold, point is skipped)

Definition at line 535 of file normals.cpp.

cv::Vec4f but_plane_detector::Normals::getNormalLSQAround	(	int	i,
		int	j,
		int	step,
		float	depthThreshold
	)

Function computes normal for point (i, j) using least squares regression with using only outer neighborhood ring

Parameters:

	i	row index
	j	column index
	step	Neighborhood to compute with
	depthThreshold	Threshold for depth difference outlier marking (if depth of neighbor is greater than this threshold, point is skipped)

Definition at line 578 of file normals.cpp.

cv::Vec4f but_plane_detector::Normals::getNormalLTS	(	int	i,
		int	j,
		int	step,
		float	depthThreshold,
		float	outlierThreshold,
		int	maxIter
	)

Function computes normal for point (i, j) using least trimmed squares regression

Parameters:

	i	row index
	j	column index
	step	Neighborhood to compute with
	depthThreshold	Threshold for depth difference outlier marking (if depth of neighbor is greater than this threshold, point is skipped)
	outlierThreshold	Threshold for marking outliers in RANSAC (maximum difference from proposed plane)
	maxIter	Maximum RANSAC iterations

Definition at line 629 of file normals.cpp.

cv::Vec4f but_plane_detector::Normals::getNormalLTSAround	(	int	i,
		int	j,
		int	step,
		float	depthThreshold,
		float	outlierThreshold,
		int	maxIter
	)

Function computes normal for point (i, j) using least trimmed squares regression with using only outer neighborhood ring

Parameters:

	i	row index
	j	column index
	step	Neighborhood to compute with
	depthThreshold	Threshold for depth difference outlier marking (if depth of neighbor is greater than this threshold, point is skipped)
	outlierThreshold	Threshold for marking outliers in RANSAC (maximum difference from proposed plane)
	maxIter	Maximum RANSAC iterations

Definition at line 699 of file normals.cpp.

unsigned int but_plane_detector::Normals::getQuantVector	(	std::vector< cv::Vec3f > &	vec,
		cv::Vec3f &	vector
	)			`[private]`

Helper function which returns bin index for given vector (not used, TODO)

Parameters:

	vec	Vector of computed quantization vectors
	vector	Vector whos bin we are computing

Definition at line 905 of file normals.cpp.

void but_plane_detector::Normals::initQuantVectors	(	int	n_bins,
		std::vector< cv::Vec3f > &	vec
	)			`[private]`

Helper function which initializes quantization vectors (not used, TODO)

Parameters:

	n_bins	Number of quantization bins
	vec	Vector of computed quantization vectors

Definition at line 885 of file normals.cpp.

Plane< float > but_plane_detector::Normals::LeastSquaresPlane ( std::vector< cv::Vec3f > & points ) [static]

Function computes plane using least squares regression from given point set

Parameters:

points

Vector of Vec3f points

Returns:: Plane object

See also:: Plane()

Definition at line 777 of file normals.cpp.

void but_plane_detector::Normals::nextStep	(	int	step,
		int &	x,
		int &	y,
		int &	plusX,
		int &	plusY
	)			`[private]`

Helper function for "Around" functions - sets next point on outer ring

Parameters:

	step	Maximum distance from center (neighborhood)
	x	Current x offset
	y	Current y offset
	plusX	Next x shift
	plusY	Next y shift

Definition at line 854 of file normals.cpp.

Member Data Documentation

sensor_msgs::CameraInfoConstPtr but_plane_detector::Normals::m_cam_info

Saved camera info (ROS)

Definition at line 397 of file normals.h.

cv::Mat but_plane_detector::Normals::m_planes

Computed plane equation approximations (so the normals also) for each point

Definition at line 392 of file normals.h.

cv::Mat but_plane_detector::Normals::m_points

Computed real points - scene coordinates (in meters)

Definition at line 387 of file normals.h.

int but_plane_detector::Normals::m_quantbins

Helper if vector quantization is on - not used in this version (TODO)

Definition at line 407 of file normals.h.

std::vector<cv::Vec3f> but_plane_detector::Normals::m_quantVectors

Helper if vector quantization is on - not used in this version (TODO)

Definition at line 402 of file normals.h.

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

but_plane_detector::Normals Class Reference

Public Member Functions

Static Public Member Functions

Public Attributes

Private Member Functions

Detailed Description

Constructor & Destructor Documentation

Member Function Documentation

Member Data Documentation