GT2004ImageProcessor Class Reference

The lines image processor recognizes characteristic lines in the image. More...

#include <GT2004ImageProcessor.h>

Inheritance diagram for GT2004ImageProcessor:

[legend]Collaboration diagram for GT2004ImageProcessor:

[legend]List of all members.


Helpers for grid drawing
void	plot (const unsigned char *p, Drawings::Color color)
const unsigned char *	last
Drawings::Color	lineColor
Public Member Functions
	GT2004ImageProcessor (const ImageProcessorInterfaces &interfaces)
	Constructor.
virtual void	execute ()
	Executes the module.
virtual bool	handleMessage (InMessage &message)
	Handles an incoming message.
Private Member Functions
void	scanColumns ()
	The function scans columns for line points.
void	scanRows ()
	The function scans rows for line points.
void	scan (const Vector2< int > &start, const Vector2< int > &end, bool vertical, bool noLines)
	The function scans a line for line points.
void	clusterRobots (const unsigned char *bottomPoint, bool redFound, bool blueFound)
	The function clusters points of red and blue robots.
void	filterPercepts ()
	The function filters the percepts, i.e.
void	filterLinesPercept (LinesPercept &percept, int type, const CameraMatrix &cameraMatrix, const CameraMatrix &prevCameraMatrix, const Image &image)
	The function filters the line-percepts, i.e.
double	calcEdgeAngle (const Vector2< int > &pointInImage, const Vector2< int > &pointOnField, double scanAngle, const RingBuffer< const unsigned char *, 7 > &pixelBuffer, const bool againstScanline=false, int channel=0) const
	The function calculates the angle of an edge at an edge point.
Vector2< int >	getCoords (const unsigned char *p) const
	The function converts an address to pixel coordinates.
	DECLARE_DEBUG_IMAGE (imageProcessorPlayers)
	DECLARE_DEBUG_IMAGE (imageProcessorGeneral)
	DECLARE_DEBUG_COLOR_CLASS_IMAGE (segmentedImage1)
	DECLARE_DEBUG_IMAGE (imageProcessorBall)
	DECLARE_DEBUG_IMAGE (imageProcessorGradients)
Private Attributes
double	xFactor
	Factor to convert the pixel coordinate space to the anglular coordinate space.
double	yFactor
	Factor to convert the pixel coordinate space to the anglular coordinate space.
int	yThreshold
	Brightness increase threshold.
int	vThreshold
	Brightness decrease threshold.
int	orangeCount
	Number of columns with ball points.
int	noOrangeCount
	Number of columns without a ball point.
int	noRedCount
	Number of columns without a red robot point.
int	noBlueCount
	Number of columns without a blue robot point.
int	noGoalCount
	Number of columns without a opponent goal seen.
int	closestBottom
	Closest bottom point on the grid.
Vector2< int >	firstRed
	First red robot point in a cluster.
Vector2< int >	closestRed
	Closest red robot point in a cluster.
Vector2< int >	lastRed
	Last red robot point in a cluster.
Vector2< int >	firstBlue
	First blue robot point in a cluster.
Vector2< int >	closestBlue
	Closest blue robot point in a cluster.
Vector2< int >	lastBlue
	Last blue robot point in a cluster.
Vector2< int >	firstFlag
	First flag point in a cluster.
Vector2< int >	lastFlag
	Last flag point in a cluster.
bool	goalAtBorder
	Is the first goal point at the image border?
int	longestBallRun
Vector2< int >	ballCandidate
CameraMatrix	cmTricot
	Camera matrix without tricot height.
CameraMatrix	prevCameraMatrix
	The camera matrix of the previous image.
CameraMatrix	prevCmTricot
	The tricot matrix of the previous image.
ColorCorrector	colorCorrector
	The color correction tool.
GT2004BeaconDetector	beaconDetector
	The beacon detector.
GT2004GoalRecognizer	goalRecognizer
	The goal recognizer.
GT2004BallSpecialist	ballSpecialist
	The ball specialist.
GT2004EdgeSpecialist	edgeSpecialist
	The edge specialist.
ImageInfo	imageInfo
	Additional information about the current image.
double	angleBetweenDirectionOfViewAndGround
int	numberOfScannedPixels
Matrix2x2< double >	rotation2x2

Detailed Description

The lines image processor recognizes characteristic lines in the image.

Four types of lines are distinguished: edges between the skyblue goal and the field, edges between the yellow goal and the field, edges between the border and the field, and edges between the field lines and the field.

The module scans vertical and horizontal lines in the image from top to bottom and from left to right. As the green of the field is very dark, all edges are characterized by a big difference of the y-channel of adjacent pixels. An increase in the y-channel followed by a decrease is an indication for an edge.

The projection of the pixels on the field plane is used to determine their relative position to the robot.

Author:: Matthias Juengel
Thomas Röfer

Definition at line 48 of file GT2004ImageProcessor.h.