AruCo Markers - Revision history

S130067: /* How to use the functions from the umission */

2019-08-26T19:17:33Z

How to use the functions from the umission

← Older revision		Revision as of 21:17, 26 August 2019
Line 43:		Line 43:
	* robotTurnCommand is a string that can be send to the robot, so that it turn and is positioned parallel to a on of the markers axes		* robotTurnCommand is a string that can be send to the robot, so that it turn and is positioned parallel to a on of the markers axes
	* robotDriveCommand drives the robot the marker		* robotDriveCommand drives the robot the marker
	* ~~rob_Ar~~ is the homogeneous transformation and rotation matrix that transform a point in the marker frame to a point in the robot frame.		* rob_Ar_H is the homogeneous transformation and rotation matrix that transform a point in the marker frame to a point in the robot frame.
	* done indicate that the robotTurnCommand has been found		* done indicate that the robotTurnCommand has been found
	* markerUnitVecX, markerUnitVecY, markerUnitVecCaliperZ; is the unitvectors of the marker represented in the robot frame		* markerUnitVecX, markerUnitVecY, markerUnitVecCaliperZ; is the unitvectors of the marker represented in the robot frame

S130067: /* How to use the functions from the umission */

2019-08-26T19:16:14Z

How to use the functions from the umission

← Older revision		Revision as of 21:16, 26 August 2019
Line 43:		Line 43:
	* robotTurnCommand is a string that can be send to the robot, so that it turn and is positioned parallel to a on of the markers axes		* robotTurnCommand is a string that can be send to the robot, so that it turn and is positioned parallel to a on of the markers axes
	* robotDriveCommand drives the robot the marker		* robotDriveCommand drives the robot the marker
	* ~~rob_Ar_H30~~ is the homogeneous transformation and rotation matrix that transform a point in the marker frame to a point in the robot frame.		* rob_Ar is the homogeneous transformation and rotation matrix that transform a point in the marker frame to a point in the robot frame.
	* done indicate that the robotTurnCommand has been found		* done indicate that the robotTurnCommand has been found
	* markerUnitVecX, markerUnitVecY, markerUnitVecCaliperZ; is the unitvectors of the marker represented in the robot frame		* markerUnitVecX, markerUnitVecY, markerUnitVecCaliperZ; is the unitvectors of the marker represented in the robot frame

S130067 at 18:49, 26 August 2019

2019-08-26T18:49:58Z

← Older revision		Revision as of 20:49, 26 August 2019
Line 26:		Line 26:
	To estimate the position of the markers is a camera matrix and distance coefficients needed. The file "ILoveCameraCalibration" contains these variables. At the beginning of the program is the variable loaded from the file. The camera matrix and distance coefficients can also be obtained by running the camera calibration process. The camera calibration process changes the variables in the file.		To estimate the position of the markers is a camera matrix and distance coefficients needed. The file "ILoveCameraCalibration" contains these variables. At the beginning of the program is the variable loaded from the file. The camera matrix and distance coefficients can also be obtained by running the camera calibration process. The camera calibration process changes the variables in the file.
	It is recommended to run the camera calibration process, to be sure that the values in the file not have been obtained on another robot.		It is recommended to run the camera calibration process, to be sure that the values in the file not have been obtained on another robot.
			Go to the [[AruCo_Markers#Camera_Calibration]] section for more details on camera calibration.

	== How to use the functions from the umission ==		== How to use the functions from the umission ==

S130067 at 18:36, 26 August 2019

2019-08-26T18:36:43Z

← Older revision		Revision as of 20:36, 26 August 2019
Line 1:		Line 1:
	= ArUco functions and camera calibration overview =		= ArUco functions and camera calibration overview =
			[[File:Test6image_5b.jpg\|180px\|thumb\|right\|Block with 3 ArUco marker visible]]
	The ArUco functions in the camera class are meant as an inspiration, to give a starting point to people, who want to use the OpenCv ArUco library.		The ArUco functions in the camera class are meant as an inspiration, to give a starting point to people, who want to use the OpenCv ArUco library.

Line 33:		Line 34:
	The function takes a picture and runs through all the AruCo functions. The results are saved in an array of structs called arUcoVal. The values are directly accessible from umission. For example, can the distance to the AruCo marker 30, be accessed in the following way:		The function takes a picture and runs through all the AruCo functions. The results are saved in an array of structs called arUcoVal. The values are directly accessible from umission. For example, can the distance to the AruCo marker 30, be accessed in the following way:
	cam->arUcoVal[30].disteance2marker;		cam->arUcoVal[30].disteance2marker;

			The struct ArUcoVal contains the values related to one ArUco marker
			* markerId is the ID of the marker ectracted from the vector markerIds found by the cv::aruco::detectMarkers function
			* frame is the picture where the markers are detected.
			* rVec and tVec are also found by cv::aruco::detectMarkers.
			* label is a sting printet on the picture with rVec, tVec ect.
			* robotTurnCommand is a string that can be send to the robot, so that it turn and is positioned parallel to a on of the markers axes
			* robotDriveCommand drives the robot the marker
			* rob_Ar_H30 is the homogeneous transformation and rotation matrix that transform a point in the marker frame to a point in the robot frame.
			* done indicate that the robotTurnCommand has been found
			* markerUnitVecX, markerUnitVecY, markerUnitVecCaliperZ; is the unitvectors of the marker represented in the robot frame
			* disteance2marker id the distance from the robot to the marker in meters
			* angle2markerX and angle2markerY are the angle between the robot and the x or y axis of the marker frame respectively */

	== Rotation and Translation ==		== Rotation and Translation ==
			The function [https://docs.opencv.org/4.1.1/d9/d6a/group__aruco.html#ga84dd2e88f3e8c3255eb78e0f79571bd1 estimatePoseSingleMarkers] calculates a rotation vector.
			rVec=[a b c]
			The rotation vector is a compact notation of the Rodrigues notation. The Rodrigues notation notates a 3D rotation by rotation axis v and an angle θ. The object is rotated around v with the angle θ. The compact notation uses only 3 values to represent the rotation.
			θ = sqrt(a²+b²+c²)
			and
			v = [a/θ b/θ c/θ]
			The [https://docs.opencv.org/4.1.1/d9/d0c/group__calib3d.html#ga61585db663d9da06b68e70cfbf6a1eac Rodrigues] function from the OpenCv library is used to convert the compact Rodrigues notation to a rotation matrix.

			This rotation matrix together with the translation vector forms a homogeneous transformation and rotation matrix <sub>Ar</sub><sup>cam</sup>'''H'''. When a point <sup>Ar</sup>P in the marker coordinates system is post multiplied on this matrix <sub>Ar</sub><sup>cam</sup>'''H''', will the result be the point in 3d coordinate system of the camera. A similar transformation matrix <sub>cam</sub><sup>rob</sup>'''H''' transform coordinate from camera to robot coordinate system. Together
			<sub>Ar</sub><sup>rob</sup>'''H''' = <sub>cam</sub><sup>rob</sup>'''H''' x <sub>Ar</sub><sup>cam</sup>'''H'''
			is formed. ArRobH transform vectors from marker coordinate to robot coordinate system.

	== Camera Calibration ==		== Camera Calibration ==
	The Ucamera class also include functions to find the camera matrix, by using a checkerboard to calibrate the camera.		The Ucamera class also include functions to find the camera matrix, by using a checkerboard to calibrate the camera.
	follow the following procedure to calibrate the camera:		follow the following procedure to calibrate the camera:
	# Measure the size of a single square on the checkerboard and update chessboardDimensions. (be as accurate as possible and use a Caliper)		# Measure the size of a single square on the checkerboard and update chessboardDimensions. (Be as accurate as possible and use a Caliper)
	# Count the number of squares in each row and column and subtract 1. And update chessboardDimensions. (~~A normal board used for a game of chess has Size(7~~,7) )		# Count the number of squares in each row and column and subtract 1. And update chessboardDimensions. [[File:chessboard.png\|180px\|thumb\|right\|Checkerboard with dimensions (6,8) marked]]
	# Then uncomment the function cameraCalibrationProcess(cameraMatrix, distanceCoefficients);		# Then uncomment the function cameraCalibrationProcess(cameraMatrix, distanceCoefficients);
	# Run the program.		# Run the program.
Line 47:		Line 71:
	# Press space to save the frame (in the image window, not the terminal).		# Press space to save the frame (in the image window, not the terminal).
	# Move the checkerboard to a new position and repeat from step 5.		# Move the checkerboard to a new position and repeat from step 5.
	# When more than 15 frames have been saved the ~~enter~~ button can be pushed to calibrate. 15 is the minimum frames the code will accept, but it is advisable to take 20+. The more frames the better result.		# When more than 15 frames have been saved the Enter button can be pushed to calibrate. 15 is the minimum frames the code will accept, but it is advisable to take 20+. The more frames the better result.
	# When the terminal writes “calibrated and saved", exit the calibration by pressing ESC.		# When the terminal writes “calibrated and saved", exit the calibration by pressing ESC.
	# The cameraMatrix and the distanceCoefficients are saved to a file called ILoveCameraCalibration. The function loadCameraCalibration loads the values from the file. The function loadCameraCalibration is called by the UCamera constructor		# The cameraMatrix and the distanceCoefficients are saved to a file called ILoveCameraCalibration. The function loadCameraCalibration loads the values from the file. The function loadCameraCalibration is called by the UCamera constructor

S130067: information about different aspects with the ArUco and camera calibration functions

2019-08-26T16:13:27Z

information about different aspects with the ArUco and camera calibration functions

@@ Line 1: / Line 1: @@
-==== ArUco functions and camera calibration overview ====
+= ArUco functions and camera calibration overview =
 The ArUco functions in the camera class are meant as an inspiration, to give a starting point to people, who want to use the OpenCv ArUco library.
-=== Settings ===
+The ArUcu module from the OpenCv library is used to detect the markers. The library can be downloaded on [https://docs.opencv.org/4.1.1/d5/dae/tutorial_aruco_detection.html this] website, where there also is some information about the functions from the library.
-Some of the settings is dependable of the printed ArUcu markers and the hardware (camera) used to detect. So the first thing to do, when starting a new project is to check and if necessary update this parameter.
+The objective of the implementation of the functions from the OpenCv library is to detect markers on a 3D printed block. The block has 6 ArUco markers on it. One on each side. The markers are placed in the following way:
-=== How to use the functions from the umission ===
+The functions are only tested with this specific block, but it should also work with other markers if the last digit in their ID corresponds to those on the blocks. eg the numbers {50, 51, 53, 54, 55, 56}
-When a new detection of AcUco markers is wantet, simply call the function:
+Some of the settings are dependable of the printed ArUcu markers and the hardware (camera) used to detect. So the first thing to do, when starting a new project is to check and if necessary update this parameter.
   cam->startWebcamMonitoring();
-The function takes a picture and run trough all the AruCo functions. The results is saved in an array of structs called arUcoVal. The values are directly azzessable fra umission. For example can the distance to the AruCo marker 30, be accessed in the following way:
+The function takes a picture and runs through all the AruCo functions. The results are saved in an array of structs called arUcoVal. The values are directly accessible from umission. For example, can the distance to the AruCo marker 30, be accessed in the following way:
   cam->arUcoVal[30].disteance2marker;

S130067 at 11:59, 26 August 2019

2019-08-26T11:59:56Z

← Older revision		Revision as of 13:59, 26 August 2019
Line 11:		Line 11:
	When a new detection of AcUco markers is wantet, simply call the function:		When a new detection of AcUco markers is wantet, simply call the function:
	cam->startWebcamMonitoring();		cam->startWebcamMonitoring();

	The function takes a picture and run trough all the AruCo functions. The results is saved in an array of structs called arUcoVal. The values are directly azzessable fra umission. For example can the distance to the AruCo marker 30, be accessed in the following way:		The function takes a picture and run trough all the AruCo functions. The results is saved in an array of structs called arUcoVal. The values are directly azzessable fra umission. For example can the distance to the AruCo marker 30, be accessed in the following way:
	cam->arUcoVal[30].disteance2marker;		cam->arUcoVal[30].disteance2marker;

S130067: Created page with "==== ArUco functions and camera calibration overview ==== The ArUco functions in the camera class are meant as an inspiration, to give a starting point to people, who want to ..."

2019-08-26T11:59:23Z

Created page with "==== ArUco functions and camera calibration overview ==== The ArUco functions in the camera class are meant as an inspiration, to give a starting point to people, who want to ..."

New page

==== ArUco functions and camera calibration overview ====
The ArUco functions in the camera class are meant as an inspiration, to give a starting point to people, who want to use the OpenCv ArUco library.

=== Settings ===

Some of the settings is dependable of the printed ArUcu markers and the hardware (camera) used to detect. So the first thing to do, when starting a new project is to check and if necessary update this parameter.

=== How to use the functions from the umission ===

When a new detection of AcUco markers is wantet, simply call the function:
cam->startWebcamMonitoring();

The function takes a picture and run trough all the AruCo functions. The results is saved in an array of structs called arUcoVal. The values are directly azzessable fra umission. For example can the distance to the AruCo marker 30, be accessed in the following way:
cam->arUcoVal[30].disteance2marker;