Calibration procedure for establishing an extrinsic relationship between lidar and camera sensors
In one or more embodiments, a method for calibration between a lidar sensor and a camera comprises determining translation parameters of extrinsic calibration parameters by using a location of the camera with respect to the lidar sensor. The method further comprises orienting a target chessboard such that it is aligned with axes of a lidar coordinate system. Also, the method comprises optimizing a best fit transformation between a camera coordinate system and a target chessboard coordinate system. In addition, the method comprises determining a rotation matrix using the best fit transformation between the camera coordinate system and the target chessboard coordinate system. Additionally, the method comprises extracting Euler angles of the extrinsic calibration parameters from the rotation matrix. Also, the method comprises collecting, by the lidar sensor and the camera, calibration scenes using the target chessboard. Further, the method comprises optimizing the extrinsic calibration parameters by using the calibration scenes.
1. A method for calibration between a lidar sensor and a camera, the method comprising:
determining, by at least one processor, translation parameters of extrinsic calibration parameters by using a location of the camera with respect to the lidar sensor;
orienting a target chessboard such that the target chessboard is aligned with at least two axes of a lidar coordinate system;
optimizing, by the at least one processor, a best fit transformation between a camera coordinate system and a target chessboard coordinate system;
determining, by the at least one processor, a rotation matrix using the best fit transformation between the camera coordinate system and the target chessboard coordinate system;
extracting, by the at least one processor, Euler angles of the extrinsic calibration parameters from the rotation matrix;
collecting, by the lidar sensor and the camera, calibration scenes using the target chessboard; and
optimizing, by the at least one processor, the extrinsic calibration parameters by using the calibration scenes.
2. The method of claim 1 , wherein the method further comprises displaying, on a display, a graphical user interface.
3. The method of claim 2 , wherein the method further comprises selecting, via the graphical user interface, a parent directory for data curation.
4. The method of claim 2 , wherein the method further comprises selecting, via the graphical user interface, the lidar sensor and the camera for the calibration.
5. The method of claim 2 , wherein the method further comprises loading, via the graphical user interface, camera intrinsic parameters.
6. The method of claim 1 , wherein the method further comprises generating camera intrinsic parameters.
7. The method of claim 1 , wherein the method further comprises measuring the location of the camera with respect to the lidar sensor.
8. The method of claim 1 , wherein the method further comprises imaging, by the camera, the target chessboard to generate an initial image of the target chessboard.
9. The method of claim 1 , wherein the optimizing, by the at least one processor, of the best fit transformation between a camera coordinate system and a target chessboard coordinate system is performed using a Levenberg-Marquardt optimization.
10. The method of claim 1 , wherein the collecting, by the lidar sensor and the camera, the calibration scenes comprises:
transmitting, by the lidar sensor, at least one transmit signal to the target chessboard located at N number of locations, where N is an integer, from the camera to obtain at least one lidar measurement; and
imaging, by the camera, the target chessboard to generate imaging measurements of the target chessboard for each of the N number of locations.
11. The method of claim 10 , wherein the at least one transmit signal is an optical signal or an infrared signal.
12. The method of claim 1 , wherein the optimizing, by the at least one processor, of the extrinsic calibration parameters by using the calibration scenes is performed using a differential evolution optimization.
13. The method of claim 1 , wherein the optimizing, by the at least one processor, of the extrinsic calibration parameters by using the calibration scenes comprises classifying lidar points and camera pixels as black or white.
14. The method of claim 1 , wherein the optimizing, by the at least one processor, of the extrinsic calibration parameters by using the calibration scenes comprises determining an orientation of the target chessboard.
15. The method of claim 14 , wherein the orientation of the chessboard is one of a diamond orientation or a rectangular orientation.
16. A system for calibration between a lidar sensor and a camera, the system comprising:
a target chessboard;
at least one processor to determine translation parameters of extrinsic calibration parameters by using a location of the camera with respect to the lidar sensor, to optimize a best fit transformation between a camera coordinate system and a target chessboard coordinate system, to determine a rotation matrix using the best fit transformation between the camera coordinate system and the target chessboard coordinate system, to extract Euler angles of the extrinsic calibration parameters from the rotation matrix, and to optimize the extrinsic calibration parameters by using calibration scenes collected by the lidar sensor and the camera; and
the lidar sensor and the camera to collect the calibration scenes using the target chessboard.
17. The system of claim 16 , wherein the system further comprises a display to display a graphical user interface.
18. The system of claim 17 , wherein the graphical user interface comprises a selection for a parent directory for data curation.
19. The system of claim 17 , wherein the graphical user interface comprises a selection for the lidar sensor and a selection for the camera.
20. The system of claim 17 , wherein the graphical user interface comprises a selection for loading camera intrinsic parameters.