TY - GEN
T1 - A camera self-calibration method based on IOS-PSO
AU - Xu, Jianping
AU - Deng, Fang
N1 - Publisher Copyright:
© 2015 IEEE.
PY - 2016/1/13
Y1 - 2016/1/13
N2 - When reconstructing the 3D coordinates of an object in the real world, the necessary step is camera calibration. However, traditional methods cannot achieve online calibration , and active vision methods can not realize specific camera motion. In this paper, self-calibration based on Kruppa equations is adopted for camera calibration and IOS-PSO is presented for the optimization of the cost function. Although it is quick for PSO to converge to the globally optimal solution, heavily numerical oscillation occurs when there are high-dimensional independent variables. For more accurate results, the searching intervals are respectively divided into four small intervals and the searching results follow the defined rules. The experiment results show that the method achieves more accurate camera calibration results and high convergence rate.
AB - When reconstructing the 3D coordinates of an object in the real world, the necessary step is camera calibration. However, traditional methods cannot achieve online calibration , and active vision methods can not realize specific camera motion. In this paper, self-calibration based on Kruppa equations is adopted for camera calibration and IOS-PSO is presented for the optimization of the cost function. Although it is quick for PSO to converge to the globally optimal solution, heavily numerical oscillation occurs when there are high-dimensional independent variables. For more accurate results, the searching intervals are respectively divided into four small intervals and the searching results follow the defined rules. The experiment results show that the method achieves more accurate camera calibration results and high convergence rate.
KW - 3D coordinates
KW - IOS-PSO
KW - Kruppa equations
KW - camera self-calibration
UR - http://www.scopus.com/inward/record.url?scp=84966670309&partnerID=8YFLogxK
U2 - 10.1109/CAC.2015.7382550
DO - 10.1109/CAC.2015.7382550
M3 - Conference contribution
AN - SCOPUS:84966670309
T3 - Proceedings - 2015 Chinese Automation Congress, CAC 2015
SP - 489
EP - 494
BT - Proceedings - 2015 Chinese Automation Congress, CAC 2015
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - Chinese Automation Congress, CAC 2015
Y2 - 27 November 2015 through 29 November 2015
ER -