Outdoor camera calibration method using GPS coordinate based on two-stage optimization

Camera calibration is a very important and necessary step in computer vision, especially in 3D reconstruction, target tracking and positioning. In this paper, a GPS coordinate is first introduced and incorporated into Tsai two-step camera calibration method. Owing to the limitation of the accuracy of GPS receivers at present, there's a relatively large error in camera calibration parameters. A hybrid optimization algorithm (GA-PSO) by combining a genetic algorithm and particle swarm optimization is proposed further to improve camera parameters. Experimental results prove that our method is feasible and efficient.