基于改进粒子群优化算法的摄像机标定

本文在充分研究多种摄像机标定方法和粒子群优化算法（PSO）的基础上，针对传统PSO算法存在早熟和局部收敛的问题，提出了一种改进的新型粒子群优化算法在摄像机标定中的应用。该算法在基本PSO惯性权重部分加入了收缩因子，很好的改善了算法的收敛性；为了进一步提高优化的速度和可靠性，引入了多适应值函数策略。最后在OpenCV上实现了基于该改进方法的摄相机标定。实验结果表明：该摄像机标定方法有效提高了原有张正友平面标定法的标定精度，结果稳定可靠，具有一定的工程应用价值。

A method of camera calibration based on the improved particle swarm optimization algorithm

Adequately studied a variety of camera calibration method and particle swarm optimization (PSO) algorithm, against the traditional PSO algorithm has the problem of precocity and local convergence, propose a new and improved particle swarm optimization algorithm in camera calibration. This algorithm adds a shrinkage factor in Inertia weight part of the basic PSO to improve the convergence of the algorithm; the introduction of multi-fitness function mechanism in order to further improve the speed and reliability of the optimization. Finally, realized the camera calibration in OpenCV based on the new algorithm and the original Zhengyou method. Experimental results show that: the proposed camera calibration method is effectively improved the calibration accuracy of the original Zhengyou calibration method, the results are stable and reliable, so it has a certain value in engineering.