基于灰狼粒子群混合算法的相机标定优化方法

针对传统标定方法参数考虑较少、标定精度不高的缺点,将群智能算法思想应用其中,并在现有的粒子群算法和灰狼算法基础上进行优化,取长补短,提出一种混合算法.首先,基于张正友方法获得标定的参数范围,在该范围内随机生成参数作为初始值,然后以最小平均误差为准则,利用灰狼粒子群优化混合算法,建立目标函数,进一步求解相机的内外参数.研究结果表明,该算法求解结果准确、稳定,可重复操作,可以有效地提高标定精度,结果好于其他几种方法.

Optimization method for camera calibration based on a hybrid algorithm of particle swarm optimization and grey wolf optimization

Aiming at the shortcomings of the traditional calibration method, such as less parameters considered and low calibration precision,this paper applies the idea of swarm intelligence algorithm, and improves the existing particle swarm optimization algorithm and gray wolf algorithm, and proposes a hybrid optimization algorithm.Firstly, the calibrated parameter range is obtained based on Zhang''s camera calibration method, and parameters are randomly generated as initial values within the range. Then, the minimum average error is used as the criterion.And the hybrid particle swarm optimization algorithm is used to further optimize the camera parameters.Experimental results of the study show that the solution of this method is accurate, stable and repeatable.This method can effectively improve the accuracy of calibration results.The final average error of optimization is better than several other methods.