【摘要】 相机标定是为了解决计算机视觉中2D图像特征和相应的3D特征之间的映射关系。针对精确标定的困难，提出了一种基于政治优化器(Political Optimizer, PO)的分步标定算法来分步求解相机几何参数和镜头畸变系数，采用参数标定与畸变修正相结合的方法，首先在理想状态下对相机参数进行寻优，其次在考虑畸变的情况下修正畸变系数，通过迭代的方式更新个体的位置来找到最佳的解，有效地校准了相机。实验结果表明，PO算法的平均误差为0.044 706像素，具有良好的避免局部最优的优化能力，能够准确完成相机标定任务。更多还原

【Abstract】 Camera calibration is used to solve the mapping relationship between 2D image features and corresponding 3D features in computer vision. To overcome the difficulty in accurate calibration, a Political Optimizer(PO) based multi-step calibration algorithm is proposed to solve the camera geometric parameters and lens distortion coefficient step by step. By combining parameter calibration and distortion correction, the camera parameters are optimally tuned firstly. Second, the distortion coefficient is modified to in consideration of distortion. Then individual positions are iteratively updated to get the optimal solution. The camera is effectively calibrated. According to the experimental results, the PO algorithm has the lowest average reprojection error, which is only 0.044 706 pixels. It has good optimization capabilities to avoid local optimization and can complete the camera calibration task accurately.更多还原