一种立体视觉系统高精度标定方法

本文介绍使用广角镜头成像的立体视觉系统的高精度标定方法,该方法利用平面单应矩阵约束估计像机内外参数的初值,优化过程中将三维重投影误差作为评价函数,结合遗传算法完成寻优过程,以保证估计出的像机参数是全局最优,避免陷入局部极小。实验结果表明:该方法的空间定位精度与传统方法相比有很大程度的提高。

A high-presion binocular calibration method which uses wide-angle lens is introduced. It makes use of planar homogrphy constraint to estimate intrinsic and extrinsic parameter. On the basis of this initial guess, an optimization scheme is used to minimize a new cost function:3D reprojection error. All the calibration parameters are globally optimized simultaneously with genetic algorithm (GA). Experiment result shows this technique has much higher precision than traditional methods.