基于柔性立体靶标的摄像机标定

Camera Calibration Based on Flexible 3D Target

针对大视场摄像机标定中,大尺寸靶标加工困难,小尺寸靶标标定精度不高等问题,提出一种基于柔性立体靶标的摄像机标定方法。柔性立体靶标是由多个小平面靶标(又称子靶标)灵活组合的一种靶标形式,将多个小的子靶标摆放在视场周边,以其中1个子靶标坐标系为基础建立柔性立体靶标坐标系。以子靶标之间位置关系不变为约束条件,将各子靶标特征点的局部坐标统一到柔性立体靶标坐标系下,建立以重投影误差为最小的目标函数,采用非线性优化方法得到摄像机参数的最优解。仿真和实验证明采用柔性立体靶标可以达到与相同靶标区域大小的大平面靶标相当的标定结果。

The calibration of camera with wide field-of-view is a challenging problem because a target of large scale is hard to machine, while using target of small scale cannot satisfy the required accuracy. A camera calibration method based on flexible three-dimensional（3D） target is proposed to solve the calibration problem of camera with wide field- of-view. Flexible 3D target is a target consisting of several planar targets, called sub-targets, which are placed flexibly according to the range of the camera＇s view-field. The coordinate frame of the flexible 3D target is constructed based on a sub-target. By using the invariance of the relative positions between sub-targets in the flexible 3D target, the local coordinate frame of each sub-target is transformed into the coordinate frame of the flexible 3D target. By minimizing the re-projection error of feature points on all the sub-targets, the maximum likelihood estimate of the camera parameters can be obtained. Synthetic and experiments show that the accuracy of the proposed calibration method is almost the same as the calibration method using a big planar target whose area is similar to that of the flexible 3D target.