基于一维标定物的反射折射摄像机标定方法

Catadioptric Camera Calibration with One-Dimensional Objects

一维标定物是指一组任意两点距离已知的共线点.基于一维标定物的标定方法相对基于二维、三维标定物的方法更加灵活,在实际中有很高的应用价值.文中提出一种基于一维标定物的反射折射摄像机标定算法.如果一维标定物包含5个或5个以上的共线点,则通过一维标定物的3次或3次以上的一般刚体运动,就能够标定反射折射摄像机的参数.算法分为两步:首先,使用一维标定物的图像和主点满足的不变量计算主点;然后,通过一维标定物图像所隐含的正交消影点信息,线性地求解绝对二次曲线的像(IAC),并对IAC矩阵进行Cholesky分解确定尺度因子和畸变因子.此外,文中方法还能够给出镜面参数的解析表示以及一维标定物相对于视球中心的位置.模拟实验中,作者在两种镜面配置下比较了文中的方法和基于圆环点的方法,结果表明文中的方法在不同的镜面配置下都能得到较好的标定结果.真实实验也验证了文中方法的正确性和可行性.

A 1-D calibration object is meant a segment with several known-distance markers （or points）. Generally speaking, calibration methods with 1D objects are more flexible than those with 2D or 3D objects. Under the pinhole camera model, Zhang^[9] proved that the calibration is not possible with free-moving 1D objects, but can be done if one of the markers is fixed. In this paper, the authors propose a catadioptric camera calibration method using 1D objects with five or more known points. The method is capable of calibrating a camera when either the camera or the calibration object undertakes three or more general motions. The proposed algorithm consists of the following two steps： Firstly, the principal point is calculated with geometric invariants under catadioptric camera model; Secondly, for every image point of 1D object, a pair of orthogonal vanishing points is derived, by which the image of absolute conic （IAC） is computed, then the intrinsic parameter matrix is obtained by Cholesky factorization on the IAC. In addition, the analytical solutions of the mirror parameter and the 1D object＇s pose are also provided. In simulated experiments, the method can have good calibration even if the fitting of partial visible conic is not very accurate. In the real experiment, the correctness and feasibility of the proposed method is also confirmed.