基于三维标定板的相机标定方法  

Camera calibration method based on 3D calibration plate

作　　者：施佳豪 王庆[1] 冯悠扬 SHI Jiahao;WANG Qing;FENG Youyang(School of Instrument Science and Engineering,Southeast University,Nanjing 210096,China)

机构地区：[1]东南大学仪器科学与工程学院,江苏南京210096

出　　处：《传感器与微系统》2021年第6期48-51,共4页Transducer and Microsystem Technologies

基　　金：国家重点研发计划资助项目(2016YFB0502103);国家自然科学基金资助项目(61601123);江苏省自然科学基金资助项目(BK20160696)。

摘　　要：针对传统平面相机标定方法中棋盘角点坐标维度信息缺失的问题,提出一种基于三维标定板的相机标定方法。在平面标定的算法基础上,利用角点的三维坐标计算相机的内参系数。根据标定板角点的世界坐标和像素坐标的对应关系、奇异值分解(SVD)方法,求解相机的投影变换矩阵;再利用旋转向量的性质计算出相机内参的初始值;最后运用非线性优化方法对所有的标定参数进行整体优化。实验结果表明:相对于平面标定而言,三维相机标定方法的稳定性更好,标定结果更符合针孔相机模型的投影规律。Aiming at the problem of missing corner coordinate dimension in traditional camera calibration method, a camera calibration method based on three-dimensional(3 D) calibration plate is proposed.Based on the algorithm of plane calibration, the three-dimensional coordinates of the corner points are used to calculate the internal parameter of the camera.According to the correspondence between the world coordinates and pixel coordinates of the calibration plate corner points and the singular value decomposition(SVD) method, the projection transformation matrix of the camera is solved, and the initial value of the camera internal parameters is calculated by the properties of the rotation vector.Finally, the nonlinear optimization method is applied to optimize all the calibration parameters.The experimental results show that the stability of the 3 D camera calibration method is better than the plane calibration, and the calibration parameters are more in line with the pinhole camera model.

关 键 词：针孔相机 标定 奇异值分解 非线性优化 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]