面向手眼系统的摄像机内参数自标定研究  

Research on Self-Calibration of Camera Intrinsic Parameters for Hand-Eye System

作　　者：朱世雄 葛动元 姚锡凡[2] 向文江 李健[1] ZHU Shixiong;GE Dongyuan;YAO Xifan;XIANG Wenjiang;LI Jian(School of Mechanical and Automotive Engineering,Guangxi University of Science and Technology,Liuzhou 545006,China;School of Mechanical and Automotive Engineering,South China University of Technology,Guangzhou 510640,China;不详)

机构地区：[1]广西科技大学机械与汽车工程学院,柳州545006 [2]华南理工大学机械与汽车工程学院,广州510640 [3]邵阳学院机械与能源工程学院,邵阳422004

出　　处：《组合机床与自动化加工技术》2023年第7期171-175,179,共6页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然科学基金项目(51765007,51675186,81960332);广西研究生教育创新计划项目(YCSW2022438)。

摘　　要：为得到机械臂手眼系统中的摄像机内参数,提出了一种基于运动约束线性求解的自标定方法。通过控制机械臂上的视觉系统进行极点延伸运动,根据图像序列之间对应特征点的关系确定摄像机主点。而后进行一组三正交平移运动,即可根据运动约束,构建关于单应性矩阵的线性约束方程组,实现摄像机内参数的线性求解。以所研究的标定结果为基础重建三维点云模型作为比较值,与已知相机参数的双目深度传感相机重建的点云模型进行对比试验,绝大多数点云相对偏差在1.5%内,有较高的标定精度。为机械臂的手眼系统标定提供了一个新的解决方案。In order to obtain the camera intrinsic parameters in the manipulator hand eye system,a self-calibration method based on linear solution of motion constraints is discussed.The basic idea is to calibrate the intrinsic parameters based on image sequences and homography matrices.By controlling the vision system on the mechanical arm to make pole extension motion and a group of three orthogonal translation motions,the linear constraint equations about the homography matrix can be constructed according to the relationship between the corresponding feature points of the image and the calibration plate plane,so as to realize the linear solution of the camera′s internal parameters.The simulation experiment shows that this method has high robustness,and the real image experiment shows that the proposed method has high accuracy,which provides a new solution for the calibration of the manipulator hand eye system.

关 键 词：自标定 手眼系统 单应矩阵 正交平移 三维重建 

分 类 号：TH165[机械工程—机械制造及自动化] TG659[金属学及工艺—金属切削加工及机床]