摘要
针对传统单目相机标定方法精度较低的问题,提出一种基于天牛须搜索算法的单目相机标定方法;首先使用MATLAB软件的标定工具箱快速计算单目相机内外参数,然后利用天牛须搜索算法,以平均重投影误差为目标建立目标函数,对相机参数进一步优化。结果表明,基于天牛须搜索算法的相机标定方法稳定、可靠,收敛快,优于张正友平面标定法的优化最终平均误差。
Aiming at the problem of lower accuracy of the traditional monocular camera calibration method,a monocular camera calibration method based on the beetle antennae search algorithm was proposed.The calibration toolbox of MATLAB software was firstly used to quickly calculate internal and external parameters of the monocular camera.Then the beetle antennae search algorithm was applied to establish the objective function with the average reprojection error as the target,and further optimize the camera parameters.The results show that the proposed algorithm is stable and reliable with fast convergence speed,and the final average error of optimization is better than that of Zhang’s camera calibration method.
作者
王道累
孙昊
胡松
韩洋
WANG Daolei;SUN Hao;HU Song;HAN Yang(College of Energy and Mechanical Engineering,Shanghai University of Electric Power,Shanghai 200090,China;Power China Jiangxi Electric Power Engineering Co.,Ltd.,Nanchang 330000,Jiangxi,China)
出处
《济南大学学报（自然科学版）》
CAS
北大核心
2020年第6期568-574,共7页
Journal of University of Jinan(Science and Technology)
基金
国家自然科学基金项目(61502297)。
关键词
相机标定
天牛须搜索算法
内部参数
畸变系数
camera calibration
beetle antennae search algorithm
internal parameter
distortion coefficient
作者简介
第一作者:王道累(1981—),男,上海人。副教授,博士,硕士生导师,研究方向为计算机视觉、图像处理、计算机辅助设计与制造。E-mail:alfredwdl@shiep.edu.cn;通信作者:孙昊(1994—),男,山东潍坊人。硕士研究生,研究方向为图像识别、机器视觉。E-mail:sunhawk_wk@163.com。
