野外大视场双目像机外参数自标定方法  

Self-calibration method of external parameters of binocular camera with large field of view

作　　者：易见为 苏志龙 关棒磊 李璋 张东升[2] 于起峰[1] YI Jianwei;SU Zhilong;GUAN Banglei;LI Zhang;ZHANG Dongsheng;YU Qifeng(Hunan Provincial Key Laboratory of Image Measurement and Vision Navigation,College of Aerospace Science and Engineering,National University of Defense Technology,Changsha 410073,Hunan,China;School of Mechanics and Engineering,Shanghai University,Shanghai Institute of Applied Mathematics and Mechanics,Shanghai Key Laboratory of Energy Engineering Mechanics,Shanghai 200444,China)

机构地区：[1]国防科技大学空天科学学院,图像测量与视觉导航湖南省重点实验室,湖南长沙410073 [2]上海大学力学与工程学院,上海应用数学和力学研究所,上海市能源工程力学重点实验室,上海200444

出　　处：《实验力学》2022年第6期775-783,共9页Journal of Experimental Mechanics

基　　金：国家自然基金重大科研仪器研制项目(11727804);国家自然科学基金(11902349);湖南省自然科学基金(2020JJ5645)资助。

摘　　要：针对野外大视场双目像机外参数标定困难的情况,提出利用无人机携带标志作为场景点进行现场外参数自标定的方法。通过控制无人机在双目像机大视场范围内飞行,利用双目像机对无人机拍摄并跟踪,无人机的位置可作为动态场景点进行外参数标定,该方法易满足动态场景点在大视场中均匀散布的要求;同时针对图像同名点对存在野值点以及优化算法容易收敛到局部最优解的问题,采用了图割优化方法进行野值点剔除,并通过全局最优解方法进行外参数精确求解。对比验证实验表明,该方法的标定精度优于传统方法,能够有效解决野外大视场双目像机外参数标定难题。Aiming at the difficulty of external parameter calibration of field binocular camera with large field of view,a method of external parameter calibration using UAV signs as scene points is proposed.The UAV was controlled to fly in the large field of view of the binocular camera.Then the UAV was photographed and tracked by the binocular camera.The position of the UAV can be used as the control point for external parameter calibration.This method is easy to meet the requirements of uniform distribution of control points in the large field of view.There are outliers with the same name point pair of the image and the algorithm is easy to converge to the local optimal solution.In order to solve the problems,the graph cut optimization method was used to eliminate the outliers,and then the external parameters were accurately solved by the global optimal solution method.Comparative verification experiments show that the calibration accuracy of this method is better than the traditional method,and can effectively solve the problem of external parameter calibration of field large field binocular camera.

关 键 词：外参数标定 对极几何约束 随机样本一致性算法(RANSAC) 全局优化 

分 类 号：O348.1[理学—固体力学]