双目立体视觉测量系统的标定

发布时间：2018-03-22 03:33

本文选题：视觉测量　切入点：大视场　出处：《光学精密工程》2016年02期 　论文类型：期刊论文

【摘要】：考虑传统的自标定方法虽然无需场景信息即可实现摄像机标定,但是标定精度较低,故本文提出了一种新的大视场双目视觉测量系统自标定方法。该方法无需高精度标定板或者标定物,仅需利用空间中常见的平行线和垂直线建立摄像机参数与特征线间的约束方程,即可实现摄像机的内参数与旋转矩阵标定;同时利用空间中距离已知的3个空间点即可线性标定两摄像机间的平移向量。通过标定实验对本文提出的方法进行了验证。结果表明:该方法标定精度能够达到0.51%,可以较高精度地标定双目测量系统。由于避免了大视场测量系统标定中大型标定物制造困难,以及摄像机自标定过程中算法冗杂,标定精度不高等问题,该方法操作简便,精度较好,适用于大视场双目测量系统的在线标定。
[Abstract]:Considering that the traditional self-calibration method can achieve camera calibration without scene information, but the calibration accuracy is low. Therefore, a new self-calibration method for large field binocular vision measurement system is proposed in this paper, which does not require high precision calibration board or calibration object. The calibration of camera inner parameters and rotation matrix can be realized only by establishing the constraint equations between camera parameters and characteristic lines by using common parallel and vertical lines in space. At the same time, the translation vector between two cameras can be calibrated linearly by using three spatial points with known distance in space. The proposed method is verified by calibration experiments. The results show that the calibration accuracy of this method can reach the level of accuracy. The binocular measurement system can be calibrated with high accuracy. The algorithm is miscellaneous and the calibration accuracy is not high in the process of camera self-calibration. This method is easy to operate and has good precision. It is suitable for on-line calibration of large field of view binocular measurement system.
【作者单位】：大连理工大学教育部精密特种加工实验室;
【基金】：国家自然科学基金面上项目(No.51375075);国家自然科学基金仪器专项基金资助项目(No.51227004) 国家973重点基础研究发展计划资助项目(No.2014CB46504) 国家创新群体科学基金资助项目(No.51321004) 辽宁省自然科学基金资助项目(No.20140128010)
【分类号】：TP391.41

【相似文献】