一种基于OpenCV和Tsai的摄像机两步标定新方法

为实现成形加工系统中摄像机标定技术,对开放计算机视觉函数库OpenCV进行了分析。在Tsai方法基础上,结合传统摄像机标定原理,充分考虑了透镜的径向畸变和切向畸变影响及求解算法,给出了一种适应于成形加工的MVS摄像机标定新方法。实验结果表明,该算法不仅具有良好的跨平台移植性,而且提高了标定精度和计算效率,能高效地获得相应的标定参数,可以满足一般成形加工中机器视觉系统的需要。

A new method of camera calibration based on OpenCV and Tsai

In order to complete the camera calibration in forming processing,the OpenCV(open source computer vision library)was discussed.Based on algorithm of Tsai and principle of traditional camera calibration,a new algorithm called MVS camera calibration method was presented,in which the influences and solving methods of lens radial and tangential distortion was well considered.The result shows that the method has a good property for the application to multi-platform,and is better than the other two methods in precision and efficiency of calibrating the camera extrinsic and intrinsic parameters.It can meet the needs of application for machine vision to traditional forming processing areas.