计算机视觉
人工智能
球(数学)
计算机科学
机器视觉
摄像机切除
摄像机自动校准
雅可比矩阵与行列式
偏移量(计算机科学)
校准
光轴
视觉伺服
光学
数学
物理
图像(数学)
应用数学
程序设计语言
统计
数学分析
镜头(地质)
作者
Fangbo Qin,De Xu,Fei Shen,Dapeng Zhang,Xilong Liu
标识
DOI:10.1109/coase.2016.7743505
摘要
The calibration of the micro-vision system is an fundamental issue in the micro-vision based measurement and control areas. The calibration methods for conventional macro-vision system are not suitable for telecentric micro-vision system. This paper proposes a calibration method with ball objects for the telecentric micro-vision system, which is used for the 3-dimensional relative pose measurement. The ball objects are low-cost and easy to be made. The intrinsic parameters and depth of field of camera are calibrated with a one-ball object. The angle between the camera's optical axis and the linear stage's axis is considered, and the image offset matrix caused by this angle is calibrated using the one-ball object. Furthermore, the image Jacobian matrices of two cameras are obtained by calibrating the two cameras' rotation transformation using a two-ball object. The experimental results verify the calibration methods' effectiveness and accuracy.
科研通智能强力驱动
Strongly Powered by AbleSci AI