摘要
提出了一种基于结构光立体视觉的焊缝测量方法 ,并给出了一种简单的摄像机与激光器标定方法。它采用安装在机器人末端的两台摄像机和一只激光器构成测量系统 ,激光器发出的光束经平凸柱面镜变成光平面 ,在焊件上形成条纹。两台摄像机同步采集该条纹图像 ,结合机器人末端位姿与摄像机参数 ,对于条纹上的每个特征点可以获得三组空间坐标。对这三组坐标进行信息融合 ,可以更加精确的获得特征点的空间坐标。进行的V形焊缝测量与跟踪试验表明 ,该方法具有实时性好、测量精度较高、标定简单。
A method of weld seam measurement based on structured light stereovision is proposed, which employs two charge coupled device cameras and a laser emitter fitted on the end-effector of the robot. A plano-convex cylinder lens is utilized to convert the laser beam to plane, to forms a stripe on the welding works. Two frames of stripe image are captured by the cameras synchronously. Through the position and pose of the end-effector with the parameters of the cameras and the plane, which can be obtained easily by a simple calibration method, three groups of position values in Cartesian space for each point in the stripe can be derived. Then the accurate three-dimensional position of the edge points in the weld seam can be obtained from them by using information fusion algorithm. An experiment that V type weld seam is measured and tracked automatically is successfully conducted.
出处
《焊接学报》
EI
CAS
CSCD
北大核心
2004年第5期45-48,52,共5页
Transactions of The China Welding Institution
基金
国家 8 63计划项目 ( 2 0 0 2AA42 2 160 )
中国科学院自动化研究所创新基金资助项目 ( 2F0 2J0 2 )
关键词
结构光
立体视觉
视觉测量
标定
焊缝测量
Automation
Calibration
Cameras
Charge coupled devices
Lasers
Light
Measurements
Stereo vision
