基于编码结构光的角焊缝视觉检测
Fillet weld visual inspection based on coded structured light
-
摘要: 针对目前角焊缝外形检测方法中存在的缺点,提出了一种基于编码结构光的角焊缝外形尺寸检测方法,采用安装在机器人手臂末端的投影结构光视觉传感机构对焊缝进行测量.投影仪在测量位置向焊缝投射一幅伪随机编码彩色条纹图像与一幅用于彩色补偿的空白图像,通过彩色补偿消除焊缝表面彩色纹理信息对结构光解码的干扰,将补偿后的变形条纹图像恢复出焊缝表面三维数据,并通过几何关系计算出角焊缝的焊缝宽度、焊脚长度等尺寸参数.结果表明,基于编码结构光的角焊缝视觉检测方法能够实现角焊缝外形尺寸的非接触式高速测量,并且测量稳定性好,测量可靠度高,在焊缝视觉检测领域将会成为一种十分有效的方法.Abstract: In order to overcome the limitations of fillet weld visual inspection,an automatic post-weld shape dimension approach based on coded structured light was proposed.A structured light vision system is mounted on the end-effector of the robot and it acquires dimensional data of the target object.The LCD projector firstly projects a pseudo random color stripes image onto the workpiece and then projects a blank image for color compensation,which reduces the interference of color texture on structured light decoding.After color compensation,point-cloud data of tested workpiece is reconstructed and finally the weld shape information is obtained by means of geometrical computation.The experimental results show that the proposed method is accurate and reliable for fillet weld non-contact shape inspection.