零部件上下盖合装的手眼视觉系统设计

Hand-eye vision system design with component upper and lower covers

针对工业生产线上零部件上下盖自动合装的问题,搭建了一套基于手眼(eye-in-hand)系统的机器视觉装配实验平台.首先对视觉系统进行手眼标定,确定相机内外参数和手眼关系;然后利用视频处理方式获取进入视场内零部件的前景图像,获得其图像数据,并对数字图像进行相应的处理,即通过背景差分法获取前景图像,并进行连通域标记和区域轮廓提取来获得数字图像的特征信息;最后根据其特征用矩形拟合计算其质心坐标以及与机器人x方向偏转角度,使六轴机器人在合装零部件时能自动进行纠偏和补偿.实验结果表明,移动零部件的装配成功率为100%,能满足工业现场产品装配精度需求.

Aiming at the problem of automatic assembly of component upper and lower covers on industrial production lines, a machine vision assembly experimental platform based on the eye-in-hand system was built. Firstly, hand-eye calibration was performed on the visual system to determine the internal and external parameters of the camera and the hand-eye relationship. Then the foreground image of the components entering the field of view and its image data were obtained by using the video processing method to perform corresponding processing on the digital image. The foreground image was obtained by the background difference method. The connected region marker and the region contour were extracted to obtain the digital image feature information. Finally, according to its characteristics, the coordinates of its centroid and the deflection angle with the x-axis of the robot were calculated by means of a rectangular fitting so that the six-axis robot could automatically perform correction and compensation when the components were assembled. The experimental results show that the assembly success rate of the moving parts is 100%, which can meet the assembly precision requirements of industrial field products.