弧焊机器人中焊缝的识别处理  

Weld seam recognition and processing used by arc welding robot

作　　者：孙俊伟[1] 孙育伟[1] 李建仁[1] 柴玉东[1] 张磊磊 

机构地区：[1]兰州理工大学,甘肃兰州730050

出　　处：《电焊机》2009年第4期104-105,124,共3页Electric Welding Machine

摘　　要：图像的处理与识别在焊接机器人的视觉系统中起着关键的作用,在本试验中通过摄像头拍摄焊缝,获取原始图像,然后通过图像采集卡将拍摄的图片采集到计算机内部,对原始图进行灰度处理、滤波、阈值处理,提取出焊缝的初始轮廓,由计算系统进行处理,提取出轮廓特征,然后转换为计算机容易识别的数据,机器人焊接系统根据这些有用数据对目标对象进行控制、监测或其他操作,以实现焊接生产智能化、提高焊缝质量和焊接生产率。提供给弧焊机器人的视觉处理系统的这些数据,可以提高弧焊机器人的焊接精度,实现自动纠偏,并对焊接系统提供可靠的数据。Image processing and recognition in the welding robot vision system plays a key role in this experiment through the weld camera to obtain the original image,and then through the image acquisition card will be collecting pictures to the computer inside ,the original image gray handle,fiher,threshold treatment,extracted weld initial contour,by the computer system to carry out process, extracted outside features,then converted easily identifiable data to computer,robot welding system useful data on the basis of these target for control, monitor or other operations, in order to realize intelligent welding production, improve productivity ,weld quality and welding efficiency.Provided to the arc welding robot visual system deal with these data,we can improve the accuracy of arc welding robot, automatic correction of the welding system to provide reliable data.

关 键 词：焊缝 图像识别 视觉系统 

分 类 号：TG409[金属学及工艺—焊接]