摘要
针对激光拼焊焊缝宽度测量中,仅依据结构光单一特征进行检测鲁棒性不强的问题,提出了基于结构光光纹信息、焊缝纹理信息、图像灰度信息、图像序列信息等多视觉特征信息进行焊缝宽度测量的方法。先依据图像灰度信息确定感兴趣区域,再针对正面焊缝图像,通过骨架法提取结构光光纹中心线,计算中心线的畸变位置识别焊缝特征点;针对背面焊缝图像,采用纹理能量模板对焊缝图像滤波确定焊缝区,然后计算焊缝区和结构光条纹的交集,识别焊缝特征点;识别的特征点通过序列图像信息进行校正,最后依据特征点计算焊缝宽度。搭建了激光拼焊焊缝宽度测量系统,并分别对焊缝正面和背面的宽度进行了测量试验。试验结果表明,该方法精度高,可靠性强,实时性好,有助于实现激光拼焊高速、高精度的焊缝质量在线检测。
Weld width is one of the important evaluation indexes of weld quality. To improve the weaker robust ability in the existing method which only based on laser stripe deformation for weld width of weld shape, a new method was proposed based on combination of laser stripe, weld texture, image grey values and image series in the paper. Firstly, the region of interest of the image was based on the image grey values. Secondly, for front weld, the centre line of the laser stripe was extracted based on skeletonization and the feature points were identified according to the deformation of the centre line. For back weld, the weld region was extracted based on texture energy filter and the feature points were identified according to the intersection of the weld region and laser stripe. And then the feature points were corrected by the information of image series. Lastly, the weld width was calculated according to the features points and laser stripe. A quality inspection system of weld shape for tailored blank laser welding was built and the weld width measurement experiments of front weld and back weld were carried out respectively. The results showed that the new method had higher accuracy, better robust ability and good real-time performance which was helpful to realize high speed and high precision weld quality inspection.
出处
《焊接技术》
2017年第1期64-69,共6页
Welding Technology
基金
国家自然科学基金项目(51405481)
辽宁省教育厅科研项目(LJZ2016016)
江苏省产学研合作项目(BY2015063-01)
沈阳建筑大学学科涵育计划(XKHY2-32)
关键词
焊缝宽度
质量检测
激光拼焊
多视觉特征信息
weld width
quality inspection
tailored blank laser welding
multi-vision feature
