基于激光扫描的焊缝成形尺寸测量

摘要/Abstract

摘要：

提出了一种基于激光扫描获取焊缝成形尺寸的视觉测量方法.搭建了一套焊缝表面成形视觉成像系统并进行系统标定；实时采集投射激光条纹的焊缝视觉图像，采用中值滤波、二值化分割、骨络化和形态学图像处理方法实现了图像中线激光条纹的提取和细化；以细化的条纹图像为信息源来表征焊缝成形轮廓，并建立了焊缝宽度和焊缝长度的检测算法.实测结果表明，所提出的激光条纹辅助视觉测量方法能够实现焊缝成形尺寸的非接触自动测量，测定的焊缝宽度及焊缝长度可以满足焊缝成形表征的需求.

关键词: 线激光, 条纹视觉图像, 焊缝尺寸, 视觉测量

Abstract:

A method for visual measurement was proposed to acquire welding formation sizes based on laser line scanning. A visual system for detecting welding formation was developed and simultaneously calibrated, based on which, images of welding formation and laser stripe which were cast on the surface of weld were captured in real time. And extracting and thinning of laser stripes in the image were accomplished by means of a series of image preprocessing methods, such as median filtering, threshold segmentation, skeletonization and morphological processing. The laser stripe being made as information source, the outline of weld forming was revealed exactly, based on which, a series of algorithms for realtime detection of the width and length of weld were established. The measured results show that visual measurement method aided by laser stripe can acquire the width and length of weld automatically, and the measured width and length of weld can meet the demand of weld formation evaluation.

Key words: laser line, visual image of stripe, welding size, visual measurement

中图分类号:

TG 115.8

张鹏贤，李明，代敦三. 基于激光扫描的焊缝成形尺寸测量[J]. 上海交通大学学报（自然版）, 2014, 48(05): 636-640.

ZHANG Pengxian,LI Ming,DAI Dunsan. Measurement of Welding Formation Size Based on Laser Line Scanning

[J]. Journal of Shanghai Jiaotong University, 2014, 48(05): 636-640.

参考文献

［1］雷正龙，吕涛，陈彦斌，等. 基于扫描激光视觉传感的焊缝图像特征信息识别［J］．焊接学报，2013, 34(5): 5458.

LEI Zhenglong, L Tao, CHEN Yanbin, et al. Features extraction for weld image of scanning laser sensing ［J］. Transaction of the China Welding Institution, 2013, 34(5): 5458.

［2］Kasban H, Zahran O, Arafa H, et al. Welding defect detection from radiography images with a cepstral approach ［J］. NDT and E International, 2011, 44(2): 226231.

［3］张鹏贤，卢忠建，曹成虎，等. 基于双目立体视觉埋弧焊焊缝成形的表征［J］．焊接学报，2012, 33(2): 9396.

ZHANG Pengxian, LU Zhongjian, CAO Chenghu, et al. Characterization of submerged arc welding seam formation based on binocular stereo vision［J］. Transaction of the China Welding Institution, 2012, 33(2): 9396.

［4］伏喜斌，林三宝，杨春利，等. 基于激光视觉传感的角焊缝外形尺寸检测［J］. 焊接学报, 2008, 29(7): 4750.

FU Xibin, LIN Sanbao, YANG Chunli, et al. External dimensions detection of weld joint based on laser vision sensing ［J］. Transaction of the China Welding Institution, 2008, 29(7): 4750.

［5］黄维. 基于信度评价和结构光视觉的焊缝三维测量研究［D］. 上海：上海交通大学材料科学与工程学院，2013.

［6］敬泽，薛方正，李祖枢，等. 基于单目视觉的空间目标位置测量［J］. 传感器与微系统, 2011, 30(3):125127.

JING Ze, XUE Fangzheng, LI Zushu, et al. Position measurement of space object based on single vision ［J］. Transducer and Microsystem, 2011, 30(3): 125127.

［7］Sonka M, Hlavac V, Boyle R. Image processing, analysis and machine vision (secondedition) ［M］. USA: Brooks/Cole, Thomson Ltd, 2001.

［8］Harris C, Stephens M. A combined corner and edge detector ［C］// Processing of the Fourth Alvey Vision Conference. Manchester, UK: Plessey Research Roke Manor, 1988: 147151.

［9］王军. 基于双目立体视觉的测量技术研究［D］. 杭州: 浙江理工大学信息电子学院，2009.

[1]	经小龙, 胡源, 郭为忠. 臂载线结构激光手眼矩阵的精确标定——多坐标系转换法[J]. 上海交通大学学报, 2020, 54(6): 607-614.
[2]	田亮，罗宇，王阳. 基于遗传算法优化BP神经网络的TIG焊缝尺寸预测模型[J]. 上海交通大学学报（自然版）, 2013, 47(11): 1690-1696.
[3]	胡文1, 胡春2, 王石刚1. 大型系泊链五环长的视觉测量系统[J]. 上海交通大学学报（自然版）, 2011, 45(10): 1552-1556.
[4]	张泽, 段广仁, 孙勇. 基于对偶四元数的交会对接相对位姿测量算法[J]. 上海交通大学学报（自然版）, 2011, 45(03): 398-402.