摘要
针对超声衍射时差法(TOFD)检测图像的近表面缺陷衍射波易与直通波混叠的特点,提出一种对接焊缝近表面缺陷自动化识别方法。为了准确定位缺陷位置,开展图像处理相关技术的研究,完成图像消噪和直通波矫直处理,在此基础上利用灰度分布统计法消除与近表面缺陷波重叠的直通波,并利用图像分割算法提取出超声TOFD检测图像中的焊缝缺陷,从而实现对焊缝缺陷的自动化识别。为验证所提方法的可靠性,开展了相关的实验研究。实验结果表明,所提方法能够有效地解决超声TOFD检测近表面盲区问题,能够提取出检测图像中与直通波混叠的近表面缺陷。
Based on the characteristics that the near-surface defect diffraction wave and the lateral wave are easy to be superimposed in the TOFD testing image,an automatic identification method for near-surface defect of the weld is proposed.In order to accurately locate the defect depth,the related technology of image processing is carried out,and the image denoising and straightening are completed,on this basis,the gray distribution statistical method is used to eliminate the direct wave overlapping with the near-surface defect wave,and the image segmentation algorithm is used to extract the ultrasonic TOFD to detect the weld defects in the image,so as to realize the automatic recognition of the near-surface defect of the weld.In order to verify the reliability of the proposed method,related experimental studies were carried out.The experimental results showed that the proposed method can effectively solve the problem of near-surface blind area detection using ultrasonic TOFD,and can extract the nearsurface defects in the detected image that are superimposed with lateral waves.
作者
周红明
张小杰
ZHOU Hongming;ZHANG Xiaojie(l.Faculty of Engineering,Lishui University,Lishui 323000,Zhejiang;Zhejiang Chendiao Machinery Co.,Ltd,Lishui 323000,Zhejiang)
出处
《丽水学院学报》
2020年第5期83-89,共7页
Journal of Lishui University
基金
浙江省科技厅公益技术研究项目“复杂载荷下海底油气管道剩余强度分析方法研究”(2016C37055)。
关键词
超声TOFD
焊缝
近表面缺陷
检测盲区
图像分割
自动化识别
ultrasonic TOFD
weld
near-surface defect
blind area of detection
image segmentation
automatic identification
