基于图像识别的高强度螺栓松动检测方法的研究

郭珍珠; 赵伟; 陈涵深; 吕硕

doi:10.13204/j.gyjzG21042001

基于图像识别的高强度螺栓松动检测方法的研究

doi: 10.13204/j.gyjzG21042001

郭珍珠¹,
赵伟^1,2, ,,
陈涵深²,
吕硕³

1. 江苏科技大学土木工程与建筑学院, 江苏镇江 212000;
2. 浙江交通职业技术学院, 教育部公路水运钢结构桥梁应用协同创新中心, 杭州 311112;
3. 青岛理工大学土木工程学院, 山东青岛 266525

基金项目:

浙江省基础公益研究计划项目(LGG21E080001)

浙江省交通运输厅科技计划项目(2022013)。

详细信息

作者简介:
郭珍珠,女,1995年出生,硕士研究生。

通讯作者:
赵伟,男,631238675@qq.com。

计量
- 文章访问数: 606
- HTML全文浏览量: 25
- PDF下载量: 12
- 被引次数: 0
出版历程
- 收稿日期: 2021-04-20
- 网络出版日期: 2022-06-30
- 刊出日期: 2022-02-20

Research on a Detection Method for Loosening of High-Strength Bolts Based on Image Recognition Techniques

1. School of Civil Engineering and Architecture, Jiangsu University of Science and Technology, Zhenjiang 212000, China;
2. Zhejiang College of Communications Ministry of Education, Collaborative Innovation Center for Application of Highway and Waterway Steel Bridge, Hangzhou 311112, China;
3. School of Civil Engineering, Qingdao University of Technology, Qingdao 266525, China

摘要

摘要: 针对钢桥高强度螺栓人工批量巡检效率低、接触传感设备成本高昂的问题,提出了一种基于图像识别的高强度螺栓松动检测方法。利用螺栓角点位置识别算法对螺栓图像样本进行白色掩膜构建、掩膜小型噪点剔除和感兴趣区域分割等处理,确定螺栓角点位置坐标,结合相机成像相似映射原理推算螺栓松动角度,进而根据螺栓松动角度评估预紧力损失。对不同型号螺栓在不同水平视距下旋转10°、20°、30°采集样本,并将其导入算法进行试验验证。结果表明,基于该方法的螺栓松动角度检测准确率达90%以上,满足工程检测要求,并能够有效评估高强度螺栓预紧力损失。
- 螺栓松动 /
- 算法 /
- 图像 /
- 特征点
Abstract: Aiming at the low efficiency of manual batch inspection for high-strength bolts steel bridge and the high cost of contact sensing equipment, a detection method for the loosening of high strength bolts based on image recognition was proposed. Through the recognition algorithm for the position of bolt corner, the bolt pattern samples were processed by constructing white mask, removing small noise and segmenting ROI of the region of interest to determine the position coordinates of bolt corner. The bolt loosening angle was calculated by using the camera imaging similarity mapping principle, and the preload loss was evaluated according to the relation between the loosening angle and the preload. Then specimens were collected by rotating 10°, 20°, and 30°of different types of bolts at different horizontal viewing angles, and inputed them into the algorithm for experimental verification.The results showed that the accuracy of bolt looseness detection based on the method was over 90%, which could meet the requirements of engineering inspection and effectively evaluate the loss of preload of high-strength bolts.
- loosening of bolt /
- algorithm /
- images /
- feature points

HTML全文

参考文献(14)

[1]	虞薇芳.主次梁承压型高强度螺栓连接节点受力性能研究[D].武汉:武汉理工大学,2009.
[2]	王涛,罗毅,刘绍鹏,等.基于压电主动传感方式的螺栓松动检测实验研究[J].传感技术学报,2013,26(8):1059-1063.
[3]	唐超.基于PZT阻抗法的钢结构桥梁高强度螺栓损伤监测试验研究[D].长沙:长沙理工大学,2017.
[4]	陈佳雷.基于压电陶瓷智能垫片的海洋结构螺栓松动监测研究[D].大连:大连理工大学,2019.
[5]	李宇宏.基于深度学习的路面病害自动检测及评估方法研究[D].苏州:江苏大学,2020.
[6]	刘鹏.基于激光投射和图像识别的深基坑位移监测方法研究[D].大连:大连理工大学,2019.
[7]	王达磊,彭博,潘玥,等.基于深度神经网络的锈蚀图像分割与定量分析[J].华南理工大学学报(自然科学版),2018,46(12):121-127,146.
[8]	PARK J H, HUYNH T C, CHOI S H, et al. Vision-based technique for bolt-loosening detection in wind turbine tower[J]. Wind Struct, 2015,21(6):709-726.
[9]	CHA Y J, YOU K, CHOI W. Vision-based detection of loosened bolts using the hough transform and support vector machines[J]. Autom. Constr, 2016,71:181-188.
[10]	RAMANA L,CHOI W,CHA Y J.Automated vision-based loosened bolt detection using the cascade detector[J].Sensors and Instrumentation,2017(5):23-28.
[11]	ZHAO X F,ZHANG Y,WANG N.Bolt loosening angle detection technology using deep learning[J].Structural Control and Health Monitoring,2019,26(1):1-14.
[12]	李健,丁小奇,陈光,等.基于改进高斯滤波算法的叶片图像去噪方法[J].南方农业学报,2019,50(6):1385-1391.
[13]	周雪梅,潘多.一种基于形态学结构元素优化的车牌快速检测算法[J].西南师范大学学报(自然科学版),2020,45(7):130-136.
[14]	山本晃.螺纹联接的理论与计算[M].郭可谦,高素娟,译.上海:上海科学技术文献出版社, 1984:45-52.