基于尺度不变特征变换算法(SIFT)立体匹配的结构三维振动位移监测

杜永峰; 马天军; 韩博; 李虎

doi:10.13204/j.gyjzG23061302

基于尺度不变特征变换算法(SIFT)立体匹配的结构三维振动位移监测

doi: 10.13204/j.gyjzG23061302

杜永峰^1,2,
马天军^1, ,,
韩博¹,
李虎¹

1. 兰州理工大学防震减灾研究所, 兰州 730050;
2. 兰州理工大学土木工程减震隔震技术研发甘肃省国际科技合作基地, 兰州 730050

基金项目:

国家自然科学基金项目（52178291）。

详细信息

作者简介:
杜永峰,男,1962年出生,博士,教授。

通讯作者:
马天军,419553801@qq.com。

计量
- 文章访问数: 74
- HTML全文浏览量: 17
- PDF下载量: 1
- 被引次数: 0
出版历程
- 收稿日期: 2023-06-13
- 网络出版日期: 2024-02-28

3D Vibration Displacement Monitoring of Structures Base on SIFT Stereo Matching

DU Yongfeng^1,2,
MA Tianjun^{1
, ,},
HAN Bo¹,
LI Hu¹

1. Institute of Earthquake Protection and Disaster Mitigation, Lanzhou University of Technology, Lanzhou 730050, China;
2. International Research Base of Seismic Mitigation and Isolation of Gansu Province, Lanzhou University of Technology, Lanzhou 730050, China

摘要

摘要: 为解决传统单目视觉位移测量方法无法获得结构三维运动信息的不足，将双目立体视觉应用于结构的三维位移监测中。使用Python语言，在Pycharm平台上进行算法编译，采用Shi-Tomasi角点检测算法与尺度不变特征变换算法相结合，实现左右两幅图像特征点的立体匹配。通过图像预处理，提取匹配的目标特征点坐标，进而实现结构三维位移监测，并将该方法应用在五层框架模型的振动台试验中，获得地震动作用下结构的三维位移时程。试验结果表明：该方法测得的位移时程曲线与拉线式位移计数据吻合较好，Z向最大峰值位移相差1.29 mm，误差绝对值在8%以内，在频域也有较好的表现。并通过改变基线距离和相机的偏转角度来验证该方法的鲁棒性，表明了该方法用作结构三维位移全过程监测的可行性。
- 尺度不变特征变换 /
- 双目视觉 /
- 特征点匹配 /
- 三维位移监测
Abstract: In order to solve the shortage of traditional monocular vision displacement measurement methods that cannot obtain 3D motion information of structures, binocular stereo vision was applied to the 3D displacement monitoring of structures. Using Python language, the algorithm was compiled on Pycharm platform, and Shi-Tomasi corner point detection algorithm was combined with scale invariant feature transformation algorithm to achieve stereo matching of feature points of left and right images. Through image pre-processing, the coordinates of the matched target feature points were extracted, and then the three-dimensional displacement monitoring of the structure was realized, and the method was applied to the shaking table test of a five-story frame model to obtain the three-dimensional displacement time history of the structure under the action of ground shaking. The test results showed that the displacement time history curve measured by the method matched well with the data of the pull-wire displacement meter, and the maximum peak displacement in Z-direction differed by 1.29 mm with the absolute value of error within 8%, and also had a good performance in the frequency domain. Besides, the robustness of the method was verified by changing the baseline distance and the deflection angle of the camera, which showed the feasibility of the method used as the whole process monitoring of the 3D displacement of the structure.
- scale-invariant feature transform /
- binocular vision /
- feature point matching /
- three-dimensional displacement monitoring

HTML全文

参考文献(23)

[1]	鲍跃全, 李惠. 人工智能时代的土木工程[J]. 土木工程学报, 2019, 52(5):1-11.
[2]	SPENCER JR B F, HOSKERE V, NARAZAKI Y. Advances in computer vision-based civil infrastructure inspection and monitoring[J]. Engineering, 2019, 5(2):199-222.
[3]	FENG D, FENG M Q. Vision-based multipoint displacement measurement for structural health monitoring[J]. Structural Control and Health Monitoring, 2016, 23(5):876-890.
[4]	单宝华, 王海, 袁文厅, 等. 地震动作用下三层框架模型倒塌的平行双目视觉监测[J]. 振动与冲击, 2018, 37(12):48-52.
[5]	叶肖伟, 董传智. 基于计算机视觉的结构位移监测综述[J]. 中国公路学报, 2019, 32(11):21-39.
[6]	DONG C, CELIK O, CATBAS F N, et al. Structural displacement monitoring using deep learning-based full field optical flow methods[J]. Structure and Infrastructure Engineering, 2020, 16(1):51-71.
[7]	周颖, 张立迅, 刘彤, 等. 基于计算机视觉的结构系统识别[J]. 土木工程学报, 2018, 51(11):17-23.
[8]	周云, 程依婷. 基于数字图像相关理论的非接触式结构位移测量方法[J]. 湖南大学学报(自然科学版), 2021, 48(5):1-9.
[9]	杜永峰, 李向雄, 张超, 等. 基于计算机视觉的复杂隔震结构支座变形监测及分析[J]. 工业建筑, 2022, 52(10):46-52.
[10]	朱前坤, 陈建邦, 张琼, 等. 基于计算机视觉人行桥挠度影响线非接触式识别[J]. 工程力学, 2021, 38(8):145-153.
[11]	YOON H, ELANWAR H, CHOI H, et al. Target-free approach for vision-based structural system identification using consumer-grade cameras[J]. Structural Control and Health Monitoring, 2016, 12(23):1405-1416.
[12]	LI H, ZHANG B. Application of integrated binocular stereo vision measurement and wireless sensor system in athlete displacement test[J]. Alexandria Engineering Journal, 2021, 60(5):4325-4335.
[13]	YUAN C, XIONG B, LI X, et al. A novel intelligent inspection robot with deep stereo vision for three-dimensional concrete damage detection and quantification[J]. Structural Health Monitoring, 2022, 21(3):788-802.
[14]	SHAO Y, LI L, LI J, et al. Computer vision based target-free 3D vibration displacement measurement of structures[J/OL]. Engineering Structures, 2021, 246[2023-06-13].https://doi.org/10.1016/j.engstruct.2021.113040.
[15]	邱志成, 汪先锋. 基于双目视觉的振动测量及控制[J]. 振动、测试与诊断, 2018, 38(1):51-58 ,205.
[16]	张广军. 视觉测量[M]. 北京:科学出版社, 2008.
[17]	ZHANG Z. A flexible new technique for camera calibration[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(11):1330-1334.
[18]	李亚旋. 双目立体视觉三维重建的立体匹配算法研究[D]. 太原:太原理工大学, 2020.
[19]	王笛, 胡辽林. 基于双目视觉的改进特征立体匹配方法[J]. 电子学报, 2022, 50(1):157-166.
[20]	LOWE D G. Distinctive image features from scale-invari-antkey point[J].International Journal of Computer Vision, 2004, 60(2):91-110.
[21]	杨伟姣, 杨先海. 结合Harris算子和极线约束改进的SIFT立体匹配算法[J]. 高技术通讯, 2019, 29(11):1082-1086.
[22]	ARYA S, MOUNT D M, NETANYAHU N S, et al. An optimal algorithm for approximate nearest neighbor searching[J]. Association for Computing Machinery Journal, 1998, 45(6):891-923.
[23]	KHUC T. Computer vision based structural identification framework for bridge health monitoring[D]. Florida:University of Central Florida, 2016.