基于无人机及YOLOX视觉算法的大跨度钢结构吊装过程位移监测

PDF(4068 KB)

振动与冲击 ›› 2024, Vol. 43 ›› Issue (17) : 61-70.

论文

基于无人机及YOLOX视觉算法的大跨度钢结构吊装过程位移监测

李万润1,2,3，范博源1，赵文海1，杜永峰1,2,3

作者信息 +

Displacement monitoring in large-span steel structure hoisting process based on drone and YOLOX visual algorithm

LI Wanrun1,2,3, FAN Boyuan1, ZHAO Wenhai1, DU Yongfeng1,2,3

Author information +

文章历史 +

摘要

在大跨度钢结构吊装施工过程中，节点位移及结构变形关系到吊装施工的安全和质量。对于传统接触式监测方法存在的耗时、耗力且维护费用高等问题，提出了一种以无人机为载体的非接触式监测方式。首先，针对大跨度钢结构吊装过程中无人机近距采集视角受限的问题，采用Harris图像拼接算法进行全景拼接，并与图像加权融合相结合，消除图像拼接中产生的不利光标及拼接缝，实现整体、高精度的大跨度结构图像的无缝拼接；其次，采用加入CBAM(Convolutional Block Attention Module)双通道注意力机制的YOLOX视觉算法解决复杂背景下不同像素面积的小目标图像识别、坐标提取和位移监测；最后，对四种不同检测模型进行对比评估，并通过对比实验室不同工况试验和实际工程验证该方法在施工环境下对大跨度钢结构测点位移监测的可行性。试验结果表明，加入CBAM注意力机制的YOLOX检测模型的平均精度及置信度均优于其他三种网络模型，且视觉识别的位移信息与Leica全站仪的误差均在亚毫米级内，满足实际工程精度的要求，实现了复杂背景下的小目标位移监测，具备较高的经济效益和广泛的应用前景。

Abstract

In the process of lifting large span steel structures, node displacements and structural deformations are related to the safety and quality of the lifting construction. For the traditional contact monitoring methods, which are time-consuming, labour-intensive and expensive to maintain, a non-contact monitoring method is proposed with a drone as the carrier. Firstly, to address the problem of limited proximity of the UAV during the lifting of large-span steel structures, the Harris image stitching algorithm is used for panoramic stitching and combined with image weighting fusion to eliminate unfavourable cursors and stitching seams in the image stitching and achieve seamless stitching of overall, high-precision images of large-span structures. Secondly, the YOLOX vision algorithm incorporating the CBAM (Convolutional Block Attention Module) dual-channel attention mechanism is adopted to solve the problem of small target image recognition, coordinate extraction and displacement monitoring with different pixel areas under complex backgrounds. Finally, the four different testing models were compared and evaluated. The experimental results show that the average accuracy and confidence of the YOLOX detection model with CBAM attention mechanism are better than the other three network models, and the errors of the visually identified displacement information and the Leica total station are within sub-millimetre level, which meet the requirements of practical engineering accuracy and achieve small target displacement monitoring in complex backgrounds, with high economic benefits and wide application prospects.

导出引用

李万润1, 2, 3, 范博源1, 赵文海1, 杜永峰1, 2, 3. 基于无人机及YOLOX视觉算法的大跨度钢结构吊装过程位移监测[J]. 振动与冲击, 2024, 43(17): 61-70

LI Wanrun1, 2, 3, FAN Boyuan1, ZHAO Wenhai1, DU Yongfeng1, 2, 3. Displacement monitoring in large-span steel structure hoisting process based on drone and YOLOX visual algorithm[J]. Journal of Vibration and Shock, 2024, 43(17): 61-70

参考文献

[1] 常乐,黄翔,赵有山,等.跨度120m三角锥体钢结构施工滑移监测方法研究[J].建筑结构学报,2022,43(09):251-259. CHANG Le, HUANG Xiang, ZHAO Youshan, et al. Research on construction slip monitoring method of spatial steelstructure with triangular cone shape spanning 120 meters [J]. Journal of Building Structures, 2022,43(09):251-259. [2] 苗吉军,侯晓燕,王典武,等.大跨度异形钢桁架安装过程试验研究及实时监控[J].建筑结构,2013,43(06):22-27. MIAO Jijun, HOU Xiaoyan, WANG Dianwu, et al. Experimental research and real-time monitoring on construction process of large span deformed steel truss [J]. Building construction, 2013,43(06):22-27. [3] 池玉辰,王胡鹏,陈鑫,等.基于LoRa无线技术的钢拱肋吊装监测系统设计[J].自动化与仪器仪表,2021(09):83-86. CHI Yuchen, WANG Hupeng CHEN Xin, et al. Design of steel arch rib hoisting monitoring system based on LoRa wireless technology [J]. Automation and Instrumentation, 2021(09):83-86. [4] 罗尧治,王小波,杨鹏程,等.某钢结构廊桥施工吊装过程监测[J].施工技术,2010,39(02):10-13. LUO Yaozhi, WANG Xiaobo, YANG Pengcheng, et al. Monitoring on the Lifting of a Steel Structure Gallery[J]. Construction techniques, 2010,39(02):10-13. [5] 余加勇，邵旭东，孟晓林,等．基于自动型全站仪的桥梁结构动态监测试验[J]．中国公路学报，2014,27(10);55-64+92. YU Jiayong, SHAO Xudong, MENG Xiaolin, et al. Experiment of Dynamic Monitoring of Bridge Structures Using Robotic Total Station [J]. Chinese Journal of Highways, 2014,27(10);55-64+92. [6] 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. YE Xiaowei, DONG Chuanzhi. Review of Computer Vision-based Structural Displacement Monitoring [J]. Chinese Journal of Highways, 2019,32(11):21-39. [7] 董传智. 基于机器视觉的桥梁健康监测与状态评估[D].浙江大学,2016. DONG Chuanzhi. Machine Vision-based Bridge Health Monitoringand Condition Assessment [D]. Zhejiang University, 2016. [8] 季云峰.无目标计算机视觉技术在斜拉索振动测试中的应用研究[J].振动与冲击,2013,32(20):184-188+202. JI Yunfeng. Application of non-target computer-vision-based technique in stay-cable vibration measurement [J]. Journal of Vibration and Shock, 2013,32(20):184-188+202. [9] 朱前坤,王军营,杜永峰,等.基于计算机视觉的结构应变无靶标鲁棒监测[J/OL].建筑结构学报:1-10[2022-12-23]. ZHU Qiankun, WANG Junying, DU Yongfeng, et al. Unmarked robust monitoring of structural strain based oncomputer vision [J]. Journal of Building Structures: 1-10[2022-12-23]. [10] 董传智,叶肖伟,刘坦.非接触式结构动力特性识别方法及试验验证[J].振动与冲击,2017,36(01):188-193. DONG Chuanzhi， YE Xiaowei， LIU Tan． Noncontact structural dynamic characteristics identification method and its test verification［J］． Journal of Vibration and Shock，2017,36(01):188-193. [11] Zhang Y, Zhao X, Liu P. Multi-point displacement monitoring based on full convolutional neural network and smartphone[J]. IEEE Access, 2019, 7: 139628-139634. [12] Zhang Y, Liu P, Zhao X. Structural displacement monitoring based on mask regions with convolutional neural network[J]. Construction and Building Materials, 2021, 267: 120923. [13] 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. [14] 李万润,赵文海,李家富,等.基于光流法和无人机的大型风力机结构动力特性测试[J/OL].湖南大学学报(自然科学版):1-11[2022-11-21]. LI Wanrun, ZHAO Wenhai, LI Jiafu, et al. Structural dynamic characteristic test of large wind turbine based on optical flow method and UAV [J/OL]. Journal of Hunan University (Natural Science Edition): 1-11[2022-11-21]. [15] Ge Y, Yu X, Chen M, et al. Monitoring dynamic deformation of building using unmanned aerial vehicle[J]. Mathematical Problems in Engineering, 2021, 2021: 1-11. [16] Sun J, Peng B, Wang C C, et al. Building displacement measurement and analysis based on UAV images[J]. Automation in Construction, 2022, 140: 104367. [17] 韩怡天,冯东明,吴刚.基于机器视觉与无人机的结构动位移测量方法[J].振动与冲击,2022,41(19):1-7. HAN Yitian, FENG Dongming, WU Gang. Measurement method of structural dynamic displacement based on machine vision and UAV [J]. Journal of Vibration and Shock, 2022,41(19):1-7. [18] Zhan W, Sun C, Wang M, et al. An improved Yolov5 real-time detection method for small objects captured by UAV[J]. Soft Computing, 2022, 26: 361-373. [19] Lim J S, Astrid M, Yoon H J, et al. Small object detection using context and attention[C]//2021 International Conference on Artificial Intelligence in Information and Communication (ICAIIC). IEEE, 2021: 181-186. [20] ZHANG Y, YANG J, LV L G, et al. Research of Image Mosaic Algorithm Based on Harris Corner Detection[C]//5th International Conference on Information Engineering for Mechanics and Materials. Atlantis Press, 2015: 1294-1297. [21] Harris C, Stephens M, Detector E. Alvey vision conference[J]. Alvey Vision Club, 1988. [22] Brown M, Szeliski R, Winder S. Multi-image matching using multi-scale oriented patches[C]//2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05). IEEE, 2005, 1: 510-517. [23] Fischler M A, Bolles R C. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography[J]. Communications of the ACM, 1981, 24(6): 381-395. [24] Ge Z, Liu S, Wang F, et al. Yolox: Exceeding yolo series in 2021[J]. arXiv preprint arXiv:2107.08430, 2021. [25] Zhang J, Ke S. Improved YOLOX fire scenario detection method[J]. Wireless Communications and Mobile Computing, 2022, 2022. [26] Liu B, Huang J, Lin S, et al. Improved YOLOX-S abnormal condition detection for power transmission line corridors[C]//2021 IEEE 3rd International Conference on Power Data Science (ICPDS). IEEE, 2021: 13-16. Piscataway, New York, USA: IEEE, 2021: 13-16. [27] Liu M, Zhu C. Residual YOLOX-based ship object detection method[C]//2022 2nd International Conference on Consumer Electronics and Computer Engineering (ICCECE). IEEE, 2022: 427-431. [28] Wang J, Tang C, Li J. Towards Real-time Analysis of Marine Phytoplankton Images Sampled at High Frame Rate by a YOLOX-based Object Detection Algorithm[C]//OCEANS 2022-Chennai. IEEE, 2022: 1-9. [29] Woo S, Park J, Lee J Y, et al. Cbam: Convolutional block attention module[C]//Proceedings of the European conference on computer vision (ECCV). 2018: 3-19. [30] LI L, FANG B, ZHU J. Performance Analysis of the YOLOv4 Algorithm for Pavement Damage Image Detection with Different Embedding Positions of CBAM Modules[J]. Applied Sciences, 2022, 12(19): 10180.