基于视觉的工人高处攀爬不安全行为识别模型

doi:10.16265/j.cnki.issn1003-3033.2025.02.0278

中国安全科学学报 ›› 2025, Vol. 35 ›› Issue (2): 144-151.doi: 10.16265/j.cnki.issn1003-3033.2025.02.0278

基于视觉的工人高处攀爬不安全行为识别模型

张泽辉¹(), 张乾隆¹, 徐晓滨¹, 赵祖国², 王海泉³, 李昊⁴

¹ 杭州电子科技大学中国-奥地利人工智能与先进制造“一带一路”联合实验室,浙江杭州 310018
² 中职物联(湖北)信息科技有限公司,湖北武汉 430014
³ 中原工学院电子信息学院,河南郑州 450007
⁴ 宁夏长骏科技咨询有限公司,宁夏银川 750001

收稿日期:2024-09-10 修回日期:2024-11-22 出版日期:2025-02-28
作者简介:
张泽辉 (1994—),男,湖南衡阳人,博士,副研究员,主要从事计算机视觉、人员行为识别、工业安全管控等方面的研究。E-mail: zhangzehui@hdu.edu.cn。
徐晓滨教授
王海泉教授
基金资助:
国家重点研发计划资助项目(2022YFE0210700); 国家水运安全工程技术研究中心开放基金资助(A202403); 浙江省自然科学基金资助(LTGG24F030004); 浙江省“尖兵”,“领雁”研发攻关计划资助项目(2024C03254)

Unsafe behavior recognition model of high climbing workers based on vision

ZHANG Zehui¹(), ZHANG Qianlong¹, XU Xiaobin¹, ZHAO Zuguo², WANG Haiquan³, LI Hao⁴

¹ China-Austria Belt and Road Joint Laboratory on Artificial Intelligence and Advanced Manufacturing, Hangzhou Dianzi University, Hangzhou Zhejiang 310018, China
² Secondary Vocational Internet of Things (Hubei) Information Technology Co., Ltd., Wuhan Hubei 430014, China
³ School of Electronic Information, Zhongyuan University of Technology, Zhengzhou Henan 450007, China
⁴ Ningxia Changjun Technology Consulting Co.,Ltd., Yinchuan Ningxia 750001, China

Received:2024-09-10 Revised:2024-11-22 Published:2025-02-28

摘要/Abstract

摘要：

为精准识别高处作业人员攀爬过程中出现的不安全行为,提出一种基于视觉的工人高处攀爬不安全行为识别模型,由人体姿态估计算法和一维卷积不安全行为识别算法组成。人体姿态估计研究者采用量子化自编码器对人体关键点进行结构化建模,实现人体关键点坐标的检测;结合高处作业安全行为知识,基于一维卷积神经网络(1DCNN)模型构建高处攀爬不安全行为识别算法,并通过实验进行验证。结果表明:该模型在人员无遮挡情况和有部分遮挡情况下,分别达到93.91%和90.34%的精度;与支持向量机(SVM)、K最邻近算法(KNN)相比,该模型具有更强的泛化能力。

关键词: 计算机视觉, 工人高处攀爬, 不安全行为, 识别模型, 一维卷积神经网络(1DCNN), 人体姿态估计

Abstract:

In order to accurately identify unsafe behaviors during the climbing process of high-altitude workers, this paper proposed an unsafe behavior recognizing method for high climbing workers based on vision, which included the human pose estimation and the one-dimensional convolutional unsafe behavior recognition models. Quantized autoencoder was used to structurally model human key points in human pose estimation, enabling the detection of human key point coordinates. Combining with safety behavior knowledge in high climbing operations, the unsafe behavior recognition model was constructed based on one-dimensional convolutional neural network model, and it was validated by industrial data experiments. Experimental results show that the accuracy of this method is 93.91% and 90.34% on unobstructed and partially obstructed datasets, respectively. Moreover, compared with support vector machines (SVM) and K-nearest neighbor (KNN), this method has stronger generalization capability.

Key words: computer vision, worker high-altitude climbing, unsafe behavior, recognition model, one-dimensional convolutional neural network(1DCNN), human pose estimation

中图分类号:

X910安全人体学

张泽辉, 张乾隆, 徐晓滨, 赵祖国, 王海泉, 李昊. 基于视觉的工人高处攀爬不安全行为识别模型[J]. 中国安全科学学报, 2025, 35(2): 144-151.

ZHANG Zehui, ZHANG Qianlong, XU Xiaobin, ZHAO Zuguo, WANG Haiquan, LI Hao. Unsafe behavior recognition model of high climbing workers based on vision[J]. China Safety Science Journal, 2025, 35(2): 144-151.

图/表 14

图1

图2

图3

图4

图5

表1

图6

图7

图8

图9

表2

图10

图11

图12

参考文献 24

[1]	国家统计局. 中华人民共和国2023年国民经济和社会发展统计公报[EB/OL].(2024-02-29). https://www.stats.gov.cn/xxgk/sjfb/tjgb2020/202402/t20240229_1947923.html.
[2]	SABRINA S O, WILLAMES D A S, BIANCA M V. Fatal fall-from-height accidents: statistical treatment using the human factors analysis and classification system-HFACS[J]. Journal of Safety Research, 2023: DOI: 10.1016/j.jsr.2023.05.004.
[3]	YANG Jingjing, YE Gui, XIANG Qingting, et al. Insights into the mechanism of construction workers' unsafe behaviors from an individual perspective[J]. Safety Science, 2021: DOI: 10.1016/j.ssci.2020.105004.
[4]	郑霞忠, 王晓宇, 陈述, 等. 高处坠落事故的人因失误与干预策略研究[J]. 中国安全生产科学技术, 2017, 13(6):139-144.
	ZHENG Xiazhong, WANG Xiaoyu, CHEN Shu, et al. Study on human error and intervention strategies of high falling accidents[J]. Journal of Safety Science and Technology, 2017, 13(6):139-144.
[5]	ZHOU Longfei, ZHANG Lin, KONZ N. Computer vision techniques in manufacturing[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2022, 53(1):105-117.
[6]	伍麟, 郝鸿宇, 宋友. 基于计算机视觉的工业金属表面缺陷检测综述[J]. 自动化学报, 2024, 50(7):1261-1283.
	WU Lin, HAO Hongyu, SONG You. A review of metal surface defect detection based on computer vision[J]. Acta Automatica Sinica, 2024, 50(7):1261-1283.
[7]	FANG Weili, DING Lieyun, LOVE P E D, et al. Computer vision applications in construction safety assurance[J]. Automation in Construction, 2020: DOI: 10.1016/j.autcon.2019.103013.
[8]	张萌, 韩豫, 刘泽锋. 深度学习下建筑工人高空安全防护装备检测方法[J]. 中国安全科学学报, 2022, 32(5):140-146. doi: 10.16265/j.cnki.issn1003-3033.2022.05.1141
	ZHANG Meng, HAN Yu, LIU Zefeng. Detection method of high-altitude safety protective equipment for construction workers based on deep learning[J]. China Safety Science Journal, 2022, 32(5):140-146. doi: 10.16265/j.cnki.issn1003-3033.2022.05.1141
[9]	程淑红, 马晓菲, 张仕军, 等. 基于多任务分类的吸烟行为检测[J]. 计量学报, 2020, 41(5):538-543.
	CHENG Shuhong, MA Xiaofei, ZHANG Shijun, et al. Smoking detection algorithm based on multitask classification[J]. Acta Metrologica Sinica, 2020, 41(5):538-543.
[10]	LI Li, ZHONG Boxuan, HUTMACHER J C, et al. Detection of driver manual distraction via image-based hand and ear recognition[J]. Accident Analysis & Prevention, 2020:DOI: 10.1016/j.aap.2020.105432.
[11]	ZHANG Xiaojun. Application of human motion recognition utilizing deep learning and smart wearable device in sports[J]. International Journal of System Assurance Engineering and Management, 2021, 12(4):835-843. doi: 10.1007/s13198-021-01118-7
[12]	XIE Yadong, LI Fan, WU Yue, et al. HearSmoking: smoking detection in driving environment via acoustic sensing on smartphones[J]. IEEE Transactions on Mobile Computing, 2021, 21(8):2847-2860.
[13]	SHEN Jie, XIONG Xin, LI Ying, et al. Detecting safety helmet wearing on construction sites with bounding-box regression and deep transfer learning[J]. Computer-Aided Civil and Infrastructure Engineering, 2021, 36(2):180-196.
[14]	SENER A S, INCE I F, BAYDARGIL H B, et al. Deep learning based automatic vertical height adjustment of incorrectly fastened seat belts for driver and passenger safety in fleet vehicles[J]. Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering, 2022, 236(4):639-654.
[15]	GB 30871—2022,危险化学品企业特殊作业安全规范[S].
	GB 30871-2022, Safety specifications of special work in hazardous chemicals enterprises[S].
[16]	刘勃, 孔韦韦, 肖家钦, 等. 基于RBF神经网络的跌倒检测算法[J]. 计算机技术与发展, 2022, 32(6):167-178.
	LIU Bo, KONG Weiwei, XIAO Jiaqin, et al. Fall detection algorithm based on rbf neural network[J]. Computer Technology and Development, 2022, 32(6):167-178.
[17]	GENG Zigang, WANG Chunyu, WEI Yixuan, et al. Human pose as compositional tokens[C]. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023: DOI: 10.1109/CVPR52729.2023.00071.
[18]	HU Qiyu, ZHANG Guangyi, QIN Zhijin, et al. Robust semantic communications with masked VQ-VAE enabled codebook[J]. IEEE Transactions on Wireless Communications, 2023, 22(12):8707-8722.
[19]	LIU Ze, LIN Yutong, CAO Yue, et al. Swin transformer: hierarchical vision transformer using shifted windows[C]. Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021:DOI: 10.1109/ICCV48922.2021.00986.
[20]	YU Jianbo, ZHOU Xingkang. One-dimensional residual convolutional autoencoder based feature learning for gearbox fault diagnosis[J]. IEEE Transactions on Industrial Informatics, 2020, 16(10):6347-6358.
[21]	池毅, 陈光武. 基于一维卷积神经网络的实时道岔故障诊断[J]. 计算机工程与应用, 2022, 58(20):293-299. doi: 10.3778/j.issn.1002-8331.2103-0348
	CHI Yi, CHEN Guangwu. Real-time turnout fault diagnosis based on one-dimensional convolutional neural network[J]. Computer Engineering and Applications, 2022, 58(20):293-299. doi: 10.3778/j.issn.1002-8331.2103-0348
[22]	SONG Yifan, ZHANG Zhang, SHAN Caifeng, et al. Constructing stronger and faster baselines for skeleton-based action recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 45(2):1474-1488.
[23]	SONG Yifan, ZHANG Zhang, SHAN Caifeng, et al. Richly activated graph convolutional network for robust skeleton-based action recognition[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2020, 31(5):1915-1925.
[24]	ZHAO Baining, LAN Haijuan, NIU Zhenwen, et al. Detection and location of safety protective wear in power substation operation using wear-enhanced YOLOv3 algorithm[J]. IEEE Access, 2021: DOI: 10.1109/ACCESS.2021.3104731.

网络模型参数	参数配置值
检测数	2
输入尺寸	(34,1)
批大小	50
优化器	Adam
学习率	0.001
卷积层数	2
池化层数	2
损失函数	交叉熵损失
训练轮次	80

检测模型	A	P	R
PCT-SVM	88.44	81.63	73.97
PCT-KNN	89.12	81.85	74.31
PCT-1DCNN	92.18	96.16	95.32

基于视觉的工人高处攀爬不安全行为识别模型

Unsafe behavior recognition model of high climbing workers based on vision

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 14

参考文献 24

相关文章 15

编辑推荐

Metrics

本文评价

[1]	王琛. 煤化工企业安全管理体系运行研究[J]. 中国安全科学学报, 2024, 34(S1): 33-38.
[2]	王海军, 齐庆杰, 梁运涛, 齐庆新, 刘英杰, 孙祚. 我国煤矿重特大事故统计分析及对策建议[J]. 中国安全科学学报, 2024, 34(9): 9-18.
[3]	吴立舟, 李华, 李典斌, 吴昱锦, 刘攀旺, 薛曦澄. 基于计算机视觉的塔吊危险区域入侵预警[J]. 中国安全科学学报, 2024, 34(7): 139-146.
[4]	纪执安, 周云奕, 张玉媛, 郭新然, 石坤. 基于改进YOLOv5的工业现场不安全行为检测[J]. 中国安全科学学报, 2024, 34(7): 38-43.
[5]	王军武, 何娟娟, 宋盈辉, 刘一鹏, 陈兆, 郭婧怡. 基于RF-SFLA-SVM的装配式建筑高空作业工人不安全行为预警[J]. 中国安全科学学报, 2024, 34(3): 1-8.
[6]	汪磊, 韦子欣, 邹颖. 民航飞行员不安全行为中的差错与违规分类管理方法[J]. 中国安全科学学报, 2024, 34(12): 8-15.
[7]	李瑚均, 杨珂, 罗讯, 陈辉华, 张建设, 苑东亮. 隧道工人认知偏差对其不安全行为的影响研究:风险感知的中介效应[J]. 中国安全科学学报, 2024, 34(10): 24-29.
[8]	倪国栋, 杨栎, 安瑶瑶, 郑雨茁. 不安全行为信息在建筑工人群体中的传染机制[J]. 中国安全科学学报, 2024, 34(1): 43-52.
[9]	王瑶涵, 宋泽阳, 张利冬. 基于卷积神经网络的安全标识分类算法研究[J]. 中国安全科学学报, 2023, 33(S1): 263-269.
[10]	李华, 吴立舟, 薛曦澄, 钟兴润. 基于计算机视觉的高处临边作业安全巡检[J]. 中国安全科学学报, 2023, 33(9): 69-75.
[11]	莫俊文, 王锐锐, 滕仓国. 高海拔铁路工程施工人员不安全行为评估[J]. 中国安全科学学报, 2023, 33(8): 30-38.
[12]	李乃文, 房小凯, 牛莉霞. 非正式群体凝聚力对矿工不安全行为的影响研究[J]. 中国安全科学学报, 2023, 33(7): 9-15.
[13]	王超, 徐楚昕, 董杰, 王志锋. 基于ST-GCN的空中交通管制员不安全行为识别[J]. 中国安全科学学报, 2023, 33(5): 42-48.
[14]	胡飞翔, 周建亮, 林欣燕, 张家琪, 戴可欣. 情绪对建筑工人不安全行为影响的认知试验[J]. 中国安全科学学报, 2023, 33(4): 75-83.
[15]	常捷, 张国维, 陈文江, 袁狄平, 王永生. 基于YOLO-V3算法的加油站不安全行为检测[J]. 中国安全科学学报, 2023, 33(2): 31-37.