基于YOLOv8可变形卷积的工人安全帽检测系统
Deformable convolution system for the detection of workers' safety helmets using YOLOv8
- 物联网学报 2026年
- 作者机构：
天津城建大学计算机与信息工程学院，天津 300384
- 作者简介：
陈帅宇，chenshuaiyu0906@163.com
- 基金信息：天津市哲学社会科学规划项目(TJGL19XSX-045);国家级大学生创新创业训练计划项目(202410792012)
DOI：
中图分类号： TN391.4收稿：2025-03-06，
修回：2025-06-09，
录用：2025-08-07，
- 稿件说明：
移动端阅览
孙叶美, 李淼, 段聪聪, 等. 基于YOLOv8可变形卷积的工人安全帽检测系统[J/OL]. 物联网学报, 2026.
Sun Yemei, Li Miao, Duan Congcong, et al. Deformable convolution system for the detection of workers' safety helmets using YOLOv8[J/OL]. Chinese Journal on Internet of Things, 2026.
摘要
建筑工地的复杂环境给工人安全帽佩戴检测带来了极大的挑战。为精准辨识复杂施工环境中作业人员安全帽佩戴情况，有效降低安全事故发生率，提出了一种基于可变形卷积的工人安全帽佩戴检测算法。该算法在YOLOv8骨干网络部分融合可变形卷积模块，缓解了传统标准卷积层使用固定几何结构提取特征时限制模型建模能力的问题，提升了对复杂施工环境的分析识别能力；通过图像几何变换和像素级处理方法，丰富训练样本数量以提升检测模型在不同工地场景中的泛化能力。将该算法部署到Jetson TX2，编写目标检测软件，实现了实际建筑工地场景的现场检测。实验表明：该模型能在复杂施工环境下完成工人的安全帽佩戴检测，在标准数据集上识别准确率为95.2%，且具有一定鲁棒性，可向各类施工现场进行推广。
Abstract
The complex environment of construction sites presents a huge challenge for detecting unsafe behaviors among workers. In order to accurately identify the wear of helmets for workers in complex construction environments
and effectively reduce the incidence of safety accidents
a detection algorithm based on deformable convolution for the identification of safety helmets worn by workers is proposed. The algorithm fuses deformable convolutional modules in the YOLOv8 backbone network part
alleviating the limitations of traditional standard convolution layers
which rely on fixed geometric structures for feature extraction. This enhancement improves the model's ability to analyze and identify complex construction environments. By employing image geometric transformations and pixel-level processing
the algorithm enriches the number of training samples
thereby enhancing the generalization capability of the detection model across various site scenarios. The algorithm is deployed on the Jetson TX2
and the target detection software is developed to facilitate on-site detection of the actual construction site environment. The results indicate that the model can effectively detect unsafe behaviors among workers in complex construction environments
achieving a recognition accuracy of 95.2% on a standard dataset. Furthermore
the model demonstrates robustness and can be applied across various construction sites.
关键词
Keywords
references
王林 , 孙礼涛 , 翁鸿飞 , 等 . 一种建筑施工现场不安全行为识别系统 [P ] . 四川省 : CN202021009212.3 , 2021-01-05 .
Wang L , Sun LT , Weng HF , et al . A kind of unsafe behavior recognition system for building construction site [P ] . Sichuan : CN202021009212.3 , 2021-01-05 .
Memarzadeh M , Golparvar-Fard M , Niebles J C . Automated 2D detection of construction equipment and workers from site video streams using histograms of oriented gradients and colors [J ] . Automation in Construction , 2013 , 32 : 24 - 37 .
Zhu Z , Ren X , Chen Z . Integrated detection and tracking of workforce and equipment from construction jobsite videos [J ] . Automation in Construction , 2017 , 81 : 161 - 171 .
Du S , Shehata M , Badawy W . Hard hat detection in video sequences based on face features, motion and color information [C ] // 2011 3rd international conference on computer research and development . ieee , 2011 , 4 : 25 - 29 .
Shrestha K , Shrestha P P , Bajracharya D , et al . Hard‐hat detection for construction safety visualization [J ] . Journal of Construction Engineering , 2015 , 2015 ( 1 ): 721380 .
Rubaiyat A H M , Toma T T , Kalantari-Khandani M , et al . Automatic detection of helmet uses for construction safety [C ] // 2016 IEEE/WIC/ACM International Conference on Web Intelligence Workshops (WIW) . IEEE , 2016 : 135 - 142 .
刘云波 , 黄华 . 施工现场安全帽佩戴情况监控技术研究 [J ] . 电子科技 , 2015 , 28 ( 04 ): 69 - 72 .
Liu Y B , Huang H . Research on monitoring technology of helmet wearing situation in construction site [J ] . Electronic Science and Technology , 2015 , 28 ( 04 ): 69 - 72 .
Mneymneh B E , Abbas M , Khoury H . Automated hardhat detection for construction safety applications [J ] . Procedia engineering , 2017 , 196 : 895 - 902 .
卢颖 , 吕希凡 , 郭良杰 , 等 . 基于Kinect的地铁乘客不安全行为识别方法与实验 [J ] . 中国安全生产科学技术 , 2021 , 17 ( 12 ): 162 - 168 .
Lu Y , Lu X F , Guo L J , et al . Kinect-based subway passenger unsafe behavior recognition method and experiment [J ] . China Safety Production Science and Technology , 2021 , 17 ( 12 ): 162 - 168 .
杨赛烽 . 基于Kinect的罐笼内矿工不安全行为识别方法研究 [D ] . 江苏 : 中国矿业大学(江苏) , 2019 .
Yang S F . Research on the identification method of miners' unsafe behavior in cage based on Kinect [D ] . Jiangsu : China University of Mining and Technology (Jiangsu) , 2019 .
赵江平 , 王垚 . 基于图像识别技术的不安全行为识别 [J ] . 安全与环境工程 , 2020 , 27 ( 01 ): 158 - 165 .
Zhao J P , Wang Y . Unsafe behavior recognition based on image recognition technology [J ] . Safety and Environmental Engineering , 2020 , 27 ( 01 ): 158 - 165 .
常捷 , 张国维 , 陈文江 , 等 . 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 ( 02 ): 31 - 37 .
CHANG J , ZHANG G W , CHEN W J , et al . Detection of unsafe behavior at gas stations based on YOLO-V3 algorithm [J ] . Chinese Journal of Safety Science , 2023 , 33 ( 02 ): 31 - 37 .
纪执安 , 周云奕 , 张玉媛 , 等 . 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 ( 07 ): 38 - 43 .
JI Z A , ZHOU Y Y , ZHANG Y Y , et al . Detection of unsafe behaviors at industrial sites based on improved YOLOv5 [J ] . Chinese Journal of Safety Science , 2024 , 34 ( 07 ): 38 - 43 .
Liu S , Huang D . Receptive field block net for accurate and fast object detection [C ] // Proceedings of the European conference on computer vision (ECCV) . 2018 : 385 - 400 .
王贞 , 邱杭 , 吴斌 , 等 . 基于CCG-YOLOv8的施工场景下安全帽佩戴检测 [J ] . 武汉理工大学学报 , 2024 , 46 ( 6 ): 73 - 80 .
WANG Z , QIU H , WU B , et al . Detection of helmet wearing in construction scenarios based on CCG-YOLOv8 [J ] . Journal of Wuhan University of Technology , 2024 , 46 ( 6 ): 73 - 80 .
Ren S , He K , Girshick R , et al . Faster R-CNN: Towards real-time object detection with region proposal networks [J ] . IEEE transactions on pattern analysis and machine intelligence , 2016 , 39 ( 6 ): 1137 - 1149 .
张艳 , 孙晶雪 , 孙叶美 , 等 . 基于分割注意力与线性变换的轻量化目标检测 [J ] . 浙江大学学报(工学版) , 2023 , 57 ( 06 ): 1195 - 1204 .
Zhang Y , Sun X J , Sun Y M , et al . Lightweight target detection based on segmented attention and linear transformation [J ] . Journal of Zhejiang University (Engineering Edition) , 2023 , 57 ( 06 ): 1195 - 1204 .
孙勇 . 建筑工人不安全行为智能检测系统研究与设计实现 [D ] . 深圳 : 深圳大学 , 2020 .
Sun Y . Research and design implementation of intelligent detection system for construction workers' unsafe behavior [D ] . Shenzhen : Shenzhen University , 2020 .
王晨 . 面向作业安防的智能检测技术研究与实现 [D ] . 西安 : 西安工业大学 , 2022 .
Wang C . Research and Implementation of Intelligent Detection Technology for Job Security [D ] . Xi'an : Xi'an University of Technology , 2022 .
Varghese R , Sambath M . YOLOv8: A Novel Object Detection Algorithm with Enhanced Performance and Robustness [C ] // 2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS) . IEEE , 2024 : 1 - 6 .
张田楷 . 基于改进YOLOv8的安全帽佩戴检测方法的研究 [D ] . 大连 : 大连交通大学 , 2024 .
Zhang T K, Research on helmet wearing detection method based on improved YOLOv8 [D ] . Dalian : Dalian Jiaotong University , 2024 .
房鑫 , 陈兵旗 , 彭书博 , 等 . 基于改进YOLOv4的前方车辆检测方法 [J ] . 传感器与微系统 , 2024 , 43 ( 10 ): 155 - 159 .
FANG X , CHEN B Q , PENG S B , et al . A forward vehicle detection method based on improved YOLOv4 [J ] . Sensors and Microsystems , 2024 , 43 ( 10 ): 155 - 159 .
温王鹏 , 罗文婷 , 李林 , 等 . 基于YOLOv7和图像分块的车道线破损检测算法 [J ] . 传感器与微系统 , 2024 , 43 ( 09 ): 131 - 134+139 .
WEN W P , LUO W T , Li L , et al . Lane line breakage detection algorithm based on YOLOv7 and image chunking [J ] . Sensors and Microsystems , 2024 , 43 ( 09 ): 131 - 134+139 .
胡海宁 , 黄雷阳 , 杨洪刚 , 等 . 改进YOLOv8n的轻量型蜂窝陶瓷缺陷检测算法 [J ] . 激光与光电子学进展 , 2024 , 61 ( 22 ): 160 - 169 .
Hu H N , Huang L Y , Yang H G , et al . Improved YOLOv8n defect detection algorithm for lightweight honeycomb ceramics [J ] . Advances in Lasers and Optoelectronics , 2024 , 61 ( 22 ): 160 - 169 .
Khanam R , Hussain M . Yolov11: An overview of the key architectural enhancements. arxiv 2024 [J ] . arxiv preprint arxiv: 2410.17725 .
罗臻 , 蒋瑜 , 李维诚 . 基于改进YOLO11n的道路缺陷检测模型 [J/OL ] . 微电子学与计算机 , 1 - 13 [ 2025-02-28 ] .
Luo Z , Jiang Y , Li W C . Road defect detection model based on improved YOLO 11 n[J/OL ] . Microelectronics and Computers , 1 - 13 [ 2025-02-28 ] .
阮庭海 , 樊卫华 . 基于YOLOv8的自动驾驶车道检测改进算法 [J ] . 计算机测量与控制 , 2025 , 33 ( 02 ): 31 - 36 .
Ruan T H , Fan W H . Improved algorithm for automatic driving lane detection based on YOLOv8 [J ] . Computerized Measurement and Control , 2025 , 33 ( 02 ): 31 - 36 .
庞国涛 , 邓晓平 , 于海洋 , 等 . 基于改进YOLOv5s的预制叠合板生产质量视觉检测研究 [J ] . 土木建筑工程信息技术 , 2025 , 17 ( 01 ): 37 - 42 .
PANG G T , DANG X P , YU H Y , et al . Research on visual inspection of prefabricated laminated panel production quality based on improved YOLOv5s [J ] . Information Technology for Civil Engineering and Construction , 2025 , 17 ( 01 ): 37 - 42 .
李希 , 潘誉 . 基于改进YOLOv8的探地雷达管线目标检测方法 [J ] . 计算机与现代化 , 2025 ,( 01 ): 94 - 99 .
LI X , PAN Y . Ground-penetrating radar pipeline target detection method based on improved YOLOv8 [J ] . Computer and Modernization , 2025 ,( 01 ): 94 - 99 .
Lei F. ; Tang F. ; Li S. Underwater Target Detection Algorithm Based on Improved YOLOv5 . J. Mar. Sci. Eng. 2022 , 10 , 310 .
0
浏览量
0
下载量
0
CSCD
- 网络PDF下载
- XML下载
- Meta-XML下载
- BibTeX下载
- Endnote下载
- RefMan 下载
- NoteExpress下载
- NoteFirst下载
关联资源
相关文章
相关作者
相关机构
- 技术支持由北京北大方正电子有限公司提供 京ICP备09064830号-19
京公网安备11010802024621 - 本系统建议在Chrome、 IE9+ 以上版本浏览器阅读本站内容，360浏览器请切换至极速模式
- Cookies帮助我们提供服务并提供个性化体验。使用本网站，即表示您同意我们使用Cookies
- 总访问量：187980 今日访问量：67