基于无人机检视的公路工程施工人员安全帽佩戴实时检测算法

doi:10.11896/jsjkx.250100047

计算机科学 ›› 2025, Vol. 52 ›› Issue (11A): 250100047-7.doi: 10.11896/jsjkx.250100047

• 计算机图形学&多媒体 • 上一篇下一篇

基于无人机检视的公路工程施工人员安全帽佩戴实时检测算法

文明¹, 吴兴堂², 尚宇豪², 甄键³, 于富才¹

1 北京市应急管理科学技术研究院北京 101101
2 华北电力大学控制与计算机工程学院北京 102206
3 浙江省长三角基础设施科学研究院杭州 310005

出版日期:2025-11-15 发布日期:2025-11-10
通讯作者: 吴兴堂(xtangwu@ncepu.edu.cn)
作者简介:wenming@bjtu.edu.cn
基金资助:
北京市自然科学基金(9244040)

Real-time Helmet Detection Algorithm for Roadway Engineering Construction Based on UAV Visual Inspection

WEN Ming¹, WU Xingtang², SHANG Yuhao², ZHEN Jian³, YU Fucai¹

1 Beijing Academy of Emergency Management Science and Technology,Beijing 101101,China
2 School of Control and Computer Engineering,North China Electric Power University,Beijing 102206,China
3 Yangtze Delta Institute of Infrastructure,Hangzhou 310005,China

Online:2025-11-15 Published:2025-11-10
Supported by:
Beijing Natural Science Foundation of China(9244040).

摘要/Abstract

摘要： 为保障公路工程施工人员作业安全,减少施工安全事故,实时检测施工人员是否佩戴安全帽已成为重要的安全监管手段。公路工程施工具有点多、线长、面广的特点,且面临穿山越岭、跨江跨河等复杂地势,传统固定摄像头的覆盖存在局限性,且成本较高。无人机作为一种灵活、低成本且具备高可视性的影像采集工具,能够有效弥补这一不足,特别适用于传统手段难以覆盖的施工现场高风险区域。针对基于无人机采集图像的安全帽检测,在光照变化、目标尺度和形状变化较大的情况下容易出现误检、漏检的问题,提出了一种基于改进扩展差分高斯(XDOG)的YOLOv5安全帽实时检测算法。针对复杂施工环境中安全帽与背景或其他物体难以区分的问题,XDOG模块通过提取图像的边缘信息,增强了待检测安全帽的结构与细节特征。随后,差分高斯结果通过归一化和非线性激活处理,消除了环境中的亮度变化和噪声干扰。为了与YOLOv5网络兼容,采用1×1卷积层调整增强后的特征图通道数,并通过残差连接与原始图像特征进行融合,从而提高了网络的鲁棒性和准确性。实验结果表明,相较于传统的YOLOv5和YOLOx等模型,XDOG-YOLOv5在mAP@50和mAP@50-95等指标上均有显著提升,显著提高了施工作业人员安全帽检测的精度。

关键词: 安全帽检测, YOLOv5, 差分高斯, 公路工程, 无人机

Abstract: To ensure the safety of highway engineering construction personnel and reduce safety risks during the construction process,real-time detection of helmet usage has become a critical safety supervision method.Highway projects are characterized by numerous,long,and wide construction sites,often involving complex terrains such as mountain ranges and rivers.Traditional fixed-camera coverage has limitations and high costs.Drones,as flexible,low-cost,and highly visible image acquisition tools,can effectively address these challenges,especially in high-risk areas that are difficult to cover with traditional methods.This paper proposes a real-time helmet detection algorithm based on an improved eXtended Difference of Gaussians(XDOG) and YOLOv5,aiming to solve the issues of misdetection and missed detection under variable lighting conditions,scale,and shape changes in images captured by drones.In complex construction environments,the features of safety helmets are often hard to distinguish from backgrounds or other objects.The XDOG module is introduced to enhance edge information in images,thereby highlighting the structural and detailed features of helmets to be detected.The difference-of-Gaussians results are further normalized and non-linearly activated to eliminate the effects of lighting variation and noise interference in construction environments.To ensure compatibility with the YOLOv5 network,the algorithm uses a 1×1 convolution layer to adjust the number of channels in the enhanced feature maps,and a residual connection is used to fuse the enhanced feature maps with the input image,thereby improving the robustness and accuracy of the network.Experimental results show that compared to traditional YOLOv5 and YOLOx models,the XDOG-YOLOv5 significantly improves detection accuracy,with notable gains in mAP@50 and mAP@50-95,demonstrating its effectiveness in real-time helmet detection for construction personnel.

Key words: Safety helmet detection, YOLOv5, Difference of Gaussians, Roadway engineering, Unmanned aerial vehicle

中图分类号:

TP391

文明, 吴兴堂, 尚宇豪, 甄键, 于富才. 基于无人机检视的公路工程施工人员安全帽佩戴实时检测算法[J]. 计算机科学, 2025, 52(11A): 250100047-7. https://doi.org/10.11896/jsjkx.250100047

WEN Ming, WU Xingtang, SHANG Yuhao, ZHEN Jian, YU Fucai. Real-time Helmet Detection Algorithm for Roadway Engineering Construction Based on UAV Visual Inspection[J]. Computer Science, 2025, 52(11A): 250100047-7. https://doi.org/10.11896/jsjkx.250100047

参考文献

[1]YANG Y,PEI Z Z,CHEN Z,et al.Study on freeway traffic safety evaluation based on improved system engineering decision theory[J].Journal of Beijing Jiaotong University,2022,46(3):34-48.
[2]FENG P,QIAN Y R,FAN Y Y,et al.Safety helmet detection al-gorithm based on the improved Cascade R-CNN[J].Microelectronics & Computer,2024,41(1):63-73.
[3]GUO W,FAN Y G,LI X Z,et al.Detection of Nonstandard Wearing of Safety Helmet Based on Improved YOLO[J].Ordnance Industry Automation,2024,43(5):33-36,42.
[4]WANG Z B,MA H,FENG J L,et al.Helmet Detection Algorithm Based on CE-YOLOv5s[J].Computer and Modernization,2024(4):55-59,98.
[5]ZHANG G P,ZHOU J Z,MA G C,et al.Lightweight safety helmet wearing detection algorithm of improved YOLOv8[J].Electronic Measurement Technology,2024,47(17):147-154.
[6]LI Z S,XIE W Q,ZHANG L Z,et al.TowardEfficient Safety Helmet Detection Based on YoloV5 With Hierarchical Positive Sample Selection and Box Density Filtering[J].IEEE Transactions on Instrumentation and Measurement,2022,71:1-14.
[7]WAN J C,LIANG B C.Design and Implementation of an Intelligent Motorcycle Helmet for Large Vehicle Approach Intimation[J].IEEE Sensors Journal,2019,19(10):3882-3892.
[8]PATEL K,PATEL V,PRAJAPATI V,et al.Safety Helmet De-tection Using YOLO V8[C]//2023 3rd International Confe-rence on Pervasive Computing and Social Networking(ICPCSN).2023:22-26.
[9]ZHAO L,ZHANG D,LIU Y,et al.Improved YOLOv5s Network for Multi-scale safety Helmet Detection [C]//2022 11th International Conference on Communications,Circuits and Systems(ICCCAS).2022:262-266.
[10]WANG B,XIONG H,LIU L.Safety helmet wearing recognition based on improved YOLOv4 algorithm[C]//2022 IEEE 6th Information Technology and Mechatronics Engineering Conference(ITOEC).2022:1732-1736.
[11]AI X,CHEN C,WANG Y.Multi-Scene Safety Helmet Detection with Multi-Scale Spatial Attention Feature[C]//2021 7th IEEE International Conference on Network Intelligence and Di-gital Content(IC-NIDC).2021:21-25.
[12]CHEN L,MAO Y,ZHANG H.Improved the Detection Algo-rithm of Safety Helmet Wearing Based on YOLOv8[C]//2023 3rd International Conference on Electronic Information Engineering and Computer Communication(EIECC).2023:1-4.
[13]QU H,LOU S,WANG X.Safety Helmet Wearing Detection Algorithm Based on Improved Yolov5s[C]//2023 5th International Conference on Artificial Intelligence and Computer Applications(ICAICA).2023:37-40.
[14]NING S,HAN X.An improved YOLOv8-based safety helmetwearing detection algorithm[C]//2024 7th International Conference on Computer Information Science and Application Technology(CISAT).2024:75-79.
[15]XU K,LI X Z,ZHANG L,et al.Safety Helmet Wearing Detection Algorithm for Distribution Network Construction in Natural Scenarios[J].Computer Engineering and Applications,2024,60(8):320-328.
[16]ZHENG C W,LIN H.YOLOv5 Helmet Wearing DetectionMethod Based on Swin Transformer[J].Computer Measurement & Control,2023,31(3):15-21.
[17]DU Q,YANG S J,GUO Q P.Intelligent detection method of working personnel wearing safety helmets in underground mine[J].Journal of Mine Automation,2023,49(7):134-140.
[18]LI Z F,FENG S Y,GUO J,et al.Lightweight safety helmet wearing detection fusing coordinate attention and multiscale feature[J].Journal of Mine Automation,2023,49(11):151-159.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于无人机检视的公路工程施工人员安全帽佩戴实时检测算法

Real-time Helmet Detection Algorithm for Roadway Engineering Construction Based on UAV Visual Inspection

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

Metrics

本文评价

推荐阅读 0