基于Y0L0v5的安全帽佩戴检测系统研究

打开文本图片集

中图分类号：TP391.4 文献标识码：A 文章编号：2096-4706（2026）03-0159-06

Abstract： With theaccelerated urbanization processand the vigorous development of the construction industry， the safety management problematconstructionsites has become increasingly prominent.Traditional safety supervisionmethods arelow inefficiencyanddificulttoadapttothecomplexandever-changingconstructionenvironments.Toadressthisissue，this paper proposes a safety helmet wearing detection system for construction sites based on the YOLOv5 algorithm，which aims to automaticallidentify workers'safety helmet wearing status byvirtueofDeep Learing technologyandthus improve the intellgencelevelofsafetymanagementatconstructionsites.Thesystem isdevelopedwiththePythonlanguageandintegrates functional modulessuchasimagedetection，real-timevideostreammonitoringand historicalvideodetection，withauserfriendlyiterfaceandaconvenientoperationalexperience.Throughdataannotation，modeltrainingand prediction，exprimetal results show thatthesystemcan acuratelyidentifythe safety helmet wearing status incomplexconstruction environments and effectively improve the safety supervision level at construction sites.

Keywords：YOLOv5;safety helmet detection; data annotation; model training; model prediction

0 引言

随着建筑行业的快速发展，施工项目的数量急剧增加，施工现场的安全管理问题日益严峻。（剩余5487字）