通信与信息技术Issue(1)：77-79,91,4.
施工现场安全衣帽穿戴实时检测算法研究
Research on real-time detection algorithm for safety helmet and vest wearing at construction sites
杨雪1
作者信息
- 1. 中通服咨询设计研究院有限公司,江苏 南京 210000
- 折叠
摘要
Abstract
To address the prevalent issues of high missed detection rates and frequent false alarms in safety helmet and vest detec-tion within complex construction environments,this paper proposes an improved YOLOv11n method.By introducing the RepVGG module to replace the Conv module in the Bottleneck component of the C3K2 module,the network's feature extraction capability for safety gear characteristics is enhanced.Experimental results demonstrate that compared with the original YOLOv11n model,the improved model achieves 1.9%increase in precision(P),1.5%improvement in recall(R),and 3.1%enhancement in mAP value.These findings indicate that the optimized model exhibits superior performance in detecting safety gear wearing under complex environmental conditions.关键词
安全衣帽检测/目标识别/深度学习/RepVGGKey words
Safety helmet and vest detection/Object detection/Deep learning/RepVGG分类
信息技术与安全科学引用本文复制引用
杨雪..施工现场安全衣帽穿戴实时检测算法研究[J].通信与信息技术,2026,(1):77-79,91,4.
