福建电脑2025，Vol.41Issue(3)：24-29,6.DOI:10.16707/j.cnki.fjpc.2025.03.006
改进YOLOv5s算法的头盔佩戴实时检测系统
A Real-time Helmet Wearing Detection System based on Deep Learning-improved YOLOv5s Algorithm
摘要
Abstract
To solve the problem of low detection efficiency caused by factors such as weather and angle in electric vehicle helmet detection,this paper proposes a helmet wearing real-time detection system based on an improved YOLOv5 algorithm.The system design integrates high-definition camera video stream input,efficient algorithm processing,and real-time detection result output.By adopting techniques such as data augmentation,structural optimization,and loss function adjustment,the system significantly improves recognition accuracy and speed in complex environments.The test results show that the detection accuracy of the optimized YOLOv5s algorithm helmet detection system has improved by 1.1 percentage points compared to the original YOLOv5 algorithm.关键词
头盔佩戴检测/电动车安全/YOLOv5算法Key words
Helmet Wearing Detection/Electric Vehicle Safety/YOLOv5 Algorithm分类
计算机与自动化引用本文复制引用
毛雅,余锦,李倩南,吴青周..改进YOLOv5s算法的头盔佩戴实时检测系统[J].福建电脑,2025,41(3):24-29,6.基金项目
本文得到黄山学院2023年度大学生创新创业训练计划"基于深度学习——改进YOLOv5s算法的头盔佩戴实时检测系统"(No.202310375010)、安徽省创新型省份建设补助资金专项资助项目(No.2020xzx004)、安徽省高等学校科学研究重点项目(No.2023AH051371)资助. （No.202310375010）
