摘要
近年来,电动摩托车事故数量攀升,未佩戴头盔成为造成驾乘人员受伤或死亡的主要原因。为了促进市民能够树立安全骑行佩戴头盔的意识,本文研究并设计了基于YOLOv5目标检测的头盔智能检测系统。该系统分为头盔目标检测端和后台管理端。头盔目标检测端首先通过定点路况数据采集获得头盔佩戴数据集,然后使用预处理后的数据集训练YOLOv5算法模型,最后训练模型用于检测道路骑行者头盔佩戴情况。后台管理端实现了视频的回放和违规数据的处理与查看。该系统基于人工智能目标检测技术,解决了交管部门对非机动车进行监管时耗时、耗力、漏检的问题,具有重要的现实意义与实用价值。
In recent years,the number of electric motorcycle accidents has increased,and failure to wear a helmet has become the main reason for drivers and passengers injury or death.In order to promote citizens'awareness of wearing helmets,this article researches and designs a helmet intelligent detection system based on YOLOv5 target detection.This system includes helmet target detection and background management terminals.First of all,the helmet target detection side obtains the helmet wearing a data set,and then use the pre-processed data set to train the YOLOV5 algorithm model.Finally,the training model is used to detect the road rider's helmet wearing.The background management side implements the release of video and the processing and viewing of illegal data.Based on artificial intelligence target testing technology,this system solves the problem of time,labor-consuming,and missed inspections when the traffic management department is supervised by non-motorized vehicles.It has important practical significance and practical value.
作者
张雪华
谭丽娜
刘喆
ZHANG Xuehua;TAN Lina;LIU Zhe(Shandong College of Electronic Technology,Jinan 250200,China)
出处
《电子测试》
2023年第2期87-91,共5页
Electronic Test
基金
2021年度山东省职业教育教学改革研究项目(2021217)资助。
