电脑与信息技术2024,Vol.32Issue(6) :6-11.

基于轻量型YOLOv7-TMC网络的疲劳驾驶状态检测方法

Fatigue Driving State Detection Method Based on Lightweight YOLOv7-TMC Network

马明
电脑与信息技术2024,Vol.32Issue(6) :6-11.

基于轻量型YOLOv7-TMC网络的疲劳驾驶状态检测方法

Fatigue Driving State Detection Method Based on Lightweight YOLOv7-TMC Network

马明1
扫码查看

作者信息

  • 1. 沈阳航空航天大学 电子信息工程学院,辽宁 沈阳 110136
  • 折叠

摘要

为了在移动端更方便地部署,需要进一步实现疲劳驾驶检测模型的轻量化,因此,提出一种新的轻量级网络算法YOLOv7-TMC,即YOLOv7-Tiny结合MobileNet和CBAM,应用于疲劳驾驶状态检测.实验结果表明,提出的算法在精确度上有明显的提升,同时检测速度更快,能够快速准确地捕捉到驾驶员的疲劳驾驶行为,为后续的智能驾驶提供了检测精度更高、速度更快的行车辅助,具有重要的应用价值.

Abstract

In order to be more conveniently deployed on mobile devices,the lightweighting of fatigue driving detection models needed to be further implemented.Therefore,a new lightweight network algorithm,YOLOv7-TMC,which combines YOLOv7-Tiny with MobileNet and CBAM,was proposed for fatigue driving state detection.Experimental results indicated that the algorithm proposed in this paper had a significant improvement in accuracy and a faster detection speed,enabling the rapid and accurate capture of fatigue driving behavior by drivers.This provided a driving assistance system with higher detection accuracy and faster speed for subsequent intelligent driving,possessing important application value.

关键词

疲劳驾驶/深度学习/YOLOv7/轻量化

Key words

fatigue driving/deep learning/YOLOv7/light weight

引用本文复制引用

出版年

2024
电脑与信息技术
中国电子学会,湖南省电子研究所

电脑与信息技术

影响因子:0.256
ISSN:1005-1228
段落导航相关论文