Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
关于 JoVE
概览领导团队博客JoVE 帮助中心
作者
出版流程编辑委员会范围与政策同行评审常见问题投稿
图书馆员
用户评价订阅访问资源图书馆顾问委员会常见问题
研究
JoVE JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关实验视频

Updated: Jan 11, 2026

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

5.2K

使用先进的深度学习模型实时检测驾驶员活动.

Md Al Emran1, Md Ariful Islam1, Md Obaydullahn Khan1

  • 1Pabna University of Science and Technology, Pabna, Bangladesh.

Cognitive neurodynamics
|November 17, 2025
PubMed
概括

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序
Same author

Molecular Docking, Antitumor and Antimicrobial Activities of Extracts and Synthesized Silver Nanoparticles From <i>Piper chaba</i> Roots.

Food science & nutrition·2026
Same author

Integrative defect engineering and NiCo-layered double hydroxide decoration of W-doped BiVO<sub>4</sub>: a high-performance photoanode for efficient photoelectrochemical water splitting.

Journal of colloid and interface science·2026
Same author

Nurses' Roles in Supporting Digital Engagement and Self-Management in Adults with Type 2 Diabetes: A Scoping Review.

Nursing reports (Pavia, Italy)·2026
Same author

Valorizing Landfill Gas Condensate as an External Carbon Source for Denitrification in Sewage-Landfill Leachate Co-Treatment.

Water environment research : a research publication of the Water Environment Federation·2026
Same author

Evaluation of immune responses in cattle immunized with an experimentally developed inactivated alum adjuvanted Brucella abortus biovar 3 vaccine.

Journal of dairy science·2026
Same author

Fate and Transformation of Landfill Leachate Dissolved Organic Nitrogen and Its Implications for Estuarine Algal Growth.

ACS ES&T water·2026
此摘要是机器生成的。

一个新的AI模型准确地检测驾驶员的行为,如分心和昏昏欲睡,以提高道路安全. 这种先进的系统在实时应用中显示出高精度,旨在减少全球交通事故.

科学领域:

  • 计算机视觉 计算机视觉
  • 人工智能的人工智能
  • 机器学习 机器学习

背景情况:

  • 交通事故是全球主要的安全问题,通常是由司机的注意力不集中,睡眠和分心造成的.
  • 计算机视觉和人工智能为实时驾驶员监控系统提供了有希望的解决方案,以减轻这些风险.

研究的目的:

  • 开发和评估一种用于多类驾驶员活动分类的新型深度学习架构.
  • 将拟议模型的性能与已建立的深度学习模型进行比较,用于驾驶员行为分析.

主要方法:

  • 评估了四种深度学习模型:移动NetV2,DenseNet201,NASNetMobile和VGG19. 这四种深度学习模型包括:
  • 提出了一种独特的混合CNN-变压器架构,并增强了高效频道注意力 (ECA).
  • 该框架将七个关键的驾驶行为分类为:闭眼,开眼,危险驾驶,分心驾驶,饮酒,打哈欠和安全驾驶.

主要成果:

  • 基线模型DenseNet201和MobileNetV2实现了高的验证准确度 (分别为99.40%和99.31%).
  • 拟议的混合CNN转换器与ECA实现了近乎完美的验证准确率99.72%.
  • 该模型表现出无的概括性,在独立的测试集上达到100%的准确性,错误分类最小.
关键词:
开车时注意力分散的驾驶驾驶员行为分类 驾驶员行为分类有效的频道注意力.混合CNN变压器 - 变压器智能运输系统 智能运输系统转移学习转移学习

更多相关视频

Tactile Vibrating Toolkit and Driving Simulation Platform for Driving-Related Research
07:15

Tactile Vibrating Toolkit and Driving Simulation Platform for Driving-Related Research

Published on: December 18, 2020

4.9K

相关实验视频

Last Updated: Jan 11, 2026

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

5.2K
Tactile Vibrating Toolkit and Driving Simulation Platform for Driving-Related Research
07:15

Tactile Vibrating Toolkit and Driving Simulation Platform for Driving-Related Research

Published on: December 18, 2020

4.9K

结论:

  • 混合CNN-变压器与ECA有效地合并了CNN本地特征提取,基于注意力的精细化和变压器全球上下文建模,以实现强大高效的驾驶员监控.
  • 开发的系统显示了实时智能运输应用的巨大潜力,有助于减少交通事故和提高道路安全.