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

Correction: Dilawar et al. Rhizofungus <i>Aspergillus terreus</i> Mitigates Heavy Metal Stress-Associated Damage in <i>Triticum aestivum</i> L. <i>Plants</i> 2024, <i>13</i>, 2643.

Plants (Basel, Switzerland)·2026
Same author

Magnetic Particle Imaging for Pulmonary Applications: Technological Advances, Biological Insights, and Clinical Translation.

Bioengineering (Basel, Switzerland)·2026
Same author

Machine learning empowered proactive content caching in O-RAN.

Scientific reports·2026
Same author

Production of anti-inflammatory and antidiabetic oligosaccharides from okra mucilage through one-step microbial fermentation.

PloS one·2026
Same author

Unlocking Grass Stress Resistance: Fungal Endophyte-Mediated Pathogen Recognition and RNA Regulation.

International journal of molecular sciences·2026
Same author

Robust multi-agent reinforcement learning framework for intelligent PV-integrated smart energy systems under uncertainty.

Scientific reports·2026

相关实验视频

Updated: Jun 21, 2025

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

2.7K

基于人工智能模型的可穿戴传感器用于识别人类活动.

Mohammed Alarfaj1, Azzam Al Madini1, Ahmed Alsafran1

  • 1Department of Electrical Engineering, College of Engineering, King Faisal University, Al-Ahsa, Saudi Arabia.

Frontiers in artificial intelligence
|July 17, 2024
PubMed
概括

这项研究引入了一种新型的人类活动识别 (HAR) 方法,使用专门的卷积神经网络 (CNN) 用于单个传感器. 与传统分类器相比,该方法显著提高了检测人类运动模式的准确性.

关键词:
气压计气压计气压计卷积神经网络是一种卷积神经网络.落检测系统 落检测系统 落检测系统人体运动 人体运动惯性测量单位是一种惯性测量单位.机器学习是机器学习.传感器网络 传感器网络传感器 传感器 传感器

更多相关视频

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
06:49

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

8.9K
Assessing the Accuracy of Fitness Smartwatch Data for Cardiovascular and Physical Activity Monitoring: A Validation Study in Digital Health
05:51

Assessing the Accuracy of Fitness Smartwatch Data for Cardiovascular and Physical Activity Monitoring: A Validation Study in Digital Health

Published on: February 21, 2025

416

相关实验视频

Last Updated: Jun 21, 2025

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

2.7K
Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
06:49

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

8.9K
Assessing the Accuracy of Fitness Smartwatch Data for Cardiovascular and Physical Activity Monitoring: A Validation Study in Digital Health
05:51

Assessing the Accuracy of Fitness Smartwatch Data for Cardiovascular and Physical Activity Monitoring: A Validation Study in Digital Health

Published on: February 21, 2025

416

科学领域:

  • 生物医学工程 生物医学工程
  • 机器学习 机器学习
  • 可穿戴技术可穿戴技术

背景情况:

  • 人类运动检测对于医学,医疗保健和体育炼应用至关重要.
  • 现有的方法面临着各种传感器数据形状的挑战,并实现高精度.
  • 卷积神经网络 (CNN) 提供了在传感器数据中复杂的模式识别的潜力.

研究的目的:

  • 开发一种新的人类活动识别 (HAR) 系统,使用针对单个传感器类型量身定制的CNN.
  • 通过有效地处理来自加速度计,陀螺仪和气压计的各种数据形状来提高HAR的准确性.
  • 将拟议的基于CNN的方法与标准支持向量机 (SVM) 分类器的性能进行比较.

主要方法:

  • 针对每个传感器类型 (加速计,陀螺仪,气压计) 设计了个别的CNN模型,以捕捉传感器特定的特征.
  • 使用后期融合技术,将来自单个CNN模型的预测结合起来,以进行全面的活动分类.
  • 拟议的CNN方法与传统的SVM分类器进行了基准测试,使用一个对其他方法.

主要成果:

  • 晚期融合的CNN模型实现了显著更高的准确性,验证准确率为99.35%,最终测试准确率为94.83%.
  • 传统的SVM分类器的准确度较低,达到87.07% (验证) 和83.10% (最终测试).
  • 结合多个传感器,气压计和过算法,明显改善了人类运动模式识别.

结论:

  • 拟议的基于CNN的HAR方法,利用传感器特定的模型和晚期融合,大大优于传统方法.
  • 根据单个传感器数据特征定制CNN架构是提高HAR准确性的关键.
  • 这种先进的技术为各种应用中精确检测人类运动提供了一个有希望的解决方案.