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

Complete mitochondrial genome of the Woodland Brown, <i>Lopinga achine</i> Scopoli, 1763 (Nymphalidae: Satyrinae) and its phylogenetic analysis.

Mitochondrial DNA. Part B, Resources·2022
Same author

Heptaketides from the endophytic fungus <i>Pleosporales</i> sp. F46 and their antifungal and cytotoxic activities.

RSC advances·2022
Same author

Strategies comparison in response to the two waves of COVID-19 in the United States and India.

International journal for equity in health·2022
Same author

Stabilization of flavin mononucleotide by capturing its "tail" with porous organic polymers for long-term photocatalytic degradation of micropollutants.

Journal of hazardous materials·2022
Same author

Quantitative Evaluation of Retinal Microvascular Abnormalities in Patients With Type 2 Diabetes Mellitus Without Clinical Sign of Diabetic Retinopathy.

Translational vision science & technology·2022
Same author

Deep insights into the gut microbial community of extreme longevity in south Chinese centenarians by ultra-deep metagenomics and large-scale culturomics.

NPJ biofilms and microbiomes·2022

相关实验视频

Updated: Jul 17, 2025

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
08:15

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Published on: March 28, 2025

627

基于多维传感数据和深度学习算法的运动识别研究.

Jia-Gang Qiu1, Yi Li1, Hao-Qi Liu1

  • 1Institute of Artificial Intelligence in Sports, Capital University of Physical Education and Sports, Beijing 100191, China.

Mathematical biosciences and engineering : MBE
|September 7, 2023
PubMed
概括
此摘要是机器生成的。

这项研究比较了机器学习和深度学习的动作识别,使用惯性测量单元 (IMU) 数据. 深度学习,特别是动态神经网络 (DNN),在识别人类运动方面表现出卓越的表现.

关键词:
经典的机器学习算法算法.深度学习算法 深度学习算法在IMU,IMU是IMU.每天的行动,每天的行动.动作识别 运动识别 运动识别

更多相关视频

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

3.8K
Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

10.7K

相关实验视频

Last Updated: Jul 17, 2025

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
08:15

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Published on: March 28, 2025

627
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

3.8K
Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

10.7K

科学领域:

  • 人与计算机的互动.
  • 生物医学工程 生物医学工程
  • 机器学习是机器学习.

背景情况:

  • 精确的人类运动识别对于身体康复,老年护理和运动感应游戏等应用至关重要.
  • 惯性测量单元 (IMU) 传感器为捕获多维运动数据提供了一种可行的方法.
  • 评估各种机器学习和深度学习算法对于优化运动识别系统至关重要.

研究的目的:

  • 为了比较经典机器学习算法 (随机森林,K-最近邻居,决策树) 和深度学习模型 (动态神经网络,卷积神经网络,循环神经网络) 对人类运动识别的性能.
  • 为了确定IMU传感器在区分日常活动时的最佳身体位置.
  • 评估将IMU数据与用于运动识别的不同算法相结合的有效性.

主要方法:

  • 采用了六种算法:随机森林 (RF),K-最近邻居 (KNN),决策树 (DT),动态神经网络 (DNN),卷积神经网络 (CNN) 和循环神经网络 (RNN).
  • 从七个身体部位收集的利用惯性测量单元 (IMU) 数据.
  • 分析和比较不同算法和传感器放置的识别率.

主要成果:

  • 经典机器学习算法表现相似,随机森林实现了最高的识别率96.67%.
  • 深度学习模型表现出显著的性能差异,动态神经网络 (DNN) 实现了最高的97.71%的速度.
  • 腰部被确定为IMU传感器区分日常活动的最佳位置.

结论:

  • 深度学习算法,特别是DNN,与多维传感器数据相结合,与经典机器学习相比,在运动识别方面提供了更高的性能.
  • 树结构模型在经典机器学习方法中仍然具有竞争力.
  • IMU传感器和深度学习算法的集成为先进的运动识别应用提供了坚实的基础.