Lightweight Deep Learning for sEMG-Based Fingers Position Classification and Embedded System Deployment

This paper presents a lightweight deep learning approach for classifying and tracking finger positions based on surface electromyography (sEMG) signals, designed with a perspective of its implementation in embedded systems. Unlike traditional studies that focus on static hand gestures, this research...

Full description

Saved in:
Bibliographic Details
Main Authors: Victor H. Benitez, Jesus Pacheco, Guillermo Cuamea-Cruz
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10904232/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850023473209933824
author Victor H. Benitez
Jesus Pacheco
Guillermo Cuamea-Cruz
author_facet Victor H. Benitez
Jesus Pacheco
Guillermo Cuamea-Cruz
author_sort Victor H. Benitez
collection DOAJ
description This paper presents a lightweight deep learning approach for classifying and tracking finger positions based on surface electromyography (sEMG) signals, designed with a perspective of its implementation in embedded systems. Unlike traditional studies that focus on static hand gestures, this research highlights the continuous tracking of finger positions during dynamic hand movements, such as the transition between open and closed states. A 1D Convolutional Neural Network was developed and validated using experimental data, achieving a classification accuracy of 97% in controlled scenarios. The architecture of the model balances computational efficiency and classification performance, making it deployable on resource-constrained embedded technology. This feature highlights its potential for real-time applications in prosthetics, robotics, and human-computer interaction. Although further optimization is needed for better generalization to unseen data, this study emphasizes the significance of developing deployable algorithms that excel beyond simulation environments focusing on enhancing model robustness and validating its real-time performance through hardware-based implementations. The findings indicate notable advancements in connecting sophisticated machine-learning techniques with effective embedded solutions for complex, dynamic tasks.
format Article
id doaj-art-4e0cb04d733d41f9a1e6914f3edff76f
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-4e0cb04d733d41f9a1e6914f3edff76f2025-08-20T03:01:22ZengIEEEIEEE Access2169-35362025-01-0113381813819410.1109/ACCESS.2025.354582710904232Lightweight Deep Learning for sEMG-Based Fingers Position Classification and Embedded System DeploymentVictor H. Benitez0https://orcid.org/0000-0002-3926-9352Jesus Pacheco1https://orcid.org/0000-0002-8636-5902Guillermo Cuamea-Cruz2Industrial Engineering Department, Universidad de Sonora, Hermosillo, Sonora, MexicoIndustrial Engineering Department, Universidad de Sonora, Hermosillo, Sonora, MexicoIndustrial Engineering Department, Universidad de Sonora, Hermosillo, Sonora, MexicoThis paper presents a lightweight deep learning approach for classifying and tracking finger positions based on surface electromyography (sEMG) signals, designed with a perspective of its implementation in embedded systems. Unlike traditional studies that focus on static hand gestures, this research highlights the continuous tracking of finger positions during dynamic hand movements, such as the transition between open and closed states. A 1D Convolutional Neural Network was developed and validated using experimental data, achieving a classification accuracy of 97% in controlled scenarios. The architecture of the model balances computational efficiency and classification performance, making it deployable on resource-constrained embedded technology. This feature highlights its potential for real-time applications in prosthetics, robotics, and human-computer interaction. Although further optimization is needed for better generalization to unseen data, this study emphasizes the significance of developing deployable algorithms that excel beyond simulation environments focusing on enhancing model robustness and validating its real-time performance through hardware-based implementations. The findings indicate notable advancements in connecting sophisticated machine-learning techniques with effective embedded solutions for complex, dynamic tasks.https://ieeexplore.ieee.org/document/10904232/Artificial intelligenceembedded systemsfinger position trackinglightweight deep learningreal-time classificationsEMG signal analysis
spellingShingle Victor H. Benitez
Jesus Pacheco
Guillermo Cuamea-Cruz
Lightweight Deep Learning for sEMG-Based Fingers Position Classification and Embedded System Deployment
IEEE Access
Artificial intelligence
embedded systems
finger position tracking
lightweight deep learning
real-time classification
sEMG signal analysis
title Lightweight Deep Learning for sEMG-Based Fingers Position Classification and Embedded System Deployment
title_full Lightweight Deep Learning for sEMG-Based Fingers Position Classification and Embedded System Deployment
title_fullStr Lightweight Deep Learning for sEMG-Based Fingers Position Classification and Embedded System Deployment
title_full_unstemmed Lightweight Deep Learning for sEMG-Based Fingers Position Classification and Embedded System Deployment
title_short Lightweight Deep Learning for sEMG-Based Fingers Position Classification and Embedded System Deployment
title_sort lightweight deep learning for semg based fingers position classification and embedded system deployment
topic Artificial intelligence
embedded systems
finger position tracking
lightweight deep learning
real-time classification
sEMG signal analysis
url https://ieeexplore.ieee.org/document/10904232/
work_keys_str_mv AT victorhbenitez lightweightdeeplearningforsemgbasedfingerspositionclassificationandembeddedsystemdeployment
AT jesuspacheco lightweightdeeplearningforsemgbasedfingerspositionclassificationandembeddedsystemdeployment
AT guillermocuameacruz lightweightdeeplearningforsemgbasedfingerspositionclassificationandembeddedsystemdeployment