Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Bridging the Performance Gap: The Impact of Grasping Behavior and Riemannian Geometry on Prosthetic Intent Recognition.

IEEE transactions on bio-medical engineering·2026
Same author

Spatial averaging method based on adaptive weight for imaging photoplethysmography.

Journal of biomedical optics·2023
Same author

A real-time heart rate estimation framework based on a facial video while wearing a mask.

Technology and health care : official journal of the European Society for Engineering and Medicine·2022
Same author

Towards an effective framework for integrating patient-reported outcomes in electronic health records.

Digital health·2022
Same author

Activation of Cholinergic Anti-Inflammatory Pathway Ameliorates Cerebral and Cardiac Dysfunction After Intracerebral Hemorrhage Through Autophagy.

Frontiers in immunology·2022
Same author

A flexible functional module to regulate ultraviolet optical nonlinearity for achieving a balance between a second-harmonic generation response and birefringence.

Chemical science·2022

Related Experiment Video

Updated: May 24, 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

361

sEMG-Based Gesture Recognition via Multi-Feature Fusion Network.

Zekun Chen, Xiupeng Qiao, Shili Liang

    IEEE Journal of Biomedical and Health Informatics
    |March 3, 2025
    PubMed
    Summary
    This summary is machine-generated.

    A novel multi-feature fusion network (MFF-Net) improves sparse surface electromyography (sEMG) gesture recognition by integrating time, frequency, and spatial features. This model enhances accuracy and generalizes well to small datasets, including for amputees.

    More Related Videos

    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

    2.5K
    Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
    09:41

    Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

    Published on: April 21, 2023

    1.5K

    Related Experiment Videos

    Last Updated: May 24, 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

    361
    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

    2.5K
    Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
    09:41

    Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

    Published on: April 21, 2023

    1.5K

    Area of Science:

    • Biomedical Engineering
    • Machine Learning
    • Human-Computer Interaction

    Background:

    • Sparse surface electromyography (sEMG) based gesture recognition faces challenges with limited feature information and poor generalization, especially for small sample sizes.
    • Existing methods struggle to effectively extract and integrate rich feature information from sparse sEMG signals.

    Purpose of the Study:

    • To propose a multi-feature fusion network (MFF-Net) to enhance feature richness and improve generalization for sparse sEMG gesture recognition.
    • To address the limitations of insufficient feature information and poor performance on small sample datasets in sEMG recognition.

    Main Methods:

    • Developed MFF-Net incorporating Long Short-Term Memory (LSTM) and an attention mechanism.
    • Constructed three sub-networks focusing on time, frequency, and spatial domains to enhance features.
    • Employed feature splicing and stacking to strengthen inter-time and inter-channel information.

    Main Results:

    • Achieved state-of-the-art classification accuracy of 92.47% on 18 gesture recognition tasks from NinaPro DB3 and DB7 datasets.
    • Demonstrated significant improvement in gesture recognition for small sample amputee data, increasing rates from 60.35% to 84.93% (DB7) and 71.84% to 82.00% (DB3).
    • Ablation experiments confirmed the model's effectiveness in feature processing and overall performance enhancement.

    Conclusions:

    • The proposed MFF-Net effectively enriches sparse sEMG features, leading to superior gesture recognition accuracy.
    • The model exhibits strong generalization capabilities and applicability to transfer learning for amputee gesture recognition tasks with limited data.
    • MFF-Net offers a promising solution for overcoming data scarcity and improving the performance of sEMG-based human-computer interfaces.