Sensor Topology-Aware Three-Branch Fusion for sEMG Gesture Recognition
Sensors · 26.07.2026 · 26 (15) · 4733
DOI: 10.3390/s26154733 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 26.07.2026. Verfasst von Luoqi Cui und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Mechanical Engineering, Xi'an University of Technology, Xi'an 710048, China. Hauptthemen laut MEDLINE: Gestures, Electromyography, Pattern Recognition, Automated. Die Arbeit nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Xi'an University of Technology, Xi’an University of Technology.
Abstract (Originalsprache)
Surface electromyography (sEMG) is increasingly used for gesture recognition in prosthetics, rehabilitation, and human–computer interaction. Existing architectures typically force heterogeneous sEMG features into a shared latent representation, limiting their ability to capture complementary temporal, frequency-domain, and inter-electrode spatial dependencies. To better exploit these features, this paper proposes a three-branch fusion network. Unlike many existing multi-branch methods, the proposed network explicitly models the ring arrangement of armband electrodes, capturing the adjacency information in the sensor topology that linear channel representations ignore. The temporal and spectral branches use a compact multi-scale residual structure, so this topology branch is added while maintaining modest model complexity. A reliability-aware routing mechanism then adaptively assigns fusion weights to the three branches for each sample. On NinaPro DB5 Exercise B (eight-channel lower armband), the method reaches 83.99% under subject-dependent training and 85.66% under transfer learning, exceeding prior transfer learning approaches under matched conditions. Ablation experiments confirm that the three branches contribute non-redundant information and that adaptive fusion outperforms fixed combinations. The architecture also generalizes to MyoArmbandDataset under a subject-adaptive transfer learning protocol without dataset-specific hyperparameter retuning, indicating potential for wearable gesture interfaces, rehabilitation, and prosthetic control.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Mechanical Engineering, Xi'an University of Technology, Xi'an 710048, China.
- College of Science and Engineering, James Cook University, Townsville, QLD 4815, Australia.
- Faculty of Computing and Information Technology (FCIT), Sohar University, Sohar 311, Oman.
ORCID-Profile: Luoqi Cui, Yong Liu, Mohammad Reza Chalak Qazani
Förderung
- Xi'an University of Technology
- Xi’an University of Technology
Zitieren
APA
Cui, L., Liu, Y., Abdar, H. F., Gao, X., Yang, M., & Qazani, M. R. C. (2026). Sensor Topology-Aware Three-Branch Fusion for sEMG Gesture Recognition. Sensors, 26(15), 4733. https://doi.org/10.3390/s26154733
Vancouver
Cui L, Liu Y, Abdar HF, Gao X, Yang M, Qazani MRC. Sensor Topology-Aware Three-Branch Fusion for sEMG Gesture Recognition. Sensors. 2026;26(15):4733. doi:10.3390/s26154733
BibTeX
@article{Cui2026,
author = {Cui, Luoqi and Liu, Yong and Abdar, Hadi Fathollahi and Gao, Xinqin and Yang, Mingshun and Qazani, Mohammad Reza Chalak},
title = {Sensor Topology-Aware Three-Branch Fusion for sEMG Gesture Recognition},
journal = {Sensors},
year = {2026},
volume = {26},
number = {15},
pages = {4733},
doi = {10.3390/s26154733}
}
EUVIMED