Replay-Based Incremental Learning Framework for Gesture Recognition Overcoming the Time-Varying Characteristics of sEMG Signals
Sensors · 2024
DOI: 10.3390/s24227198 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, 2024. Verfasst von Xingguo Zhang und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Mechanical Engineering, Nantong University, Nantong 226019, China. Hauptthemen laut MEDLINE: Electromyography, Gestures, Algorithms. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Science and Technology Commission of Shanghai Municipality, Shanghai Municipal Education Commission, Nantong Science and Technology Bureau.
Abstract (Originalsprache)
Gesture recognition techniques based on surface electromyography (sEMG) signals face instability problems caused by electrode displacement and the time-varying characteristics of the signals in cross-time applications. This study proposes an incremental learning framework based on densely connected convolutional networks (DenseNet) to capture non-synchronous data features and overcome catastrophic forgetting by constructing replay datasets that store data with different time spans and jointly participate in model training. The results show that, after multiple increments, the framework achieves an average recognition rate of 96.5% from eight subjects, which is significantly better than that of cross-day analysis. The density-based spatial clustering of applications with noise (DBSCAN) algorithm is utilized to select representative samples to update the replayed dataset, achieving a 93.7% recognition rate with fewer samples, which is better than the other three conventional sample selection methods. In addition, a comparison of full dataset training with incremental learning training demonstrates that the framework improves the recognition rate by nearly 1%, exhibits better recognition performance, significantly shortens the training time, reduces the cost of model updating and iteration, and is more suitable for practical applications. This study also investigates the use of the incremental learning of action classes, achieving an average recognition rate of 88.6%, which facilitates the supplementation of action types according to the demand, and further improves the application value of the action pattern recognition technology based on sEMG signals.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Mechanical Engineering, Nantong University, Nantong 226019, China.
- School of Mechanical Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China.
ORCID-Profile: Yue Zhang
Förderung
- Science and Technology Commission of Shanghai Municipality
- Shanghai Municipal Education Commission
- Nantong Science and Technology Bureau
- Shanghai Sailing Program, China
- Shanghai University youth teacher training assistance scheme
- Nantong fundamental Science Research Projects
Zitieren
APA
Zhang, X., Li, T., Sun, M., Zhang, L., Zhang, C., & Zhang, Y. (2024). Replay-Based Incremental Learning Framework for Gesture Recognition Overcoming the Time-Varying Characteristics of sEMG Signals. Sensors. https://doi.org/10.3390/s24227198
Vancouver
Zhang X, Li T, Sun M, Zhang L, Zhang C, Zhang Y. Replay-Based Incremental Learning Framework for Gesture Recognition Overcoming the Time-Varying Characteristics of sEMG Signals. Sensors. 2024. doi:10.3390/s24227198
BibTeX
@article{Zhang2024,
author = {Zhang, Xingguo and Li, Tengfei and Sun, Maoxun and Zhang, Lei and Zhang, Cheng and Zhang, Yue},
title = {Replay-Based Incremental Learning Framework for Gesture Recognition Overcoming the Time-Varying Characteristics of sEMG Signals},
journal = {Sensors},
year = {2024},
doi = {10.3390/s24227198}
}
EUVIMED