Individual and Finger Posture Identification Using Fiber Bragg Grating Sensors
IEEE Sensors Journal · 2026-08 · 26 (15) · 22438-22448
DOI: 10.1109/jsen.2026.3701533 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Sensors Journal, 2026-08. Verfasst von Walter O. C. Flores und 7 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 46 Literaturangaben. Gefördert von: Coordenação de Aperfeiçoamento de Pessoal de Nível Superior, Financiadora de Estudos e Projetos (FINEP), Fundação Araucária, Multi-User Photonics Facility of Universidade Tecnológica Federal do Paraná Campus Curitiba.
Abstract (Originalsprache)
This work presents a forearm sensing scheme for volunteer and active-finger identification from repeated individual finger-extension movements using two elastic bracelets, each containing one PDMS-encapsulated FBG. The two FBG sensors were positioned at 50% and 35% of the ulnar length measured from the wrist, and their distinct Bragg wavelengths enabled simultaneous interrogation. In each sensor, the embedded FBG was integrated into the textile structure. Differences in forearm muscle deformation associated with finger extension were mechanically transferred to the bracelets and detected as Bragg-wavelength shifts. Experiments were conducted with ten volunteers, and Bragg-wavelength time-series data were acquired at 50 Hz. The resulting signals were then processed offline by filtering, baseline correction, and windowed segmentation before feature extraction. Classification was performed to identify the active finger among five individual finger-extension movements and to identify the volunteer. A multioutput k-nearest neighbors (kNN) classifier, optimized by fivefold cross-validation, achieved test-set accuracy values of 0.911 for finger identification, 0.915 for volunteer identification, 0.913 for the average of both tasks, and 0.894 for simultaneous finger-and-volunteer identification. The sensing scheme was also characterized under controlled elongation, yielding a sensitivity of $\text {S}= {\mathrm {3.103~{\,}}}\text {pm}/\mu \text {m}$ , corresponding to an equivalent strain sensitivity of $ {\mathrm {0.310~{\,}}}\text {pm}/\mu \varepsilon $ , with linearity of $\text {R}^{2}=0.999$ . Under the controlled laboratory conditions adopted in this study, the results indicate that the proposed two-sensor forearm system can support offline identification of the active finger and the volunteer.
Förderung
- Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
- Financiadora de Estudos e Projetos (FINEP), Fundação Araucária
- Multi-User Photonics Facility of Universidade Tecnológica Federal do Paraná Campus Curitiba
Zitieren
APA
Flores, W. O. C., Junior, P. V. R., Lourencao, J. P. M., Muller, M., Fabris, J. L., Lazzaretti, A. E., Junior, J. J. A. M., & Dreyer, U. J. (2026). Individual and Finger Posture Identification Using Fiber Bragg Grating Sensors. IEEE Sensors Journal, 26(15), 22438-22448. https://doi.org/10.1109/jsen.2026.3701533
Vancouver
Flores WOC, Junior PVR, Lourencao JPM, Muller M, Fabris JL, Lazzaretti AE, et al. Individual and Finger Posture Identification Using Fiber Bragg Grating Sensors. IEEE Sensors Journal. 2026;26(15):22438-22448. doi:10.1109/jsen.2026.3701533
BibTeX
@article{Flores2026,
author = {Flores, Walter O. C. and Junior, Pericles V. Rialto and Lourencao, Joao P. M. and Muller, Marcia and Fabris, Jose L. and Lazzaretti, Andre E. and Junior, Jose J. A. Mendes and Dreyer, Uilian J.},
title = {Individual and Finger Posture Identification Using Fiber Bragg Grating Sensors},
journal = {IEEE Sensors Journal},
year = {2026},
volume = {26},
number = {15},
pages = {22438--22448},
doi = {10.1109/jsen.2026.3701533}
}
EUVIMED