Presentation 2023-03-13
sEMG-based Gesture Authentication for Smartwatch
Akihisa Watanabe, Ryo Iijima, Tatsuya Mori,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We propose a gesture-based user recognition system using sEMG signals for smartwatches. We designed the system to realize sEMG authentication with an inexpensive 1-channel sEMG sensor, whereas expensive medical devices with 32 or more channels have been used in the previous researches. We conducted preprocessing, feature extraction, machine learning model construction, and evaluation of 11 types of gestures for 19 experimental participants. We conducted a user study with 19 participants on 11 types of gestures to collect sEMG data, and conducted preprocessing, feature extraction, machine learning model construction, and evaluation. The basic evaluation showed that the models can be constructed with high accuracy (AUC=95%, EER=11%), and that the percentage reduction in AUC by robustness evaluation is within 1-2%, indicating that the models are robust against attacks assumed in a real-world scenario.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Biometrics / Bio-signals / surface Electromyography (sEMG) / User recognition by gesture
Paper # ICSS2022-50
Date of Issue 2023-03-06 (ICSS)

Conference Information
Committee ICSS / IPSJ-SPT
Conference Date 2023/3/13(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Okinawaken Seinenkaikan
Topics (in Japanese) (See Japanese page)
Topics (in English) Security, Trust, etc.
Chair Katsunari Yoshioka(Yokohama National Univ.)
Vice Chair Takahiro Kasama(NICT) / Akira Yamada(KDDI labs.)
Secretary Takahiro Kasama(Okayama Univ.) / Akira Yamada(Mitsubishi Electric)
Assistant Takeshi Sugawara(Univ. of Electro-Comm.) / Yo Kanemoto(NTT)

Paper Information
Registration To Technical Committee on Information and Communication System Security / Special Interest Group on Security Psychology and Trust
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) sEMG-based Gesture Authentication for Smartwatch
Sub Title (in English)
Keyword(1) Biometrics
Keyword(2) Bio-signals
Keyword(3) surface Electromyography (sEMG)
Keyword(4) User recognition by gesture
1st Author's Name Akihisa Watanabe
1st Author's Affiliation Waseda University(Waseda Univ.,)
2nd Author's Name Ryo Iijima
2nd Author's Affiliation Waseda University(Waseda Univ.,)
3rd Author's Name Tatsuya Mori
3rd Author's Affiliation Waseda University(Waseda Univ.,)
Date 2023-03-13
Paper # ICSS2022-50
Volume (vol) vol.122
Number (no) ICSS-422
Page pp.pp.13-18(ICSS),
#Pages 6
Date of Issue 2023-03-06 (ICSS)