Tuesday, May 2025

04:00 PM - 04:20 PM

Room: LL20D

Session: Artificial Intelligence for Emerging Technologies and Applications

Thumb Gesture Recognition Method Using Wrist EMG Signals with a Machine Learning Algorithm

Student

• Yu Sheng Zeng

  Novatek Microelectronics Corp.

Description:

This paper explores using multiple electrodes on the back of a smartwatch at the wrist to capture EMG signals and employs machine learning methods to detect various thumb movements. The study results show that even with a limited number of electrodes placed only at the wrist on the back of the watch, appropriate machine learning techniques can still achieve a recognition accuracy of around 99% for thumb movements.