Related Experiment Video
Updated: Jul 8, 2025
08:15
Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
Published on: March 28, 2025
503
Comparison of Wrist and Forearm EMG for Multi-day Biometric Authentication.
Summary
Wrist electromyography (EMG) signals show comparable performance to forearm signals for daily authentication. Long-term studies indicate wrist EMG is promising for reliable, continuous biometric security in wearable devices.
Area of Science:
- Biometrics
- Human-Computer Interaction
- Wearable Technology
Background:
- Electromyography (EMG) is a novel biometric trait offering dual security (biometrics and knowledge).
- Wrist-based EMG is ideal for consumer authentication via smartwatches and fitness trackers.
- Current EMG biometrics often rely on forearm data, limiting wearable integration and long-term robustness.
Purpose of the Study:
- To compare the authentication performance of wrist versus forearm EMG signals.
- To evaluate performance in both within-day and cross-day scenarios.
- To assess the suitability of wrist EMG for long-term authentication applications.
Main Methods:
- Utilized the open GRABMyo dataset containing forearm and wrist EMG data.
- Collected data from 43 participants over three distinct days (Days 1, 8, 29).
- Analyzed data for within-day and cross-day authentication Equal-error rates (EER).
Main Results:
- Wrist EMG achieved comparable within-day authentication performance to forearm EMG.
- Cross-day analysis showed higher EER for wrist EMG compared to forearm EMG.
- Wrist EMG demonstrated a low median EER (<0.1) in cumulative cross-day analysis.
Conclusions:
- Wrist EMG signals are a viable option for daily biometric authentication.
- Further research is needed to improve cross-day robustness for long-term applications.
- Wrist EMG holds significant promise for continuous, long-term authentication in wearable devices.
