Related Experiment Video
Updated: Oct 10, 2025
08:15
Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
Published on: March 28, 2025
827
Deep Learning-based User Authentication with Surface EMG Images of Hand Gestures
Summary
This study introduces a novel user authentication method using surface electromyogram (sEMG) images of hand gestures and deep anomaly detection. The approach effectively distinguishes clients from imposters, demonstrating viability for secure system access.
Area of Science:
- Biometrics
- Machine Learning
- Signal Processing
Background:
- User authentication is critical for system security.
- Existing methods face challenges with sophisticated threats.
- Surface electromyogram (sEMG) signals offer unique biometric data.
Purpose of the Study:
- To propose a novel user authentication method using sEMG images and deep anomaly detection.
- To evaluate the effectiveness of different sEMG image generation techniques.
- To assess the performance in client vs. imposter classification.
Main Methods:
- Acquiring multi-channel sEMG signals during hand gestures.
- Converting sEMG signals into sEMG images.
- Employing a deep anomaly detection model for classification.
Main Results:
- The proposed method successfully classifies users as clients or imposters.
- Different sEMG image generation methods show varying performance.
- Experimental results validate the proposed authentication approach.
Conclusions:
- The sEMG image-based deep anomaly detection method is a viable biometric authentication technique.
- This approach offers a promising direction for enhanced system security.
- Further research can optimize sEMG image generation for improved accuracy.
