Related Experiment Video
Updated: Jun 1, 2026
08:15
Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
Published on: March 28, 2025
[Research on finger key-press gesture recognition based on surface electromyographic signals]
Juan Cheng1, Xiang Chen, Zhiyuan Lu
1Department of Electronics Science & Technology, Univ. of Science & Technology of China, Hefei 230027, China.
Summary
Researchers achieved over 75% accuracy in recognizing 16 finger key-press gestures using surface electromyographic (SEMG) signals. This demonstrates the feasibility of SEMG for myoelectric control of virtual keyboards.
Area of Science:
- Biomedical Engineering
- Signal Processing
- Human-Computer Interaction
Context:
- Surface electromyographic (SEMG) signals offer a non-invasive method for detecting muscle activity.
- Pattern recognition of human gestures is crucial for advanced human-computer interfaces.
- Virtual keyboard interaction can be enhanced through novel control mechanisms.
Purpose:
- To investigate the feasibility and repeatability of recognizing finger key-press gestures using SEMG signals.
- To define and classify 16 distinct right-hand key-press gestures based on PC keyboard standards.
- To evaluate the performance of SEMG-based gesture recognition across different training periods.
Summary:
- This study explored pattern recognition of 16 right-hand finger key-press gestures using SEMG signals from the forearm.
- Two pattern recognition experiments were conducted with 6 subjects, utilizing 4 SEMG sensors.
- Average classification rates exceeded 75.8% using same-day templates, with 5-day training yielding comparable accuracies.
Impact:
- Confirms the feasibility and repeatability of SEMG-based key-press gesture classification.
- Highlights the potential for SEMG in developing myoelectric control systems for virtual keyboards.
- Provides a foundation for more intuitive and efficient human-computer interaction through gesture recognition.
