Real-Time Hand Gesture Recognition Using Surface Electromyography and Machine Learning: A Systematic Literature...

Area of Science:

• Human-Computer Interaction (HCI)
• Biomedical Engineering
• Machine Learning

Background:

• Daily life increasingly involves computing systems, necessitating natural interaction methods.
• Human-Computer Interaction (HCI) aims to bridge communication gaps between humans and computers.
• Hand Gesture Recognition (HGR) is a key HCI modality, interpreting hand movements.

Purpose of the Study:

• To systematically review the state-of-the-art in real-time hand gesture recognition models utilizing surface electromyography (EMG) and machine learning.
• To analyze and standardize common components of machine learning-based HGR systems using EMG data.
• To identify current trends and research gaps in EMG-based HGR.

Main Methods:

• A systematic literature review was conducted, adhering to the Kitchenham methodology.
• 65 primary studies on real-time EMG-based HGR were selected and assessed.
• Analysis focused on model structure, data acquisition, preprocessing, feature extraction, classification, and evaluation metrics.

Main Results:

• Standardized concepts across various components of EMG-based HGR systems were established.
• Key aspects analyzed include data acquisition, signal processing, feature extraction, and classification algorithms.
• The review identified specific trends and gaps in the current research landscape.

Conclusions:

• The study provides a comprehensive overview of EMG-based real-time HGR, consolidating knowledge and methodologies.
• Identified gaps highlight opportunities for future research in improving accuracy, efficiency, and applicability of these systems.
• This review serves as a foundational resource for researchers and developers in the field of gesture recognition using EMG signals.