Study on Hand Gesture Interface Using Multi-Channel EMG

This project explores the use of multi-channel Electromyography (EMG) signals from the MYO Thalmic bracelet for hand gesture classification. The dataset consists of time-series data from eight EMG channels, each capturing muscle activity during various hand gestures. The goal is to classify gestures such as hand at rest, clenched fist, wrist flexion, wrist extension, radial deviations, ulnar deviations, and extended palm. Machine learning algorithms are implemented to process the EMG signals and accurately predict the corresponding gesture. The frontend of the application is designed using HTML, CSS, and JavaScript, while the backend leverages Python with Flask for data processing and handling user interactions. The system provides a user-friendly interface for gesture classification, making it accessible for various applications such as assistive technologies and human-computer interaction. The project evaluates model performance based on accuracy and computational efficiency, aiming to provide a reliable tool for real-time gesture detection.

Keywords: Multi-channel EMG, Hand Gesture Classification, MYO Thalmic Bracelet, Machine Learning, Flask, Frontend Development, Python, Gesture Recognition, EMG Signal Processing, Real-time Prediction.