EMG Based Human Machine Integration for IoT Based Instruments

Introducing the application of hand gestures in human-computer interaction, this study presents a novel approach that employs artificial neural networks (ANNs) to analyze muscle activity and distinguish hand movements from electromyograms (EMGs). The emphasis on creating a user-friendly and stress-free environment is reflected in the limitation of EMG detection to three specific muscle groups in the arm. The analysis of EMG signals yields parameters related to the temporal patterns of muscle contractions, each uniquely associated with a specific gesture. An experimental study was conducted to assess the system’s ability to accurately recognize six distinct arm movements. The results demonstrate that the algorithm achieves an impressive 98% accuracy rate in identifying all six motions. One notable advantage of this system is its ease of training and seamless real-time execution once the training process is completed.