EMG Biometric Systems Based on Different Wrist-Hand Movements

Electromyogram (EMG) acquisition and analysis is growing in importance with human attempts to interact control equipment such as robots, prostheses or virtual environments. In some cases, only approved users should be permitted these capabilities. For applications, securing EMG-based a major open question - the best of authors’ knowledge, no prior art exists which can identify individuals from wide range wrist-hand gestures EMG readings within wearable device. This paper addresses this problem. Techniques are presented allow used biometric, allowing verify themselves. An EMG-sensing armband attached lower forearm anonymously authenticate member an group, themselves uniquely. development extensive biometric system, three datasets similar sensing different sessions were exploited. verification, accuracy up 93% achieved, 92% achieved for identification. The system also shown operate real-time on ARM Cortex A-53 embedded processor suitable housing device, incurring latencies 1.06 ms 1.61 verification identification respectively. These metrics comfortably sufficient use real-time, battery-powered authentication devices.