Micro Hand Gesture Recognition System Using Ultrasonic Active Sensing

In this paper, we propose a micro hand gesture recognition system using ultrasonic active sensing, which uses micro dynamic hand gestures within a time interval for classification and recognition to achieve Human-Computer Interaction (HCI). The implemented system called Hand-Ultrasonic-Gesture (HUG) consists of ultrasonic active sensing, pulsed radar signal processing, and time-sequence pattern recognition by machine learning. We adopted lower-frequency (less than 1MHz) ultrasonic active sensing to obtain range-Doppler image features, detecting micro fingers motion at a fine resolution of range and velocity. Making use of high resolution sequential range-Doppler features, we propose a state transition based Hidden Markov Model for classification in which high dimensional features are symbolized, achieving competitive accuracy of nearly 90% and significantly reducing the computation complexity and power consumption. Furthermore, we utilized the End-to-End neural network model for classification and reached the accuracy of 96.32%. Besides offline analysis, a real-time prototype was released to verify our methods potential of application in the real world.