Ultrasonic hand gesture detection and tracking using CFAR and Kalman filter

This paper presents a new approach to hand gesture recognition using constant false alarm rate (CFAR) and Kalman filter based on a compact ultrasonic phased array. A 45mm-long linear ultrasonic phased array consists of a single transducer element as the transmitter and eight transducer elements as the receiver. Firstly, the depth and azimuth of the targets are determined by offline receiver beamforming. Secondly, the CFAR detection technique in radar system is developed to ensure reliable identification of hand gestures. In addition, the motion trajectories of the targets are tracked by using the Kalman filter. Experimental results show that the proposed approach can detect and track hand gestures effectively, which could be applicable to human computer interaction.