Mallard Detection using Microphone Array and Delay-and-Sum Beamforming

This paper presents a method of estimating the direction and position of mallards from voice information using arrayed microphones combined with delay-and-sum beamforming. Rice farming using mallards saves labor because mallards work instead of farmers. Nevertheless, the number of mallards declines when they are preyed upon by natural enemies such as crows, kites, and weasels. We consider that efficient management can be achieved from locating and identifying the locations of mallards and natural enemies using voice information that can be widely sensed in a paddy field. For this study, we developed a prototype system with 32 microphones installed as an array installed on our originally designed and assembled mount. We obtained benchmark sound datasets in an outdoor environment that comprises 18 grid points of 40 m width and 30 m depth. For direction estimation, no detection angle error occurred at 9 points. Estimation errors were increased at sharp-angle positions from the microphone array. For position estimation, our method correctly estimated 8 points.