A jellyfish distribution management system using an unmanned aerial vehicle and unmanned surface vehicles

In this paper, we propose a jellyfish distribution management system using an UAV (unmanned aerial vehicle) and USVs (unmanned surface vehicle). The UAV was designed to satisfy the requirements for flight in ocean environment. The target jellyfish, aurelia aurita, is recognized through a convolution neural network and its distribution is calculated. A high-speed deep neural network architecture has been developed to have reliable recognition accuracy and fast operation speed. We also propose a method for selecting candidates that are inputs to the proposed network. The recognition accuracy of the jellyfish is increased by removing the probability value of the meaningless class among the probability vectors of the evaluated input image and re-evaluating it by normalization. The jellyfish distribution is calculated based on the unit jellyfish image thus recognized. The distribution level is defined by using the novelty concept of the distribution map buffer. The proposed algorithm has a frame rate of 8 Hz or higher without any GPU and a recognition rate of over 90%.