Embedded Online Fish Detection and Tracking System via YOLOv3 and Parallel Correlation Filter

Nowadays, ocean observatory networks, which gather and provide multidisciplinary, long-term, 3D continuous marine observations at multiple temporal spatial scales, play a more and more important role in ocean investigations. In this paper, we first perform image enhancement to produce depth information and benefit many vision algorithms and advanced image editing. We try to develop a novel underwater fish detection and tracking strategies combining you only look once(YOLO) latest detection algorithm YOLOv3 algorithm and parallel Correlation Filter. We demonstrated on the NVIDIA Jetson TX2 for online fish detection and tracking, enabling a fast system and rapid experimentation. It has been shown in the experiments that the developed scheme of this paper achieves consistent performance improvements on online fish detection and tracking for ocean observatory network.