A Deep Learning-based ROV Surveillance System for Monitoring Illegal Fishing

Abstract

The issue of illegal fishing is a global concern that requires attention. This paper focuses on marine biodiversity and sustainable development and proposes a deep learning-based computer vision technology for monitoring illegal fishing using ROV surveillance systems. By combining deep learning-based technology with a remotely operated underwater vehicle (ROV), supervisors can remotely control the underwater robot to take images and send them to the AI server through a 4G/5G mobile network. The AI server quickly identifies any instances of illegal fishing by divers in underwater images. Additionally, supervisors can use the illegal fishing monitoring platform to monitor for any illicit activity fishing underwater.