Deep learning for marine recreational fisheries spatial-temporal planning

Abstract

[eng] Successful marine spatial planning relies on understanding patterns of human use, with accurate, detailed, and up-to-date information about the spatial distribution of fishing effort. In commercial vessels, tracking systems like Vessel Monitoring System (VMS) or Automatic Identification System (AIS) have helped to maintain and enhance biodiversity of areas by generating large sources spatial positional data that served for monitoring commercial marine spatial planning. Unfortunately, there is no regulation regarding location systems such as VMS or AIS for marine recreational fishing boats. Obtaining spatial data of marine recreational fishing can be difficult and time-expensive given the widespread and variable nature of the fleet. Remote cameras, computer visions systems and artificial intelligence are increasingly used to overcome cost limitations of these conventional methods. Here we show a low-cost and real-time tracking system based on photo time-lapses and deep learning algorithms (YOLOv5 and DeepSORT) to automatically detect, classify and track recreational fishing boats in coastal areas. This method allows determining the intensity and spatial-temporal distribution of recreational fishing effort, important to monitor marine protected areas and to define the sustainability of fishing activity in coastal areas.