От орбита до брега: инструменти с изкуствен интелект за откриване на нефтени разливи с помощта на платформи с отворени данни
From Orbit to Coastline: AI-Powered Tools for Oil Spill Detection Using Open-Source Platforms

Summary/Abstract: Rapid and accurate detection of marine oil spills is a critical global environmental challenge. Traditional monitoring methods, relying on ship patrols and aerial surveillance, are limited by cost, scope, and time. A paradigm shift is underway, fueled by the convergence of three transformative technologies: open-source satellite data, unmanned aerial vehicles (UAVs), and artificial intelligence (AI). This synthesis examines the current ecosystem of AI-powered tools that use freely available data from orbital platforms and drone-based imagery to automate and improve oil spill monitoring. The standard AI workflow is analyzed, from data preprocessing and classical machine learning to advanced deep learning segmentation models such as U-Net. Satellite systems provide all-weather surveillance, and unmanned aerial vehicles (UAVs) provide unparalleled high-resolution inspection and thickness estimation. Despite progress, key challenges remain, including distinguishing “lookalike” phenomena and the need for explainable AI in operational contexts. The future trajectory points to integrated, multi-platform monitoring systems where AI not only detects but also predicts spill dynamics, building a powerful open-access toolkit for marine ecosystem conservation.

• Details
• Contents