AI and satellites detect illegal landfills with 90% accuracy.

Artificial intelligence applied to satellite imagery analysis allows us to detect illegal landfills and waste disposal sites with over 90% accuracy. This is demonstrated by the testing of the innovative technology developed by the Polytechnic University of Milan as part of the international Perivallon research project. Already operational in several European countries, it accelerates environmental investigations, improving land protection and the fight against eco-mafia.

The project, funded by the Horizon Europe program, involves 24 partners from 12 European and non-EU countries, including the Safe Foundation, Arpa Lombardia, and the Carabinieri Ecological Unit. Thanks to collaboration with environmental agencies and law enforcement agencies, this technology is already being used for large-scale land monitoring.

At the heart of the innovation is the use of neural networks and advanced algorithms to analyze satellite imagery and detect hidden hazardous waste deposits in urban, industrial, and agricultural areas. One of the project's key tools is the public AerialWaste dataset, the first of its kind in the world, already downloaded over 6,500 times by international researchers. The trial conducted with Arpa Lombardia in over one hundred municipalities has demonstrated that the use of artificial intelligence dramatically reduces investigation times, allowing for quicker intervention and preventing irreversible damage to ecosystems and public health.

"Thanks to artificial intelligence and satellite imagery analysis, we can now identify signs of environmental crime on a large scale, quickly and accurately," explain Piero Fraternali and Giacomo Boracchi of the Department of Electronics, Information, and Bioengineering at the Polytechnic University of Milan. "These tools introduce a paradigm shift, enabling systematic land surveys to support law enforcement and environmental agencies."