AI-Driven Image and Video Analytics for Earth Observation Intelligence

This project cooperation aims to leverage the latest advances in artificial intelligence (AI) and machine learning (ML) to process Earth Observation (EO) data. The explosion of EO data from satellites, drones, and aerial systems presents a unique opportunity to extract actionable information in near real-time, but it also poses significant computational and algorithmic challenges.

The Network Control Laboratory contributes deep expertise in intelligent systems, data fusion, and scalable machine learning pipelines developed across two recent ESA-funded projects. The group focuses on designing models that combine convolutional neural networks (CNNs), attention mechanisms, temporal analysis, and unsupervised learning to extract features and detect patterns from both static and dynamic EO data, with the aim of providing timing support to public authorities to analyze critical data related to wildfires and landslides.

Targeted applications span environmental disaster monitoring, precision agriculture, and crisis management, particularly in regions that are hard to access or under-monitored.