09 January 2024 14h00-14h30
Mattia Stasolla Dr. ir.
Geospatial intelligence has become an important component of modern warfare. Thanks to satellite images, it is nowadays possible to monitor large areas of interest with a short revisit time, collecting valuable information that would be otherwise difficult to gather. However, the majority of the images that are currently in use come from electro-optical sensors. If on the one hand they are more easily interpretable, on the other hand they can be severely affected by atmospheric conditions and are therefore limited in terms of usable data they can provide. Moreover, the data processing is still largely carried out by visual analysis, although machine learning-based systems are more and more used to reduce the workload. Owing to its all-weather and day-and-night acquisition capabilities, in recent years synthetic aperture radar (SAR) technology has drawn the attention of Defense agencies, which are now contracting commercial providers to integrate their products into the existing government systems. Due to the fundamentally different nature of the SAR acquisition process, the interpretability of such data poses some challenges. The implementation of automated analysis routines becomes therefore crucial. In this talk we will show how time series of SAR images could be exploited for intelligence operations such as monitoring wide geographical areas to identify military activity in an automated way.