Development of a Space Objects Detection System based on Deep Learning

To ensure the security of the satellite infrastructure and reduce the risk of financial losses for both public and private entities, it has become crucial to detect and track space objects. Indeed, in recent years, the accumulation of objects in Earth's orbit, both natural and man-made, has raised growing concerns about the security and sustainability of space operations. This thesis is based on the development of a machine-learning-based detection system, developed to identify space objects within images, where the main objective is the identification of high-risk orbital regions, where the probability of impact with debris is particularly significant.