Satellite Detection

Satellite detection research focuses on automatically identifying satellites and other space objects in imagery, primarily to enhance space situational awareness and monitor maritime activities. Current methods heavily utilize convolutional neural networks (CNNs), often coupled with algorithms like optical flow or linear quadratic estimators, for object detection and tracking in both resolved and unresolved imagery. These advancements enable near real-time detection and autonomous tracking capabilities, improving the efficiency and accuracy of space surveillance and contributing to applications such as monitoring illicit ship-to-ship transfers. The development of robust and efficient algorithms is crucial for managing the increasing number of objects in orbit and ensuring maritime security.