Real Time Detection Of Anomalous Satellite Behavior from Ground-Based TelescopeImages

摘要

Air Force analysts are faced with the task of monitoring satellites with groundbased telescopes. Images are collected and analyzed in a time consuming and subjective effort to detect any behavior that is anomalous. This research maximizes use of a priori information to create an automated, real time satellite behavior classification tool. Using modeling software and knowledge of a satellite's orbit, reference imagery is created for each measured image in a satellite pass. Features are extracted from the measured and reference image pairs that provide good overall gaussian classification accuracy (85%), reduce the dimensionality of the problem (from 32,768 down to 3), and are least dependent on data partitioning. The statistical image pair classifier is tested for robustness to atmospheric distortion, and training data requirements are explored. Satellite behavior is classified by counting the classification results for the image pairs in a satellite pass. A binomial analysis of the classification technique predicts virtually 100% classification accuracy of satellite behavior. This research demonstrates the validity of model based satellite behavior analysis.