Data Fusion and Utilization of Intelligent Perception and Self-learning of Multivariate Information on Satellite

Technologies such as real-time processing of onboard data, intelligent perception, and self-learning information fusion have become the main trends in the development of future satellites. This paper studies the algorithms for intelligent perception and self-learning fusion utilization of multiple information on satellites. The basic tasks involved in the research are summarized and divided. A computational framework for intelligent perception and fusion utilization of multiple information is proposed, which includes three levels: “target level perception scene level perception evaluation and decision support”. Target detection and recognition based on single remote sensing data, potential target self-learning detection, and Feature fusion target recognition algorithm for multivariate remote sensing data. Compared with other algorithms, the experimental results show that the proposed algorithm has better performance.