Study On Information Fusion In The MMW/IR Guidance System

Target recognition, space and time alignment of multi-sensor information fusion system is becoming a hot research subject. It has great significance in theory and practice to studying method and theory about target recognition and sensor registration of multisensor information fusion system for improving integrate performance. The work of this dissertation is focused on research of target recognition ,time and space alignment based on the information fusion in the MMW/IR dual control and guide system.Firstly, the background,significance and merit of information fusion are introduced.and gives an excellent overview of existing theory and actuality, summarizes the character and configuration of the MMW/IR dual control and guide system.Secondly, invariant feature extraction of infrared image and recognition method of feature fusion are studied,then Dempster-Shafer's Evidential Reasoning and fuzzy theory are used at the report level. The results of the experiment show that feature fusion method is more stable, reliable and can efficiently improve the accuracy and the ability of fault tolerance in the system of MMW/IR.Finally,the space and time alignment of MMW/IR dual control and guide system is studied,and gives an algorithm model ,then uses the algorithm to achieve the aim of target fusion and target track. The results show that the algorithm has a good performance and an important significance in practice.