面向目标识别的水下多传感器信息融合技术比较研究趑

对于水下作战,目标识别十分困难,因此必须走多传感器融合、多技术融合的道路,应分析研究不同的方法优缺点,取长补短,综合处理。本文比较了不确定性推理技术中主观Bayes方法与证据理论的特点,分别给出了基于2种技术的目标融合识别思想。提出了基于Bayes统计理论的身份识别和基于D-S理论融合身份识别的框架。研究了基于模糊神经网络的多传感器信息融合技术,提出了模糊神经网络信息融合紧密结合与松散结合的2种处理框架,并结合应用讨论了其特点。

Research on arts comparing of underwater multi-sensors information fusion of target recognition oriented

It is very difficult to recognizing target underwater,when underwater warfare carrying into execution. So,It is better to keep paths of multi sensors information fusion and multi arts fusion.It should compare with differient arts,and take advantages of various techniques.Then process them synthetically. Among uncertainty reason methods,the Subjective Bayes Method and the Proof Theory Method are compared with in this paper. Respectively,target fusion thoughts are given according to these two kinds of methods.Schemes of information fusion recognization based on statistical theory of Bayes and D-S theory are performed. Furthermore,multi-sensors information fusion based on fuzzy neural network is studied.Two schemes of processing networks,which are compact and incompact are given.Its characteristics are discussed combining with applications.