D-S证据理论在多传感器目标识别中的改进  

Improvement of D-S Theory Evidence in Multi-sensor Target Identification System

作　　者：董彦佼[1] 韩元杰[1] 刘洁莉[1] 

机构地区：[1]桂林电子科技大学,广西桂林541004

出　　处：《弹箭与制导学报》2009年第4期220-222,共3页Journal of Projectiles,Rockets,Missiles and Guidance

基　　金：广西自然科学基金(0542052)资助

摘　　要：针对证据理论在处理冲突证据时可能得到与直觉相悖的结论,文中首先分析了基本D-S证据理论组合准则缺陷的产生原因,提出一种新的基于权值的证据调整方法,给予证据分配不同的权值,重新分配概率分配函数。算例分析结果表明,改进后的方法可以提高识别结果的可靠性,在一定程度上改善了目标识别系统的性能。It is invalid when the evidence theory is used to deal with conflicting probabilities. In this paper, it first analyzes how basic D-S theory of evidence can result in limitations. Then a new weighted D-S combination rule is proposed to solve the conflicting management in the target identification system. Different weight was allocated ervry evidence, and reallocate the BPA. The result of example shows that the new method can improve the reliability of identification result. and improves the performance of target identification system to a certain extent.

关 键 词：信息融合 证据理论 目标识别 

分 类 号：TP212.9[自动化与计算机技术—检测技术与自动化装置]