新的基于加权均值模型的证据融合方法  

New evidence fusion method based on weighted mean value model

作　　者：路申易 税爱社[1] 

机构地区：[1]后勤工程学院,重庆401331

出　　处：《传感器与微系统》2016年第7期55-57,61,共4页Transducer and Microsystem Technologies

基　　金：总后勤部军需物资油料部项目(油20130208)

摘　　要：针对D-S证据理论难以处理证据冲突的问题,提出了一种将Murphy平均融合方法和证据权方法相结合的证据融合方法。该方法将显著偏差证据的判别引入融合流程,实现对证据权重的区分量化,建立了加权的基本概率分配均值模型。仿真结果表明:该方法能有效区分证据的重要程度,提高了证据融合的准确性与收敛速度,较好地解决了冲突证据融合的问题。Aiming at problem that evidence conflicting can＇t be well solved by D-S evidence theory,a new evidence fusion method is proposed based on combination of Murphy average fusion method and weights of evidence method. This method introduces judgment of obvious deviation evidence into fusion process,weights of evidence are distinguish quantization,and mean value model for basic probability assignments of weighting is established. Simulation results show that the method can distinguish the importance of the evidence efficiently,accuracy of evidence fusion and convergence speed are also improved,solve problem of conflicting evidence very well.

关 键 词：D-S证据理论 证据融合 基本概率分配 均值模型 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]