摘要
Dempster-Shafer(D-S)证据理论在信息融合领域得到了广泛的应用。为抑制Dempster证据合成规则产生的有悖常理的证据合成结果,提出了一种基于局部冲突消除的证据合成法则。首先,通过一种基于Vague集和逐层递推算法的证据冲突衡量方法计算了证据之间的冲突程度;进而根据距离法思想对高冲突证据进行局部冲突"中和";最后,利用Dempster证据合成法则对已预处理的证据进行信息融合。算例结果验证了此种法则的有效性。
The Dempster-Shafer evidence theory is widely used in the field of information fusion.To suppress the illogical results when using Dempster＇s combination rule,a modified evidence combination rule based on local conflict elimination is proposed.First,based on Vague set and layer by layer recurrence algorithm,the conflicting degree between available bodies of the evidence is measured by a new representation method of evidential conflict.In each combination step,according to the conflicting degree,the elimination of the local conflict is judged.Then the distance method is used to modify the bodies of the evidence.Finally,the combination result can be obtained by the Dempster＇s combination rule.Some numerical examples show the efficiency of the proposed rule.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2016年第7期1594-1599,共6页
Systems Engineering and Electronics
基金
国家自然科学基金(51407128)资助课题
