基于新的证据冲突表示的信息融合方法研究

DS证据理论为不确定信息的表达和合成提供了自然而强有力的方法,被广泛应用于信息融合的各个领域。但是在证据高度冲突情况下常常会导致错误的融合结果。文章首先分析经典证据理论中冲突系数的不足,引入证据距离函数表示证据之间的差异性,并提出了一种新的证据冲突表示方法,能更有效地度量证据的冲突程度。而后基于新的冲突表示方法,提出了一种新的冲突信息融合方法,该方法保持DS组合规则不变,对融合结果进行修正,能够有效解决证据高度冲突情况下的信息融合问题。最后通过算例验证了该方法的有效性。

A Novel Information Fusion Method Based on Our Evidence Conflict Representation

Aim.The Dempster-Shafer(D-S) evidence theory is widely used in many fields of information fusion.However,erroneous results may be obtained by the classical Dempster combination rule when collected evidences are highly in conflict with each other.So we propose a novel information fusion method.Section 1 of the full paper analyzes the D-S evidence theory.Section 2 explains our new evidence conflict representation method through illustrative examples 1,2 and 3,including Table 1 and Fig.1.Section 3 gives illust...