摘要
为了有效地融合高度冲突的证据,该文首先分析了经典证据理论中冲突系数的不足,引入三角模算子用于表示证据之间的相似性,由此得到证据距离并定义证据间的马氏距离,二者结合经典冲突系数提出了新的冲突表示方法,能够更有效地度量证据的冲突程度。在此基础上提出了证据的二次修正组合方法。该方法保持D-S组合规则不变,采用序贯式实现证据组合,利用"折扣因子"及信任系数分别对组合证据体及各步融合结果进行了修正,最大程度地减小了冲突证据对融合结果的影响。试验结果表明,该方法与传统的证据理论融合方法相比较,具有更好的准确性和稳定性,能够有效地解决证据高度冲突情况下的信息融合问题。
In order to effectively combine highly conflict evidence,it first analyzes the shortcomings of conflict coefficient in the classical evidence theory,introduces triangular modular operator to represent similarity between evidences.The resulting distance is derived and the Mahalanobis distance between the evidences is defined.The two elements combine classical conflict coefficient provided a new conflict representation method which can more effectively measure the degree of conflict of evidence.The secondary correction combination method of evidence is put forward on the basis.The method of keeping the D-S combination rule unchanged,sequential implementation of evidence combination,using the ＂discount factor＂ and trust coefficient treat evidence combination body and each step of the fusion results of correction,minimize the impact of conflict evidence fusion results.The experimental results show that this method has better accuracy and stability than the traditional evidence theory fusion method,which can effectively solve the problem of information fusion in the case of high conflict of evidence.
作者
曹洁
冯锋
王进花
CAO Jie FENG Feng WANG Jin-hua(School of Computer and Communication,Lanzhou University of Technology,Lanzhou 730050,Chin)
出处
《自动化与仪表》
2017年第9期5-9,19,共6页
Automation & Instrumentation
基金
国家自然科学基金资助项目(61263031)
甘肃省自然科学基金资助项目(1506RJZA10)
关键词
证据理论
冲突系数
D-S组合规则
三角模算子
二次修正
evidence theory
conflict coefficient
D-S combination rule
triangular modular operator
secondary correction
