Author Login Editor-in-Chief Peer Review Editor Work Office Work

Computer Engineering ›› 2010, Vol. 36 ›› Issue (8): 191-193. doi: 10.3969/j.issn.1000-3428.2010.08.067

• Artificial Intelligence and Recognition Technology • Previous Articles     Next Articles

Evidence Combination Rule Based on Thought of Nearest Neighbors

WANG Le, WU Gen-xiu, JI Jun, HU Zhen   

  1. (School of Mathematics and Information Science, Jiangxi Normal University, Nanchang 330022)
  • Received:1900-01-01 Revised:1900-01-01 Online:2010-04-20 Published:2010-04-20

一种基于近邻思想的证据合成规则

王 玏,吴根秀,纪 军,胡 真   

  1. (江西师范大学数学与信息科学学院,南昌 330022)

Abstract: According to the insufficiencies existing in D-S evidence theory when they deal with the conflicting evidences, many scholars present modified rules based on reallocation of the basic probability assigned to conflict. On the base of analyzing and comparing a few common combination rules, a new combination rule is presented to mend the disadvantages of the rules. This new rule realizes information fusion rationally with the use of effectiveness of evidential sources and the distance between them. Experimental results show that the new combination rule overcomes the insufficiencies. The outputs are constringent on the aspect of rationality, and the rule can be applied to combine the total conflict function.

Key words: evidence theory, information fusion, combination rule, conflict

摘要: D-S证据合成规则在遇到冲突证据时常有悖常理。针对现有规则的不足,通过理论分析和对几种常用组合规则的比较,提出一种基于近邻思想的冲突证据修正规则,综合考虑证据源的有效性和证据源间的距离,进行信息融合。实验结果表明,该规则对合理的合成方向具有较快的收敛速度,在证据源完全冲突的情况下也能适用。

关键词: 证据理论, 信息融合, 合成规则, 冲突

CLC Number: