计算机工程与应用 ›› 2015, Vol. 51 ›› Issue (24): 61-66.

• 理论研究、研发设计 • 上一篇    下一篇

一种基于Pignistic概率距离的合成公式

马丽丽,张  芬,陈金广   

  1. 西安工程大学 计算机科学学院,西安 710048
  • 出版日期:2015-12-15 发布日期:2015-12-30

Synthetic rule of evidence based on Pignistic probability distance

MA Lili, ZHANG Fen, CHEN Jinguang   

  1. School of Computer Science, Xi’an Polytechnic University, Xi’an 710048, China
  • Online:2015-12-15 Published:2015-12-30

摘要: 针对多证据源信息融合过程中证据源间存在的冲突问题,提出了一种基于Pignistic概率距离的合成公式。利用Pignistic概率距离构造证据可信度;再利用证据可信度修正证据体,以改进合成公式;利用改进的合成公式对证据源进行融合。算例结果表明,改进合成公式的融合结果合理有效,与其他方法相比有更好的适用性、可靠性和较快的运算速度。

关键词: 可信度, Pignistic概率距离, 合成公式, D-S证据理论, 信息融合

Abstract: Aiming at the conflict problem among evidences in the process of information fusion, one synthetic rule of evidence based on Pignistic probability distance is presented. Evidence reliability is constructed by using the Pignistic probability, and then the body of evidence is altered by using the evidence reliability to improve the synthetic rule. Evidences are fused by employing the improved synthetic rule. Results of numerical examples show that the fusion results obtained through the improved synthetic rule are reasonable and effective. Compared with some other methods, it has good characters in applicability, reliability, and faster processing speed.

Key words: reliability, Pignistic probability distance, synthetic rule, Dempster-Shafer theory, information fusion