基于改进D-S证据组合规则的目标识别算法  

Target-Recognition Algorithm Based on Improved D-S Evidence Combination Rule

作　　者：周莉[1] 唐文静[1] 郭伟震 ZHOU Li TANG Wen-jing GUO Wei-zhen(School of Information and Electrical Engineering, Ludong University, Shandong Yantai 264025, Chin)

机构地区：[1]鲁东大学信息与电气工程学院,山东烟台264025

出　　处：《北京邮电大学学报》2016年第5期47-50,共4页Journal of Beijing University of Posts and Telecommunications

基　　金：国家自然科学基金项目(61273152);国家自然科学基金青年科学基金项目(61304052)

摘　　要：为了更好地解决高冲突证据的融合问题,提出一种3条证据直接融合的改进D-S算法.该算法首先根据证据支持贴近度函数给出识别框架下各焦元支持度的计算方法;其次根据三维证据直接融合产生的冲突因子的性质及各焦元的支持度,提出一种基于D-S证据组合规则的冲突信息加权分配算法;最后以多传感器多目标识别系统为背景进行仿真实验.理论分析和仿真结果表明,基于三维证据直接融合的改进D-S算法具有较强的抗干扰性能,能有效融合各种冲突信息,提高目标识别概率.In order to better solve the fusion problem of high conflict evidence,an improved D-S algorithm fusing three pieces of evidence directly was proposed. Firstly,the calculation method of the support of each focal element under the identification framework is given according to the function of evidence supporting measurement of similarity. Secondly,a weighted assignment algorithm of conflicting information based on D-S evidence combination rule is put forward according to the properties of the conflicting factor generated in the direct fusion process of three-dimensional evidence and the support degree of each focal element. Finally,simulation is implemented under the background of multi-sensor multi-target recognition system. Analysis and simulation show that the improved D-S algorithm based on direct fusion of three-dimensional evidence has a strong anti-interference performance,and it can fuse various types of conflicting information effectively and improve the target recognition probability.

关 键 词：D-S证据组合规则 目标识别 高冲突信息 证据支持贴近度 

分 类 号：TN391[电子电信—物理电子学]