一种基于度量层信息的基本信任分配构造方法

贾宇平; 杨威; 付耀文; 庄钊文

doi:10.3724/SP.J.1146.2008.01105

一种基于度量层信息的基本信任分配构造方法

doi: 10.3724/SP.J.1146.2008.01105

计量
- 文章访问数: 3476
- HTML全文浏览量: 87
- PDF下载量: 866
- 被引次数: 0
出版历程
- 收稿日期: 2008-09-05
- 修回日期: 2009-03-10
- 刊出日期: 2009-06-19

A Basic Belief Assignment Construction Method Based on the Information of Measurement Level

摘要

摘要: 在决策层融合目标识别中，Dempster组合规则是一种常用的融合算子，它的有效应用取决于相应基本信任分配的合理建立。该文针对决策层融合目标识别问题中子源传感器输出的度量层信息，分析了基本信任分配的构造准则，并提出了一种基于参考向量与度量层信息之间相似度的基本信任分配构造方法。在仿真数据与空中雷达目标实测数据上的实验结果表明，该方法构造的基本信任分配能够有效地通过Dempster组合规则进行融合。
- 目标识别;基本信任分配;Dempster组合规则;度量层信息
Abstract: Dempster rule of combination is a useful fusion operator for decision level recognition fusion. The effective application of this rule depends on the rational construction of corresponding basic belief assignment. Considering the measurement information from each sensor in decision level recognition fusion, this paper suggests a group of principles for constructing basic belief assignment, and then presents a strategy of basic belief assignment construction based on the similarity degree between referential vector and the measurement information. The experiments on artificial data and real data of radar aerial target demonstrate the basic belief assignments constructed by the presented method can be fused by Dempster rule of combination effectively.

HTML全文

参考文献(1)

Dempster A. Upper and lower probabilities induced by amultivalued mapping [J].Annals of Math. Statistics.1967, 38:325-339[2]Shafer G. A Mathematical Theory of Evidence [M]. Princeton:Princeton Univ. Press, 1976.[3]Delmotte F and Smets P. Target identification based on thetransferable belief model interpretation of dempster-shafermodel [J]. IEEE Trans. on SMC, Part A: Systems andHumans, 2004, 34(4): 457-471.[4]王壮. C4ISR 系统目标综合识别理论与技术研究[D]. [博士论文], 长沙: 国防科技大学, 2001, 第2 章.[5]Xu L, Krzyzak A, and Suen C. Methods of combiningmultiple classifiers and their applications to handwrittingrecognition [J]. IEEE Trans. on SMC, 1992, 22(3): 418-434.[6]Guo H, Shi W, and Deng Y. Evaluating sensor reliability inclassification problems based on evidence theory [J].IEEETrans. on System, Man and Cybernetics-PART B:Cybernetics.2006, 36(5):970-981[7]黎湘, 付耀文, 景小军, 庄钊文. 多模导引头融合检测研究[J].系统工程与电子技术. 2000, 22(10): 25-26.[8]韩崇昭, 朱洪艳, 段战胜. 多源信息融合[M]. 北京: 清华大学出版社, 2006, 第11 章: 463-471.[9]Denoex T. A k-nearest neighbor classification rule based onDempster-Shafer theory [J].IEEE Trans. on System, Man,and Cybernetics.1995, 25(5):804-813[10]Matsuyama T. Belief formation from observation and beliefintegration using virtual belief space in Dempster-Shaferprobability model [C]. Proceedings of the 1994 IEEEInterational Conference on Multisensor Fusion andIntegration for Intelligent Systems, Las Vegas, NV, Oct. 1994:379-386.[11]Boudraa A, et al.. Dempster-Shafers basic probabilityassignment based on fuzzy membership functions [J].Electronic Letters on Computer Vision and Image Analysis,2004, 4(1): 1-9.[12]Ahmed A and Deriche M. A new technique for combiningmultiple classifiers using the dempster-shafer theory ofevidence [J]. Journal of Artificial Intelligence Reasearch, 2002,17(11): 333-361.[13]Yaghlane A, Denoeux T, and Mellouli K. Elicitation of expertopinions for constructing belief functions [C]. Proceedings ofIPMU'2006, Paris, France, Jul. 2006, Vol I: 403-411.[14]Lefevre E, et al.. Knowledge modeling methods in theframework of evidence theory: An experimental comparisonfor melanoma detection [C]. IEEE InternationalConference on Systems, Man, and Cybernetics, Nashville,TN, USA, Oct. 2000,Vol. 4: 2806-2811.[15]Denoex T. Construction of predictive belief functions using afrequentist approach [C]. Proceedings of IPMU'2006, Paris,France, Jul. 2006, Vol II: 1412-1419.[16]Aregui A and Denoex T. Consonant belief function inducedby a confidence set of pignistic probabilities [C]. Proceedingsof ECSQARU2007, Hammamet, Tunisia, 2007: 344-355.[17]叶芗芸. 基于神经网络的图象识别方法及应用[D]. [博士论文],上海: 上海交通大学, 1997, 第5 章.[18]Liu W. Analyzing the degree of conflict among belieffunctions [J].Artificial Intelligence.2006, 170:909-924[19]郭华伟, 施文康, 邓勇, 陈智军. 证据冲突: 丢弃, 发现或化解[J]. 系统工程与电子技术, 2007, 29(6): 890-898.[20]Smets P. Analyzing the combination of conflicting belieffunctions [J].Information Fusion.2007, 8(4):387-412[21]Haenni R and Hartmann S. Shedding new light on Zadeh'scriticism of Dempster's rule of combination [C]. 7thInternational Conference on Information Fusion, Univ. ofBerne, Bern, Switzerland, Appl. -Math. 2005: 880-884.

施引文献

资源附件(0)

访问统计