决策层特征融合decisionlevelidentityfusion.pptxVIP

Decision-Level Identity FusionTan XinLab 5, System Engineering Dept. Contents1. Introduction2. Classical inference3. Bayesian inference4. Dempster-Shafer’s method*5. Generalized Evidence Processing (GEP) Theory6. Heuristic methods for identity fusion7. Implementation and trade-offsDecision-level fusion(joint identity declaration)Feature-level fusion(Feature extraction, identity declaration)Data-level fusion(Data fused)IntroductionDecision-level fusionSeeks to process identity declarations from multiple sensors to achieve a joint declaration of identity.IntroductionFeature ExtractionSensorAAssociationIdentityDeclarationDecisionLevelFusion –IdentityFusionSensorBIdentityDeclarationSensorNIdentityDeclarationIntroductionDecision-Level Fusion TechniquesClassical inferenceBayesian inferenceDempster-Shafer’s methodGeneralized evidence processing theoryHeuristic methodsClassical inferenceTheoretical baseStatistical inference techniques seek to draw conclusions about an underlying mechanism or distribution, based on an observed sample of data.Classical inference typically assumes an empirical probability model.Empirical probability assumes that the observed frequency distribution will approximate the probability as the number of trials.heren trials, occurrence of k timesClassical inferenceOne disadvantageStrictly speaking, empirical probabilities are only defined for repeatable events.Classical inference methods utilize empirical probability and hence are not strictly applicable to nonrepeatable events, unless some model can be developed to compute the requisite probabilities.Classical inferenceMain technique – hypothesis testingDefine two hypothesis1. A null hypothesis, H0 (原假设)2. An alternative hypothesis,H1 (备择假设)Test logic1. Assume that the null hypothesis (H0) is true;2. Examine the consequences of H0 being true in the sampling distribution for statistic;3. Perform a hypothesis test, if the observation have a high probability of being observed if H0 is true, the declare