A New Pattern Recognition Method for Parametric Modelling of Random Fields

摘要

To predict more reliably the real system’s behaviour by means of simulation,rnprobabilistic models can be created that account for aleatory uncertainties. Localrneffects of uncertainty can be modelled by random fields in the n-dimensional featurernspace. In this paper general constraints for the parameterisation of random fields arernidentified. A new multi-scale method for a pattern recognition based parameterisationrnof random fields is developed: The approximated differential scale-spacern(ADSS). The ADSS yields parameter sets that describe the relevant characteristicsrnof the random field and allows synthesising of simulation samples. The capabilitiesrnof the new method are demonstrated for one- and two-dimensional applicationrnexamples.