Statistical texture image classification using two-dimensionalnonminimum-phase Fourier series based model

摘要

Using the 2-D extension of Chi's real 1-D parametricnonminimum-phase Fourier series based model (FSBM) for or as anapproximation to any arbitrary nonminimum-phase linear time-invariant(LTI) systems, we propose a system identification algorithm for 2-Dnonminimum-phase linear shift-invariant (LSI) systems supported by somesimulation results. The estimated 2-D FSBM parameters and second- andhigher-order statistics obtained using the proposed algorithm constituteeffective features for texture image classification supported by someexperimental results