A 1-D, sequence decomposition based, autoregressive hidden Markov model for dynamic signature identification and verification

A new model for use in writer identification and verification is presented. The signature, represented by a one-dimensional (1-D) spatial stochastic sequence, is decomposed into pseudo-stationary segments; a characterization which allows descriptions of abrupt and gradual changes in the contours. An autoregressive hidden Markov model is employed to describe the evolution of such changes. An experimental study is presented which demonstrates the model's effectiveness.<
>