Revistes Catalanes amb Accés Obert (RACO)

A neuro-fuzzy system for isolated hand-written digit recognition

Miguel Pinzolas, José Javier Astrain Escola, Jesús Villadangos Alonso, José Ramón González de Mendivil

Resum


A neuro-fuzzy system for isolated hand-written digit recognition using a
similarity fuzzy measure is presented. The system is composed of two main
blocks: a first block that normalizes the input and compares it with a set
of fuzzy patterns, and a second block with a multilayer perceptron to
perform a neuronal classification. The comparison with the fuzzy patterns is
performed via a fuzzy similarity measure that uses the Yager parametric
t-norms and t-conorms. Along this work, several values of the parameters
have been studied, in order to obtain the best classification. The
simplicity of the method makes it extremely quick and provides a
recognition accuracy about 90% in classification of isolated digits, making
it an attractive method for practical applications.

Text complet: HTML