A Signature Verification System Based on Spiking Neural Networks
Alimohammad Beigi, M. S. Khandel, Nasser Mozayani · 2004
Signature verification is one of the methods for authentication of users in different ap- plications. In dynamic (online) signature verification, besides spatial features extracted from the shape of signature, temporal information plays also a major role. Aritficial neural networks are among promising and powerful methods towards processing spatio-temporal information. In this paper, we present an algorithm for dynamic signature verification using spiking neural network based on time factors . MLP and LVQ netwoks are applyed In this research. Our model can be considerd as one of the simplest verification sys- tems in terms of computational costs and implementation overhead. Key Word: dynamic signature verification, temporal features , Spiking Neural networks.