SIGNATURE VERIFICATION USING MULTIPLE NEURAL CLASSIFIERS

This paper is concerned with signature verification. Three different t ypes of global features have been used for the classification of signa tures. Feed-forward neural net based classifiers have been used. The f eatures used for the classification are projection moments and upper a nd lower envelope based characteristics. Output of the three classifie rs is combined using a connectionist scheme. Combination of these feat ure based classifiers for signature verification is the unique feature of this work. Experimental results show that combination of the class ifiers increases reliability of the recognition results. Copyright (C) 1996 Pattern Recognition Society.