ONLINE SIGNATURE VERIFICATION USING NORMALIZED DYNAMIC FEATURE WITH ARTIFICIAL NEURAL NETWORK CLASSIFICATION

Handwritten signature verification system is most widely used in any financial and other documentation activities for authorization of identity of any human activity, but still these types of verification system are manly based on manual verification, that is a person only by looking compare the given signature with the test signature, so a more robust system is required which can be based on some computer based classification, so in this paper an online signature verification is proposed which is based on normalized dynamic features of the signature using artificial neural network as classification method, the proposed system used digital tablet with digital pen for acquisition of the signature and extract three dynamic features of the signature i.e., x,y c oordinate of the signature along with the pressure at different points of the signature and from this dynamic features 11 different feature set is calculated and then using this feature set a neural network is trained for classification, the proposed sy stem provide FAR of about 5% and FRR of about