SIGNATURE VERIFICATION USING SIAMESE NEURAL NETWORK ONE-SHOT LEARNING

With the acceleration of digitalization in all areas our lives, need for biometric verification methods is increasing. The fact that data unique and stronger against phishing attacks compared to password-based authentication methods, has increased its preference rate. Signature verification, which one types, plays an important role many such as banking systems, administrative judicial applications. There are 2 types signature online offline, identifying identity person detecting forgery. Online carried out during signing temporal dynamic available regarding person's signature. Offline applied by scanning image after signing, this limited spatial data. Therefore, offline process considered a more challenging task. In study, independent writer, based on One-Shot Learning, was performed using Siamese Neural Network. Due Deep Convolution Network requires large amount labeled classification, real fake distinction been achieved Learning method, can perform successful classification less numbers images. As result experiments conducted datasets, architecture, proposed approach percentage accuracy 93.23, 92.11, 89.78, 91.35 4NSigComp2012, SigComp2011, 4NSigComp2010 BHsig260 respectively.