SIGNATURE VERIFICATION USING NEURAL NETWORKS

Signatures are predominantly used to verify an individual’s authenticity. In this project we present new improved off-line signature verification system using global and texture features of signatures.The proposed technique in this research will assist individuals in characterizing signatures in order to check whether they are genuine or forged. In this proposed project an algorithm to validate signatures using CNN (Convolution neural network), Siamese neural network and VGG16 (Visual geometry group of transfer learning) models in our system. Authentic and competent forging signatures are used to test the approach. Results obtained by our proposed algorithm are more efficient than most of the existing techniques.