Signature Verification System using Siamese Neural Networks

This paper presents an efficient approach to signature verification and forgery detection using Siamese neural networks algorithm. This research involves training classifiers on a diverse dataset of signatures in different persons. Our method employs vectorization and Euclidean distance classifier for signature classification and detection tasks. Experimental results demonstrate the effectiveness of the classifiers with high accuracy and robustness against various signature styles and classifying the genuine and forgery signatures. The approach showcases notable efficiency in signature forgery detection, addressing the growing need for various fraud detecting solutions.