HSCM: A New Framework For Handwritten Signature Verification Using ConvMixer

As nations attempt to recover from the coronavirus disease (COVID-19) epidemic, digital solutions facilitate economic change and put economies on the path to green growth. Therefore, identifying an individual based on biometric characteristics has rapidly become an urgent and essential subject. Banks, intelligence agencies, and high-profile institutions are frequently use handwritten signature verification to authenticate the identification of an individual. This paper introduces an innovative model (HSCM) for the authentication of online handwritten signatures based on the ConvMixer Network. The main idea of the proposed model is to divide the signature into several patches known as strokes, which are then processed using ConvMixer to generate signature features. These features are processed using Feed-Forward network. Experimental results on the real-world dataset show that the proposed model achieved 11.6% equal error rate (EER) compared with other competitors. This vindicates the efficacy and reliability of the proposed model in signature verification).