Abstract
Biometric systems input physical or personal human characteristics for identification, authentication, and security purposes. With the advancement in communication and intelligent security systems, biometrics are programmed to validate electronic signatures (E-signatures) for online and offline authentication. This article introduces a dynamic signature verification technique (DSVT) using mutual compliance (MC) between the security system and the biometric device. The security system is responsible for online and offline signature approval using personal inputs from humans. This personal verification is related to the stored online/offline signatures using certificates provided for authentication. The certificate-based authentication is valid within a session for online representation. Contrarily, this authentication is valid for persons under offline conditions. In this mode of segregation, application-level authentication verification is performed. A conventional tree classifier for dynamic signature verification is used for differentiating online and offline signatures. Moreover, the security metrics—such as signing bit, key, and size—are verified for both modes using classifier learning. For the segregated mode, the validation of the above is required to be unanimous to accelerate the dynamicity. The proposed technique’s performance is analyzed using the authentication success rate, verification failing ratio, verification time, and complexity.