AVAO Enabled Deep Learning Based Person Authentication Using Fingerprint
Rasika Deshmukh, Pravin Laxmikant Yannawar · Advances in intelligent systems research/Advances in Intelligent Systems Research · 2023
Person authentication based on biometrics has been a major aspect accountable for providing security to cyberspace.The traditional biometric-based systems are based on the usage of single modality, which are potentially devoid of the capability to provide high security.A Deep Maxout Network (DMN) is utilized for performing person authentication on the basis of fingerprint.A novel optimization algorithm, named African vultures-Aquila Optimization (AVAO) algorithm is devised for updating the weights of the DMN.The strategies of the African Vulture Optimization Algorithm (AVOA) are modified according to the expanded exploration capability of the Aquila Optimizer (AO) to develop the proposed AVAO algorithm.The introduced optimization enabled deep learning based person authentication system achieved an accuracy of 0.927, sensitivity of 0.938 and specificity of 0.930,thereby showing superior performance.