Enhanced security in ATM By Face and Iris Recognition Using Improved-KF-RBF Model

Abstract

A biometric system utilizing face recognition could enhance the security of the Automatic Teller Machine, one of the oldest and most dependable types of technology that is still in use today. A biometric system's primary use case is input authentication through database validation and identification. There are essentially five steps to the process: preprocessing, image segmentation, iris normalization, feature extraction, and training (the model). For the sake of adding complexity during preprocessing, the proposed approach used brown iris photos. Then use the ocular image to extract picture segmentation. A result of iris normalization is an iris region with consistent dimensions across all environmental and geographical contexts. To extract features, multiscale morphology is employed. Discover how to use the Improved KF-RBF model for face and iris detection in this paper's ATM section.  Contrasted with RBF and KF, this method looks antiquated. Incredibly, the data reveals an improvement with a 97.23% accuracy rate.