Face, Iris, and Fingerprint based Robust Biometric Authentication System

When it comes to identifying people automatically, bio-metrics is the field of study that uses observable traits of the human body, both innate and acquired, to do so. “Bio-metrics is a rapidly developing breakthrough that is being used to increase security and verify sensitive systems. A multimodal bio-metric system combines data from several sources to solve these issues and outperform a uni-modal system in terms of recognition accuracy. Since spoofing can’t be performed on many bio-metric sources at once, multimodal systems have a lower failure to enroll rate. As a level-based method, multimodal bio-metrics combines information at several stages, including sensing, features, matches, and decisions. Because it includes more substantive information, fusion at the score level delivers superior recognition performance, and it is both possible and practicable. The purpose of this study is to evaluate the efficacy of combining fingerprint, face, and iris bio-metrics based on matching score. The fusion rule and classification are carried out using three different soft computing methodologies; the SVM classifier, the sparse SVM classifier, and the rough set fuzzy classifier”.