SCCM: A Novel Hybrid Framework for Dual Biometric Authentication for Identity Enhancement

Abstract

The increasing need for secure identification has led to advancements in biometric authentication. Traditional single-factor methods, such as passwords or tokens, are vulnerable to cyber threats, necessitating more robust alternatives. This study proposes a Dual Biometric Authentication System (DBAS) based on SCCM (SIFT-CNN-MLP), integrating fingerprint and iris recognition to enhance security, precision, and reliability. The system leverages state-of-the-art machine learning and computer vision techniques, including Multilayer Perceptron (MLP), Convolutional Neural Networks (CNN), and Scale-Invariant Feature Transform (SIFT). MLP and CNN models analyze intricate iris patterns, while SIFT and CNN extract distinguishing features from fingerprint ridge-valley structures. By intelligently combining authentication results from both biometrics, DBAS ensures seamless access for authorized users while effectively blocking imposters. The proposed SCCM framework demonstrates high performance in terms of accuracy. For iris authentication, MLP achieved 97.33% accuracy, while CNN outperformed with 97.92% accuracy. Similarly, for fingerprint authentication, SIFT yielded 91.30% accuracy, whereas CNN excelled with 99.23% accuracy. The SCCM-based DBAS significantly enhances authentication accuracy and robustness compared to traditional methods. It is highly effective for computer logins, mobile security, and critical infrastructure protection, making it a novel, future-proof solution for secure authentication.