Abstract :

Traditional Password-based Authentication Systems Exhibit Critical Vulnerabilities Including Susceptibility To Phishing, Bruteforce Attacks, And Credential Theft. This Paper Presents A Multimodal Biometric Secure File Storage System That Integrates Face, Fingerprint, And Dual-iris Recognition For Quadruple-modal Authentication. Leveraging MTCNN And FaceNet For Face Processing, MobileNetV2 Transfer Learning For Fingerprint Feature Extraction, And A Custom Convolutional Neural Network For Iris Recognition, The System Requires Simultaneous Verification Of All Four Biometric Samples Before Granting Document Access. Implemented As A Python Flask Web Application With SQLite Persistence, The System Achieves 625ms Average Authentication Latency With GPU Acceleration, 97.6% Test Pass Rate Across 42 Test Cases, And Maintains Complete User Isolation For Stored Documents. The Integrated Analytics Dashboard Provides Real-time Performance Metrics, Confidence Scores, And Comprehensive Audit Logging. Experimental Evaluation Confirms That The AND-fusion Decision Strategy— Requiring All Modalities To Match—provides Exponentially Stronger Security Than Single-modality Or Optional Multimodal Approaches, Establishing A Practical Framework For Biometricsecured Digital Document Repositories. Keywords—multimodal Biometrics, Face Recognition, Fingerprint Recognition, Iris Recognition, Secure File Storage, Deep Learning, FaceNet, MobileNetV2, Authentication

Published:

02-4-2026

Issue:

Vol. 26 No. 4 (2026)

Page Nos:

168-173

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite