Convolutional Neural Network Enhanced Two-Factor Authentication for RFID/IOT-Based Attendance System
Year: 2025
Type: article
Abstract: This paper presents a deep learning-based approach for fingerprint biometrics and RFID authentication, designed for secure attendance systems. The proposed system employs a Tiny-YOLO CNN model to learn and recognize fingerprint patterns, achieving an average accuracy of 95%. Additionally, the system uses the same CNN model to authenticate RFID sequences, demonstrating an average accuracy of 90%. The results highlight the potential of using deep learning-based approaches for biometric authentication, particularly in resource-constrained devices such as the Raspberry Pi. The proposed system demonstrates promising performance and can be further optimized and improved for real-world deployment. This study contributes to the development of secure and efficient attendance systems, with potential applications in various fields, including education, healthcare, and finance.
Authors: Saeed Musa Yarima, Alhassan Haruna Umar, Amiru Ali, Abdullahi Mohammed Ibrahim
Language: English
Subfield: Signal Processing
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
Funder: Tertiary Education Trust Fund
OpenAlex ID: W4412724422