<![CDATA[Purpose – This study aims to design and develop an intelligent vehicle access system that enhances security through a two-factor authentication mechanism integrating face recognition and RFID-based electronic driver identification (E-SIM). Design/methods/approach – The research adopts a Research and Development (R&D) approach, including system design, implementation, and evaluation. The system is built on a Raspberry Pi 4 platform and integrates face recognition using the Histogram of Oriented Gradients (HOG) method with RFID UID verification. Additional features include GPS-based tracking and Telegram-based real-time notifications. Performance evaluation is conducted using confusion matrix metrics and experimental testing under varying environmental conditions. Findings – The proposed system achieves 95% accuracy, 95.92% precision, 94% recall, and an F1-score of 94.95%. The system demonstrates good performance in preventing unauthorized access, with only two false acceptance cases. Performance remains stable under moderate lighting and short distances but decreases under low illumination and longer distances. The GPS module provides reliable tracking with an average positioning error of approximately 5.06 meters. In terms of real-time performance, the system exhibits an average latency of approximately 6.84 seconds per authentication cycle, which remains acceptable for practical vehicle access applications. Research implications/limitations – The system demonstrates strong performance as a functional prototype; however, it remains vulnerable to face spoofing and RFID cloning due to the absence of liveness detection and encrypted communication. Environmental factors such as lighting and distance also affect recognition accuracy. Originality/value – This study contributes by integrating biometric and possession-based authentication within a standalone embedded system, enhanced with IoT features for real-time monitoring. Unlike prior single-factor approaches, the proposed system improves security robustness while maintaining practical usability.]]>
