DESIGN AND IMPLEMENTATION OF AUTOMATED BARRIER GATE SYSTEM WITH LICENSE PLATE RECOGNITION

This research involved the design and implementation of an Automated Barrier Gate System with integrated License Plate Recognition (LPR) to enhance vehicular access control at Nile University of Nigeria. A Raspberry Pi 5 served as the central controller, interfacing with a Camera, infrared sensors, and a DC motor-driven barrier. YOLOv8 was used for license plate detection, achieving 96.7% confidence, while Paddle-OCR delivered character-level accuracy of 93.2% and full-plate match accuracy of 87.5% on Nigerian plates. A custom dual-stage power supply unit using buck and boost converters generated regulated 5V and 30V outputs for digital and mechanical subsystems, respectively. These converters were simulated in Proteus and validated through physical testing, showing output deviations within ±5%. Python-based software was developed using OpenCV, PyTorch, and SQLite for real-time image processing, database matching, and barrier control. Lightweight aluminum was selected for the barrier arm to minimize mechanical load and wear. The complete system achieved an average response time of 14.6 seconds from vehicle detection to gate actuation, well within the 20-second performance target. Testing confirmed system reliability under controlled conditions. However, limitations included restricted plate format support, sequential gate operation, and scalability constraints. Despite these, the system proved to be a cost-effective and efficient solution for campus-level access control with potential for broader deployment in other secured environments.