Automatic toll collection system using RFID with vehicle classification using convolutional neural network

Abstract

The need for efficient and secure toll collection systems has prompted the development of advanced technologies that streamline toll collection and enhance traffic management. This paper presents an automatic toll collection system that integrates Radio Frequency Identification (RFID) technology with vehicle classification using convolutional neural network algorithms. The proposed system aims to improve the accuracy and efficiency of toll collection processes while reducing illegal use of Fast-tags (RFID) on unauthorized vehicles. The RFID-based component of the system facilitates contactless payment by detecting vehicles equipped with RFID tags as they approach the toll booth. The system automatically processes the payment, enabling swift passage for vehicles and minimizing delays. To enhance security and accuracy, the system incorporates a vehicle classification module based on Single Shot Detector (SSD) and You Only Look Once (YOLO) models. Cameras capture images of approaching vehicles, which are then processed by CNN algorithms trained to classify vehicles based on features such as type, make, model and Size. This classification enables the system to apply appropriate toll rates according to vehicle category and ensure compliance with toll regulations. The integration of RFID and deep learning technologies provides a robust approach to managing toll collection, minimizing fraud or evasion, and ensuring a seamless experience for drivers. The proposed system also offers valuable data insights for traffic analysis and management, contributing to smarter transportation infrastructure like toll fee SMS services and automatic toll gate opening and closing system. The results demonstrate that the system significantly improves the efficiency and accuracy of toll collection while providing a reliable and secure method for vehicle classification. The proposed system holds potential for widespread adoption, aligning with the growing demand for intelligent transportation solutions.