REAL-TIME TRAFFIC SIGN RECOGNITION AND CLASSIFICATION SYSTEM USING OPENCV

Abstract

This paper presents a Real-Time Traffic Sign Recognition and Classification System using OpenCV and Raspberry Pi, designed to assist in vehicle automation and faster decision-making for drivers or autonomous systems. The system uses a Raspberry Pi 4 Model B processor and a Pi Camera for live video capture, with OpenCV for image processing. A Convolutional Neural Network (CNN) is trained to classify traffic signs into predefined categories. The system's performance is evaluated based on recognition accuracy, processing speed, and handling multiple traffic signs simultaneously. This real-time system has potential for integration into modern vehicle automation systems and as assistive technology for drivers, contributing to safer roads and more efficient transportation networks.

To enhance the accuracy of the system, a Convolutional Neural Network (CNN) is trained using a dataset of traffic sign images. This CNN model is embedded within the system to classify the detected signs into predefined categories such as stop, yield, speed limit, or no entry. Machine learning techniques are employed to ensure that the model can adapt to different lighting conditions, camera angles, and background noise, thus making the system robust and reliable for real-world usage.

This real-time system offers significant potential for integration into modern vehicle automation systems or as an assistive technology for drivers, contributing to safer roads and more efficient transportation networks. The project demonstrates that low-cost hardware like Raspberry Pi, when combined with powerful image processing libraries like OpenCV, can effectively perform complex tasks such as traffic sign recognition, making it a viable solution for smart city applications and advanced driver assistance systems (ADAS).