In the modern world, correctly recognizing and categorizing traffic signs is essential for improving road safety and enabling autonomous driving technologies. This study uses the Raspberry Pi 3 Model B+ as the central processing unit to provide a hardware-based, real-time method for traffic sign detection and recognition. Our system integrates image collection, preprocessing, and classification into a small, embedded device intended for realistic, realworld deployment, in contrast to conventional software-heavy approaches. The system makes use of a lightweight Convolutional Neural Network (CNN) to effectively categorize traffic signs, employs a camera module to gather visual data, and applies image enhancement techniques to improve clarity. The CNN is perfect for lowpower devices like the Raspberry Pi since it is tuned to balance speed and accuracy. Our test demonstrates that the system can accurately identify and detect a variety of traffic signs in a range of environmental and lighting situations. All things considered, this study provides a scalable, economical, and energyefficient solution for intelligent automobiles and smart traffic systems. In order to achieve even greater performance, future advancements will concentrate on increasing recognition accuracy using more sophisticated deep learning models and utilizing cutting-edge AI technology.