A Literature Review on AI-Based Traffic Sign Recognition Systems

Traffic Sign Recognition (TSR) is a fundamental component of intelligent transportation systems and Advanced Driver Assistance Systems (ADAS). Accurate detection and classification of traffic signs enable safer driving, reduce human error, and support autonomous vehicle decision-making. Recent advancements in artificial intelligence, particularly deep learning and computer vision, have significantly improved TSR performance. Techniques such as Convolutional Neural Networks (CNNs), attention-based architectures, hybrid classifiers, and lightweight transformer models enable real-time recognition even on embedded platforms. However, real-world deployment still faces challenges including environmental variability, occlusion, sign degradation, regional diversity, and computational constraints. This manuscript provides a detailed review of recent AI-based TSR approaches, analyzes their advantages and limitations, identifies research gaps, and outlines future directions toward robust, scalable, and real-time TSR systems.