Embedded Artificial Intelligence in autonomous vehicles significantly enhances Advanced Driver Assistance Systems (ADAS), particularly in detecting and classifying road objects, including traffic signs. Traditional approaches to Traffic Sign Classification often address only general situations, highlighting the need for specialized solutions for specific road scenarios. This paper presents a comprehensive approach to Traffic Sign Classification, aimed at developing an intelligent real-time system that detects and classifies road objects and traffic signs using cameras embedded in moving vehicles. Our methodology integrates YOLOv8n, a state-of-the-art object detection model, leveraging the BDD100K dataset for road object detection and the German Traffic Sign Recognition Benchmark (GTSRB) for traffic sign classification. Additionally, we collected and annotated a dataset for the Moroccan Traffic Sign Recognition Benchmark (MTSRB) to address traffic signs written in Arabic for Moroccan road scenarios. The proposed system demonstrates superior performance in Moroccan road scenes compared to existing solutions. It has been evaluated and tested on both an NVIDIA GeForce GPU and an NVIDIA Jetson Xavier AGX board, confirming the reliability and efficiency of our approach.