YOLOv8 Enhanced: Pioneering Accuracy in Traffic Sign Detection and Classification

This research paper considers the You Only Look Once version 8 (YOLOv8) model as a tool for traffic sign detection and classification, which is an important feature of the autonomous driving system and the intelligent transportation system. Analyzing the latest version of the YOLO series, which seeks to maintain a suitable balance between object detection speed and accuracy, is the aim of this study since precise and timely traffic sign recognition is a critical factor for road safety and traffic management. The YOLOv8 model was trained and the accuracy of it was tested using a thorough dataset of traffic signs under different environmental conditions. The approach relied on exploiting YOLOv8's advanced deep learning architecture, transfer learning, and data augmentation to improve the model performance. Statistics revealed that YOLOv8 surpasses the previous versions with its mAP of 95.2% and processing speeds that match real-time application demands. The model had high precision and recall across multiple classes of traffic signs, especially those that are unique in shape or color. This research indicates that YOLOv8 can be an effective tool to enhance the performance of autonomous driving systems by its robust sign detection and classification. The model has opened a new frontier for research on improving the model for more accuracy and further application of the model in more elaborate areas of road safety and intelligent transportation systems.