<![CDATA[The advancement of artificial intelligence technologies has driven significant innovation in intelligent transportation systems, particularly in autonomous vehicles that require real-time object detection capabilities. This study develops a web-based traffic sign detection system using the YOLOv8 (You Only Look Once version 8) algorithm. The dataset consists of 5,224 annotated images representing 20 classes of traffic signs, collected through direct image acquisition and managed using Roboflow. The preprocessing stage includes resizing all images to 640×640 pixels and applying data augmentation to enhance model generalization. Model training was conducted on Google Colab using an optimal configuration of 50 epochs, a batch size of 32, and a learning rate of 0.001, resulting in a Precision score of 0.9406, a Recall of 0.9395, and an mAP50 of 0.9748. The trained model was integrated into a Flask-based web application to support image, video, and real-time camera detection. The results indicate that the system is capable of detecting traffic signs with high accuracy and strong computational efficiency.]]>
Copyrights © 2025
