BanglaRoadSignNet: A YOLO-Based Framework for Efficient Road Sign Detection and Recognition in Bangladesh

In developing countries like Bangladesh, the detection and recognition of road signs is a key component of intelligent transportation systems (ITS). This study addresses unique road signs and nonstandardized signage challenges, improving the detection and recognition process of traffic signs. The study highlights the increasing vehicular traffic and the lack of automation in traffic monitoring, which could improve roadway security and traffic management. This paper proposes BanglaRoadSignNet, a robust framework leveraging YOLOv11 (You Only Look Once) for accurate and simultaneous detection and recognition of road signs mainly for Bangladesh roads. BanglaRoadSignNet establishes a link between existing identification frameworks by addressing multilingual signage, occlusions, and environmental conditions, achieving high performance. The proposed model demonstrates significant performance with a precision of 94.3%, recall of 98.8%, mAP50 of 99.0%, and mAP50-95 of 92.6%, surpassing earlier versions of YOLO. The study’s major contributions are the novel architecture of YOLOv11 on a Bangladesh local dataset and its accuracy and reliability in detecting road signs, making it a practical solution for real-time ITS applications in Bangladesh.