Yolov5 Series Algorithm for Road Marking Sign Identification

Road markings and signs provide vehicles pedestrians with essential information that assists them to follow the traffic regulations. surface include pedestrian crossings, directional arrows, zebra speed limit signs, other similar text, so on, which are usually painted directly onto road surface. fulfill a variety of important functions, such as alerting drivers potentially hazardous section, directing traffic, prohibiting certain actions, slowing down. This research paper provides summary Yolov5 algorithm series for marking sign identification, includes Yolov5s, Yolov5m, Yolov5n, Yolov5l, Yolov5x. study explores wide range contemporary object detectors, ones used determine location signs. Performance metrics monitor data, including quantity BFLOPS, mean average precision (mAP), detection time (IoU). Our findings shows Yolov5m is most stable method compared methods 76% precision, 86% recall, 83% mAP during training stage. Moreover, Yolov5l achieve highest score, 87% on in testing In addition, we have created new dataset Taiwan, called TRMSD.