To address issues such as low recognition accuracy,missed detections,and false positives in road defect detection,the road defect detection model EC-YOLO was proposed.A new C2f_EMA mod-ule was constructed in the backbone network by using the C2f and EMA attention mechanisms.Channel information was preserved by reshaping channels and grouping channel dimensions.An upsampling module called CARAFE was introduced in the neck network.More feature details were retained through feature expansion and recombination.An experiment was conducted to compare EC-YOLO with main-stream target detection models.The experimental results show that the average accuracy of EC-YOLO is 3.4%higher than that of YOLOv8.
万方数据
维普
