id	author	title	date	pages	extension	mime	words	sentence	flesch	summary	cache	txt
fcis-14020	He, Chunlin; Wu, Jiaye; Yang, Yujie	Research on Expressway Pavement Crack Detection based on Improved YOLOv5s	2023	7	.pdf	application/pdf	4329	235	51	Firstly, a CBAM attention module was added to the backbone network of the improved YOLOv5 model to obtain more detailed features; Then, in the feature fusion layer, BiFPN weighted bidirectional feature pyramid network is used for multi-scale feature fusion, replacing the traditional feature pyramid (FPN)+pixel aggregation network (PAN) structure to enhance feature fusion. In common application examples, attention modules are often combined with feature fusion operations, but further experiments are needed to determine which position or positions in the network can achieve better results.	cache/fcis-14020.pdf	txt/fcis-14020.txt