Opto-Electronic Engineering, Volume. 51, Issue 5, 240028(2024)
Defects detection for cable surface of cable-stayed bridge based on improved YOLOv5s network
Figures & Tables(19)
Table 1. Prior box size of conventional YOLOv5s
View tableView in ArticleTable 1. Prior box size of conventional YOLOv5s
特征图 尺寸 先验框尺寸 特征图1 80×80 (10, 13), (16, 30), (33, 23) 特征图2 40×40 (30, 61), (62, 45), (59, 119) 特征图3 20×20 (116, 90), (156, 198), (373, 326)
Table 2. Ablation comparative experiment
View tableView in ArticleTable 2. Ablation comparative experiment
算法 AP /% mAP /% FPS a b c d 常规YOLOv5s 90.42 89.08 83.77 91.57 87.71 56 方法1 91.92 92.54 94.01 96.13 93.65 51 方法2 94.09 91.81 90.32 95.62 92.96 64 方法3 93.20 92.23 91.26 95.83 93.13 68 方法4 95.11 91.29 93.42 97.22 94.26 68
Table 3. Detection results of different networks on the surface defect dataset of the cable
View tableView in ArticleTable 3. Detection results of different networks on the surface defect dataset of the cable
网络 mAP /% FPS Faster R-CNN 90.57 3 文献[ 12 ]87.89 16 YOLOv4 89.37 20 常规YOLOv5s 87.71 59 常规YOLOv8s 92.38 51 改进YOLOv5s 94.26 68
Table 4. Detection results of different networks on VOC 2007 dataset
View tableView in ArticleTable 4. Detection results of different networks on VOC 2007 dataset
网络 mAP /% FPS Faster R-CNN 76.38 4 文献[ 12 ]71.91 21 YOLOv4 73.12 26 常规YOLOv5s 72.57 63 常规YOLOv8s 76.82 57 改进YOLOv5s 78.21 71
Tools
Get Citation
Copy Citation Text
Pengfeng Wang, Yuntang Li, Yongyong Huang, Wenkai Zhu, Jie Lin, Binrui Wang. Defects detection for cable surface of cable-stayed bridge based on improved YOLOv5s network[J]. Opto-Electronic Engineering, 2024, 51(5): 240028
Paper Information
Category: Article
Received: Jan. 26, 2024
Accepted: Apr. 24, 2024
Published Online: Jul. 31, 2024
The Author Email: Li Yuntang (李运堂)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20