DLF-YOLOF: an improved YOLOF-based surface defect detection for steel plate

Surface defects can affect the quality of steel plate. Many methods based on computer vision are currently applied to surface defect detection of steel plate. However, their real-time performance and object detection of small defect are still unsatisfactory. An improved object detection network base...

Full description

Saved in:
Bibliographic Details
Published in:Journal of iron and steel research, international international, 2024-02, Vol.31 (2), p.442-451
Main Authors: Liu, Guang-hu, Chu, Mao-xiang, Gong, Rong-fen, Zheng, Ze-hao
Format: Article
Language:English
Subjects:
Applied and Technical Physics
Engineering
Machines
Manufacturing
Materials Engineering
Materials Science
Metallic Materials
Original Paper
Physical Chemistry
Processes
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Surface defects can affect the quality of steel plate. Many methods based on computer vision are currently applied to surface defect detection of steel plate. However, their real-time performance and object detection of small defect are still unsatisfactory. An improved object detection network based on You Only Look One-level Feature (YOLOF) is proposed to show excellent performance in surface defect detection of steel plate, called DLF-YOLOF. First, the anchor-free detector is used to reduce the network hyperparameters. Secondly, deformable convolution network and local spatial attention module are introduced into the feature extraction network to increase the contextual information in the feature maps. Also, the soft non-maximum suppression is used to improve detection accuracy significantly. Finally, data augmentation is performed for small defect objects during training to improve detection accuracy. Experiments show the average precision and average precision for small objects are 42.7% and 33.5% at a detection speed of 62 frames per second on a single GPU, respectively. This shows that DLF-YOLOF has excellent performance to meet the needs of industrial real-time detection.
ISSN:1006-706X
2210-3988
DOI:10.1007/s42243-023-01059-4