Improved faster R-CNN for steel surface defect detection in industrial quality control

Abstract Steel surface defect detection constitutes a critical inspection task in industrial production. To address challenges including missed detections and low accuracy for fine defects, this study develops an enhanced Faster R-CNN algorithm. The proposed framework incorporates a feature fusion m...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuefeng Leng, Jiazhi Liu
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-12740-x
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849736291434889216
author Yuefeng Leng
Jiazhi Liu
author_facet Yuefeng Leng
Jiazhi Liu
author_sort Yuefeng Leng
collection DOAJ
description Abstract Steel surface defect detection constitutes a critical inspection task in industrial production. To address challenges including missed detections and low accuracy for fine defects, this study develops an enhanced Faster R-CNN algorithm. The proposed framework incorporates a feature fusion module and lightweight channel attention mechanism between Feature Pyramid Networks (FPN) and Region Proposal Network (RPN), substantially augmenting subtle feature extraction capabilities. Evaluated on the NEU-DET dataset, the optimized model achieves a mean average precision (mAP) of 80.2%-yielding a 12.6% improvement over the baseline-while increasing detection speed by 40.9%. This approach not only significantly elevates defect recognition accuracy but also establishes a practical framework for automated steel surface inspection systems.
format Article
id doaj-art-089194eaf05b4467b4b7e5c3572e97a3
institution DOAJ
issn 2045-2322
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-089194eaf05b4467b4b7e5c3572e97a32025-08-20T03:07:20ZengNature PortfolioScientific Reports2045-23222025-08-0115111110.1038/s41598-025-12740-xImproved faster R-CNN for steel surface defect detection in industrial quality controlYuefeng Leng0Jiazhi Liu1School of Mechanical Engineering, Liaoning Technical UniversitySchool of Mechanical Engineering, Liaoning Technical UniversityAbstract Steel surface defect detection constitutes a critical inspection task in industrial production. To address challenges including missed detections and low accuracy for fine defects, this study develops an enhanced Faster R-CNN algorithm. The proposed framework incorporates a feature fusion module and lightweight channel attention mechanism between Feature Pyramid Networks (FPN) and Region Proposal Network (RPN), substantially augmenting subtle feature extraction capabilities. Evaluated on the NEU-DET dataset, the optimized model achieves a mean average precision (mAP) of 80.2%-yielding a 12.6% improvement over the baseline-while increasing detection speed by 40.9%. This approach not only significantly elevates defect recognition accuracy but also establishes a practical framework for automated steel surface inspection systems.https://doi.org/10.1038/s41598-025-12740-xDefect detectionFeature fusionChannel attention mechanismImproved faster R-CNN algorithm
spellingShingle Yuefeng Leng
Jiazhi Liu
Improved faster R-CNN for steel surface defect detection in industrial quality control
Scientific Reports
Defect detection
Feature fusion
Channel attention mechanism
Improved faster R-CNN algorithm
title Improved faster R-CNN for steel surface defect detection in industrial quality control
title_full Improved faster R-CNN for steel surface defect detection in industrial quality control
title_fullStr Improved faster R-CNN for steel surface defect detection in industrial quality control
title_full_unstemmed Improved faster R-CNN for steel surface defect detection in industrial quality control
title_short Improved faster R-CNN for steel surface defect detection in industrial quality control
title_sort improved faster r cnn for steel surface defect detection in industrial quality control
topic Defect detection
Feature fusion
Channel attention mechanism
Improved faster R-CNN algorithm
url https://doi.org/10.1038/s41598-025-12740-x
work_keys_str_mv AT yuefengleng improvedfasterrcnnforsteelsurfacedefectdetectioninindustrialqualitycontrol
AT jiazhiliu improvedfasterrcnnforsteelsurfacedefectdetectioninindustrialqualitycontrol