Region‐based fully convolutional networks with deformable convolution and attention fusion for steel surface defect detection in industrial Internet of Things

Abstract Next‐generation 6G networks will fully drive the development of the industrial Internet of Things. Steel surface defect detection as an important application in industrial Internet of Things has recently received increasing attention from the military industry, the aviation industry and oth...

Full description

Saved in:
Bibliographic Details
Main Authors: Meixia Fu, Jiansheng Wu, Qu Wang, Lei Sun, Zhangchao Ma, Chaoyi Zhang, Wanqing Guan, Wei Li, Na Chen, Danshi Wang, Jianquan Wang
Format: Article
Language:English
Published: Wiley 2023-05-01
Series:IET Signal Processing
Subjects:
Online Access:https://doi.org/10.1049/sil2.12208
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850164720792764416
author Meixia Fu
Jiansheng Wu
Qu Wang
Lei Sun
Zhangchao Ma
Chaoyi Zhang
Wanqing Guan
Wei Li
Na Chen
Danshi Wang
Jianquan Wang
author_facet Meixia Fu
Jiansheng Wu
Qu Wang
Lei Sun
Zhangchao Ma
Chaoyi Zhang
Wanqing Guan
Wei Li
Na Chen
Danshi Wang
Jianquan Wang
author_sort Meixia Fu
collection DOAJ
description Abstract Next‐generation 6G networks will fully drive the development of the industrial Internet of Things. Steel surface defect detection as an important application in industrial Internet of Things has recently received increasing attention from the military industry, the aviation industry and other fields, which is closely related to the quality of industrial production products. However, many typical convolutional neural networks‐based methods are insensitive to the problem of unclear boundaries. In this article, the authors develop a region‐based fully convolutional networks with deformable convolution and attention fusion to adaptively learn salient features for steel surface defect detection. Specifically, deformable convolution is applied into selectively replace the standard convolution in the backbone of the region‐based fully convolutional networks, which performs significantly in scenarios with unclear defect boundaries. Moreover, convolutional block attention module is utilised in region proposal network to further enhance detection accuracy. The proposed architecture is demonstrated on two popular steel defect detection benchmarks, including NEU‐DET and GC10‐DET, which can effectively present the performance of steel surface defect detection by abundant experiments. The mean average precision on two datasets reaches 80.9% and 66.2%. The average precision of defect crazing, inclusion, patches, pitted‐surface, rolled‐in scale and scratches on NEU‐DET is 58.2%, 82.3%, 95.7%, 85.6%, 75.9%, and 87.9% respectively.
format Article
id doaj-art-e9b37631cfa547af8ab17b51ce7ead2b
institution OA Journals
issn 1751-9675
1751-9683
language English
publishDate 2023-05-01
publisher Wiley
record_format Article
series IET Signal Processing
spelling doaj-art-e9b37631cfa547af8ab17b51ce7ead2b2025-08-20T02:21:54ZengWileyIET Signal Processing1751-96751751-96832023-05-01175n/an/a10.1049/sil2.12208Region‐based fully convolutional networks with deformable convolution and attention fusion for steel surface defect detection in industrial Internet of ThingsMeixia Fu0Jiansheng Wu1Qu Wang2Lei Sun3Zhangchao Ma4Chaoyi Zhang5Wanqing Guan6Wei Li7Na Chen8Danshi Wang9Jianquan Wang10School of Automation and Electrical Engineering Institute of Industrial Internet University of Science and Technology Beijing Beijing ChinaSchool of Automation and Electrical Engineering Institute of Industrial Internet University of Science and Technology Beijing Beijing ChinaSchool of Automation and Electrical Engineering Institute of Industrial Internet University of Science and Technology Beijing Beijing ChinaSchool of Automation and Electrical Engineering Institute of Industrial Internet University of Science and Technology Beijing Beijing ChinaSchool of Automation and Electrical Engineering Institute of Industrial Internet University of Science and Technology Beijing Beijing ChinaSchool of Automation and Electrical Engineering Institute of Industrial Internet University of Science and Technology Beijing Beijing ChinaSchool of Science and Communication Engineering University of Science and Technology Beijing Beijing ChinaSchool of Science and Communication Engineering University of Science and Technology Beijing Beijing ChinaThe Graduate School of Science and Technology Nara Institute of Science and Technology Ikoma JapanState Key Laboratory of Information Photonics and Optical Communications Beijing University of Posts and Telecommunications Beijing ChinaSchool of Automation and Electrical Engineering Institute of Industrial Internet University of Science and Technology Beijing Beijing ChinaAbstract Next‐generation 6G networks will fully drive the development of the industrial Internet of Things. Steel surface defect detection as an important application in industrial Internet of Things has recently received increasing attention from the military industry, the aviation industry and other fields, which is closely related to the quality of industrial production products. However, many typical convolutional neural networks‐based methods are insensitive to the problem of unclear boundaries. In this article, the authors develop a region‐based fully convolutional networks with deformable convolution and attention fusion to adaptively learn salient features for steel surface defect detection. Specifically, deformable convolution is applied into selectively replace the standard convolution in the backbone of the region‐based fully convolutional networks, which performs significantly in scenarios with unclear defect boundaries. Moreover, convolutional block attention module is utilised in region proposal network to further enhance detection accuracy. The proposed architecture is demonstrated on two popular steel defect detection benchmarks, including NEU‐DET and GC10‐DET, which can effectively present the performance of steel surface defect detection by abundant experiments. The mean average precision on two datasets reaches 80.9% and 66.2%. The average precision of defect crazing, inclusion, patches, pitted‐surface, rolled‐in scale and scratches on NEU‐DET is 58.2%, 82.3%, 95.7%, 85.6%, 75.9%, and 87.9% respectively.https://doi.org/10.1049/sil2.12208artificial intelligencecomputer visionimage segmentationobject detection
spellingShingle Meixia Fu
Jiansheng Wu
Qu Wang
Lei Sun
Zhangchao Ma
Chaoyi Zhang
Wanqing Guan
Wei Li
Na Chen
Danshi Wang
Jianquan Wang
Region‐based fully convolutional networks with deformable convolution and attention fusion for steel surface defect detection in industrial Internet of Things
IET Signal Processing
artificial intelligence
computer vision
image segmentation
object detection
title Region‐based fully convolutional networks with deformable convolution and attention fusion for steel surface defect detection in industrial Internet of Things
title_full Region‐based fully convolutional networks with deformable convolution and attention fusion for steel surface defect detection in industrial Internet of Things
title_fullStr Region‐based fully convolutional networks with deformable convolution and attention fusion for steel surface defect detection in industrial Internet of Things
title_full_unstemmed Region‐based fully convolutional networks with deformable convolution and attention fusion for steel surface defect detection in industrial Internet of Things
title_short Region‐based fully convolutional networks with deformable convolution and attention fusion for steel surface defect detection in industrial Internet of Things
title_sort region based fully convolutional networks with deformable convolution and attention fusion for steel surface defect detection in industrial internet of things
topic artificial intelligence
computer vision
image segmentation
object detection
url https://doi.org/10.1049/sil2.12208
work_keys_str_mv AT meixiafu regionbasedfullyconvolutionalnetworkswithdeformableconvolutionandattentionfusionforsteelsurfacedefectdetectioninindustrialinternetofthings
AT jianshengwu regionbasedfullyconvolutionalnetworkswithdeformableconvolutionandattentionfusionforsteelsurfacedefectdetectioninindustrialinternetofthings
AT quwang regionbasedfullyconvolutionalnetworkswithdeformableconvolutionandattentionfusionforsteelsurfacedefectdetectioninindustrialinternetofthings
AT leisun regionbasedfullyconvolutionalnetworkswithdeformableconvolutionandattentionfusionforsteelsurfacedefectdetectioninindustrialinternetofthings
AT zhangchaoma regionbasedfullyconvolutionalnetworkswithdeformableconvolutionandattentionfusionforsteelsurfacedefectdetectioninindustrialinternetofthings
AT chaoyizhang regionbasedfullyconvolutionalnetworkswithdeformableconvolutionandattentionfusionforsteelsurfacedefectdetectioninindustrialinternetofthings
AT wanqingguan regionbasedfullyconvolutionalnetworkswithdeformableconvolutionandattentionfusionforsteelsurfacedefectdetectioninindustrialinternetofthings
AT weili regionbasedfullyconvolutionalnetworkswithdeformableconvolutionandattentionfusionforsteelsurfacedefectdetectioninindustrialinternetofthings
AT nachen regionbasedfullyconvolutionalnetworkswithdeformableconvolutionandattentionfusionforsteelsurfacedefectdetectioninindustrialinternetofthings
AT danshiwang regionbasedfullyconvolutionalnetworkswithdeformableconvolutionandattentionfusionforsteelsurfacedefectdetectioninindustrialinternetofthings
AT jianquanwang regionbasedfullyconvolutionalnetworkswithdeformableconvolutionandattentionfusionforsteelsurfacedefectdetectioninindustrialinternetofthings