Article(id=1276896904998949269, tenantId=1146029695717560320, journalId=1276577754012160025, issueId=1276896661737701828, articleNumber=null, orderNo=null, doi=10.3724/j.gyjzG26033109, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1774886400000, receivedDateStr=2026-03-31, revisedDate=null, revisedDateStr=null, acceptedDate=null, acceptedDateStr=null, onlineDate=1782365545748, onlineDateStr=2026-06-25, pubDate=1779206400000, pubDateStr=2026-05-20, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1782365545748, onlineIssueDateStr=2026-06-25, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1782365545748, creator=13701087609, updateTime=1782365545748, updator=13701087609, issue=Issue{id=1276896661737701828, tenantId=1146029695717560320, journalId=1276577754012160025, year='2026', volume='56', issue='5', pageStart='1', pageEnd='264', issueExtLink='null', onlineDate='null', pubDate='1779206400000', pubDateStr='2026-05-20', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1782365487751, creator='13701087609', updateTime=1782367237543, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext={EN=IssueExt(id=1276904000968589318, tenantId=1146029695717560320, journalId=1276577754012160025, issueId=1276896661737701828, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1276904000968589319, tenantId=1146029695717560320, journalId=1276577754012160025, issueId=1276896661737701828, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null, downloadFileDto=null}, startPage=215, endPage=231, ext={EN=ArticleExt(id=1276896905481294232, articleId=1276896904998949269, tenantId=1146029695717560320, journalId=1276577754012160025, language=EN, title=A Review of Computer Vision-Based Damage Detection in Steel Structures, columnId=null, journalTitle=Industrial Construction, columnName=null, runingTitle=null, highlight=null, articleAbstract=

Efficient and reliable structural health monitoring is essential for ensuring the safety and extending the service life of steel structures. Owing to the advantages of non-contact nature, high efficiency, and a high degree of automation, computer vision (CV) has gradually become an important technology for the inspection and maintenance of steel structures. Focusing on surface cracks and corrosion damage of steel structures, this review systematically summarizes the recent research progress in CV-based damage detection and outlines the major approaches, including image classification, object detection, and image segmentation. Particular attention is paid to key optimization strategies for small object detection, robustness under complex backgrounds, few-shot learning, and on-site deployment. Existing studies indicate that CV has significantly improved the automation, intelligence, and precision of damage detection for steel structures. However, further advances are still required in dataset standardization, model robustness to interference, generalization capability across scenarios, and lightweight real-time inference.

, authors=null, authorsList=Yikang LIU, Mingxuan ZHANG, Qianqian YU, authorCompany=null, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, fund=null), CN=ArticleExt(id=1276896908782211493, articleId=1276896904998949269, tenantId=1146029695717560320, journalId=1276577754012160025, language=CN, title=基于计算机视觉的钢结构损伤检测综述, columnId=1276896684756038168, journalTitle=工业建筑, columnName=工程诊治与运维数智化, runingTitle=null, highlight=null, articleAbstract=

高效可靠的钢结构健康监测是保证结构安全、延长服役寿命的关键。当前,计算机视觉(Computer Vision, CV)技术因其非接触、效率高和自动化程度高等优势,正逐渐成为钢结构运维检测的重要技术手段。围绕钢结构表面裂纹与腐蚀损伤,系统梳理了计算机视觉技术在钢结构损伤检测中的研究进展,归纳了图像分类、目标检测和图像分割等主要方法,重点总结了现有研究在微小目标识别、复杂背景抑制、小样本训练及工程现场部署等方面的关键优化策略。现有研究表明,计算机视觉技术有效提升了钢结构损伤检测的自动化、智能化和精细化水平,但在数据集规范化、模型抗干扰能力、跨场景泛化能力和轻量化实时推理等方面仍有进一步发展空间。

, authors=

刘逸康,硕士研究生,主要从事钢结构性能检测方向研究,

, authorsList=刘逸康, 张铭煊, 余倩倩, authorCompany=null, correspAuthors=null, authorNote=null, correspAuthorsNote=
余倩倩,博士,教授,主要从事结构性能演化与控制方向研究,
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=HWmvLy8U+DwSQGjjQc3qGg==, magXml=tbmmCPcAI0AuzsYr63cqog==, pdfUrl=null, pdf=uBVwKyWrpDCR1+V0hicpeA==, pdfFileSize=5560795, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=33whRWrml7PzQsYG8D+tcw==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=gGini9mtWc1alBlY+iPZvA==, mapNumber=null, fund=null)}, authors=[Author(id=1276896910116000178, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=2430876@tongji.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1276896910476710326, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, authorId=1276896910116000178, language=EN, stringName=Yikang LIU, firstName=Yikang, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1Department of Structural Engineering, Tongji University, Shanghai200092, China
2Key Laboratory of Performance Evolution and Control for Engineering Structures, Tongji University, Shanghai200092, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1276896912146043319, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, authorId=1276896910116000178, language=CN, stringName=刘逸康, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1同济大学建筑工程系,上海200092
2同济大学工程结构性能演化与控制教育部重点实验室,上海200092, bio={"content":"

刘逸康,硕士研究生,主要从事钢结构性能检测方向研究,

"}, bioImg=null, bioContent=

刘逸康,硕士研究生,主要从事钢结构性能检测方向研究,

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1276896909293916583, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, xref=1, ext=[AuthorCompanyExt(id=1276896909302305192, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909293916583, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1Department of Structural Engineering, Tongji University, Shanghai200092, China), AuthorCompanyExt(id=1276896909310693801, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909293916583, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1同济大学建筑工程系,上海200092)]), AuthorCompany(id=1276896909621072298, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, xref=2, ext=[AuthorCompanyExt(id=1276896909633655211, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909621072298, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Performance Evolution and Control for Engineering Structures, Tongji University, Shanghai200092, China), AuthorCompanyExt(id=1276896909646238124, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909621072298, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2同济大学工程结构性能演化与控制教育部重点实验室,上海200092)])]), Author(id=1276896912238318009, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1276896912540307900, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, authorId=1276896912238318009, language=EN, stringName=Mingxuan ZHANG, firstName=Mingxuan, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1Department of Structural Engineering, Tongji University, Shanghai200092, China
2Key Laboratory of Performance Evolution and Control for Engineering Structures, Tongji University, Shanghai200092, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1276896912615805373, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, authorId=1276896912238318009, language=CN, stringName=张铭煊, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1同济大学建筑工程系,上海200092
2同济大学工程结构性能演化与控制教育部重点实验室,上海200092, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1276896909293916583, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, xref=1, ext=[AuthorCompanyExt(id=1276896909302305192, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909293916583, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1Department of Structural Engineering, Tongji University, Shanghai200092, China), AuthorCompanyExt(id=1276896909310693801, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909293916583, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1同济大学建筑工程系,上海200092)]), AuthorCompany(id=1276896909621072298, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, xref=2, ext=[AuthorCompanyExt(id=1276896909633655211, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909621072298, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Performance Evolution and Control for Engineering Structures, Tongji University, Shanghai200092, China), AuthorCompanyExt(id=1276896909646238124, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909621072298, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2同济大学工程结构性能演化与控制教育部重点实验室,上海200092)])]), Author(id=1276896912691302847, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=qianqian.yu@tongji.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1276896913106538946, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, authorId=1276896912691302847, language=EN, stringName=Qianqian YU, firstName=Qianqian, middleName=null, lastName=YU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 3, address=1Department of Structural Engineering, Tongji University, Shanghai200092, China
3State Key Laboratory of Disaster Reduction in Civil Engineering, Tongji University, Shanghai200092, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1276896913395945923, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, authorId=1276896912691302847, language=CN, stringName=余倩倩, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 3, address=1同济大学建筑工程系,上海200092
3同济大学;土木工程防灾减灾全国重点实验室,上海200092, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1276896909293916583, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, xref=1, ext=[AuthorCompanyExt(id=1276896909302305192, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909293916583, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1Department of Structural Engineering, Tongji University, Shanghai200092, China), AuthorCompanyExt(id=1276896909310693801, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909293916583, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1同济大学建筑工程系,上海200092)]), AuthorCompany(id=1276896910027919790, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, xref=3, ext=[AuthorCompanyExt(id=1276896910040502702, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896910027919790, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3State Key Laboratory of Disaster Reduction in Civil Engineering, Tongji University, Shanghai200092, China), AuthorCompanyExt(id=1276896910048891311, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896910027919790, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3同济大学;土木工程防灾减灾全国重点实验室,上海200092)])])], keywords=[Keyword(id=1276896913500803524, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=EN, orderNo=1, keyword=steel structure), Keyword(id=1276896913819570629, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=EN, orderNo=2, keyword=damage detection), Keyword(id=1276896913899262406, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=EN, orderNo=3, keyword=computer vision), Keyword(id=1276896914247389639, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=EN, orderNo=4, keyword=deep learning), Keyword(id=1276896914348052936, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=CN, orderNo=1, keyword=钢结构), Keyword(id=1276896914658431433, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=CN, orderNo=2, keyword=损伤检测), Keyword(id=1276896914742317514, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=CN, orderNo=3, keyword=计算机视觉), Keyword(id=1276896915082056139, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=CN, orderNo=4, keyword=深度学习)], refs=[Reference(id=1276896918097760732, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=8, pageStart=10, pageEnd=13, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=袁宇峰, journalName=冶金管理, refType=null, unstructuredReference=袁宇峰. 我国钢结构产业发展现状及其用钢需求调查[J]. 冶金管理2024(8): 10-13., articleTitle=我国钢结构产业发展现状及其用钢需求调查, refAbstract=null), Reference(id=1276896918177452509, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2025, volume=null, issue=7, pageStart=62, pageEnd=68, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=中国建筑金属结构协会建筑钢结构行业可持续发展研究课题组, journalName=建筑, refType=null, unstructuredReference=中国建筑金属结构协会建筑钢结构行业可持续发展研究课题组. 中国建筑钢结构行业发展报告(2023—2024年度)[J]. 建筑2025(7): 62-68., articleTitle=中国建筑钢结构行业发展报告(2023—2024年度), refAbstract=null), Reference(id=1276896918252949982, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=幸坤涛, 赵晓青, 郭小华, journalName=null, refType=null, unstructuredReference=幸坤涛, 赵晓青, 郭小华, . 工业建筑钢结构疲劳损伤检测、评估及加固关键技术研究[C]//第十二届中国钢铁年会论文集. 北京: 2019., articleTitle=工业建筑钢结构疲劳损伤检测、评估及加固关键技术研究, refAbstract=null), Reference(id=1276896918320058847, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2025, volume=37, issue=5, pageStart=539, pageEnd=556, url=null, language=null, rfNumber=[4], rfOrder=3, authorNames=何润, 周世康, 张琦超, journalName=钢铁研究学报, refType=null, unstructuredReference=何润, 周世康, 张琦超, . 桥梁钢结构的腐蚀与防护技术研究进展[J]. 钢铁研究学报202537(5): 539-556., articleTitle=桥梁钢结构的腐蚀与防护技术研究进展, refAbstract=null), Reference(id=1276896918395556320, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=1, pageStart=1, pageEnd=16, url=null, language=null, rfNumber=[5], rfOrder=4, authorNames=WU R K, ZHANG H, YANG R Z, journalName=Journal of Sensors, refType=null, unstructuredReference=WU R KZHANG HYANG R Zet al. Nondestructive testing for corrosion evaluation of metal under coating[J]. Journal of Sensors2021,2021(1): 1-16., articleTitle=Nondestructive testing for corrosion evaluation of metal under coating, refAbstract=null), Reference(id=1276896918458470881, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=59, issue=1, pageStart=56, pageEnd=85, url=null, language=null, rfNumber=[6], rfOrder=5, authorNames=VASAGAR V, HASSAN M K, ABDULLAH A M, journalName=Corrosion Engineering, refType=null, unstructuredReference=VASAGAR VHASSAN M KABDULLAH A Met al. Non-destructive techniques for corrosion detection: a review[J]. Corrosion Engineering, Science and Technology, 202459(1): 56-85., articleTitle=Non-destructive techniques for corrosion detection: a review, refAbstract=null), Reference(id=1276896918521385442, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2025, volume=55, issue=7, pageStart=131, pageEnd=142, url=null, language=null, rfNumber=[7], rfOrder=6, authorNames=陈飞圻, 薛江, 逯鹏, journalName=工业建筑, refType=null, unstructuredReference=陈飞圻, 薛江, 逯鹏, . 基于机器视觉的钢结构工程运维关键技术研究现状[J]. 工业建筑202555(7): 131-142., articleTitle=基于机器视觉的钢结构工程运维关键技术研究现状, refAbstract=null), Reference(id=1276896918592688611, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=40, issue=10, pageStart=97, pageEnd=110, url=null, language=null, rfNumber=[8], rfOrder=7, authorNames=朱洪洲, 谭祺琦, 范世平, journalName=重庆交通大学学报 (自然科学版), refType=null, unstructuredReference=朱洪洲, 谭祺琦, 范世平, . 基于图像技术的沥青混合料细观结构研究进展[J]. 重庆交通大学学报 (自然科学版)202140(10): 97-110., articleTitle=基于图像技术的沥青混合料细观结构研究进展, refAbstract=null), Reference(id=1276896918659797476, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2016, volume=37, issue=增刊1, pageStart=113, pageEnd=117, url=null, language=null, rfNumber=[9], rfOrder=8, authorNames=宋伟, 左丹, 邓邦飞, journalName=仪器仪表学报, refType=null, unstructuredReference=宋伟, 左丹, 邓邦飞, . 高压输电线防震锤锈蚀缺陷检测[J]. 仪器仪表学报201637(增刊1): 113-117., articleTitle=高压输电线防震锤锈蚀缺陷检测, refAbstract=null), Reference(id=1276896918726906341, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2016, volume=111, issue=null, pageStart=275, pageEnd=287, url=null, language=null, rfNumber=[10], rfOrder=9, authorNames=XU Y, LI H, LI S, journalName=Corrosion Science, refType=null, unstructuredReference=XU YLI HLI Set al. 3-D modelling and statistical properties of surface pits of corroded wire based on image processing technique[J]. Corrosion Science2016111: 275-287., articleTitle=3-D modelling and statistical properties of surface pits of corroded wire based on image processing technique, refAbstract=null), Reference(id=1276896918819181030, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2013, volume=33, issue=5, pageStart=407, pageEnd=412, url=null, language=null, rfNumber=[11], rfOrder=10, authorNames=刘涛, 艾军, 张丽芳, journalName=中国腐蚀与防护学报, refType=null, unstructuredReference=刘涛, 艾军, 张丽芳, . 基于图像处理技术的钢箱梁防腐涂层寿命预测实验研究[J]. 中国腐蚀与防护学报201333(5): 407-412., articleTitle=基于图像处理技术的钢箱梁防腐涂层寿命预测实验研究, refAbstract=null), Reference(id=1276896918903067111, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2013, volume=34, issue=11, pageStart=997, pageEnd=1000, url=null, language=null, rfNumber=[12], rfOrder=11, authorNames=刘淼, 薛建军, 王玲, journalName=腐蚀与防护, refType=null, unstructuredReference=刘淼, 薛建军, 王玲, . 基于数字图像分析的碳钢腐蚀等级评定方法[J]. 腐蚀与防护201334(11): 997-1000., articleTitle=基于数字图像分析的碳钢腐蚀等级评定方法, refAbstract=null), Reference(id=1276896918970175976, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2016, volume=23, issue=11, pageStart=2867, pageEnd=2875, url=null, language=null, rfNumber=[13], rfOrder=12, authorNames=SHI T, KONG J Y, WANG X D, journalName=Journal of Central South University, refType=null, unstructuredReference=SHI TKONG J YWANG X Det al. Improved Sobel algorithm for defect detection of rail surfaces with enhanced efficiency and accuracy[J]. Journal of Central South University201623(11): 2867-2875., articleTitle=Improved Sobel algorithm for defect detection of rail surfaces with enhanced efficiency and accuracy, refAbstract=null), Reference(id=1276896919041479145, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=53, issue=24, pageStart=126, pageEnd=135, url=null, language=null, rfNumber=[14], rfOrder=13, authorNames=姚志东, 卢佳祁, 熊梦雅, journalName=建筑结构, refType=null, unstructuredReference=姚志东, 卢佳祁, 熊梦雅, . 基于计算机视觉的钢结构表面缺陷智能识别研究综述[J]. 建筑结构202353(24): 126-135., articleTitle=基于计算机视觉的钢结构表面缺陷智能识别研究综述, refAbstract=null), Reference(id=1276896919104393706, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=52, issue=10, pageStart=22, pageEnd=27, url=null, language=null, rfNumber=[15], rfOrder=14, authorNames=逯鹏, 赵天淞, 王剑, journalName=工业建筑, refType=null, unstructuredReference=逯鹏, 赵天淞, 王剑, . 基于计算机视觉的钢结构表面损伤识别与健康监测综述[J]. 工业建筑202252(10): 22-27., articleTitle=基于计算机视觉的钢结构表面损伤识别与健康监测综述, refAbstract=null), Reference(id=1276896919171502571, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=52, issue=6, pageStart=1002, pageEnd=1039, url=null, language=null, rfNumber=[16], rfOrder=15, authorNames=罗东亮, 蔡雨萱, 杨子豪, journalName=中国科学:信息科学, refType=null, unstructuredReference=罗东亮, 蔡雨萱, 杨子豪, . 工业缺陷检测深度学习方法综述[J]. 中国科学:信息科学202252(6): 1002-1039., articleTitle=工业缺陷检测深度学习方法综述, refAbstract=null), Reference(id=1276896919242805740, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=4, pageStart=143, pageEnd=156, url=null, language=null, rfNumber=[17], rfOrder=16, authorNames=杨泽青, 张明轩, 陈英姝, journalName=现代制造工程, refType=null, unstructuredReference=杨泽青, 张明轩, 陈英姝, . 基于机器视觉的表面缺陷检测方法研究进展[J]. 现代制造工程2023(4): 143-156., articleTitle=基于机器视觉的表面缺陷检测方法研究进展, refAbstract=null), Reference(id=1276896919322497517, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=40, issue=4, pageStart=967, pageEnd=977, url=null, language=null, rfNumber=[18], rfOrder=17, authorNames=程锦锋, 方贵盛, 高惠芳, journalName=计算机应用研究, refType=null, unstructuredReference=程锦锋, 方贵盛, 高惠芳. 表面缺陷检测的机器视觉技术研究进展[J]. 计算机应用研究202340(4): 967-977., articleTitle=表面缺陷检测的机器视觉技术研究进展, refAbstract=null), Reference(id=1276896919389606382, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=27, issue=2, pageStart=27, pageEnd=36, url=null, language=null, rfNumber=[19], rfOrder=18, authorNames=高艺平, 王浩, 李新宇, journalName=工业工程, refType=null, unstructuredReference=高艺平, 王浩, 李新宇, . 基于深度智能视觉的表面缺陷检测研究进展[J]. 工业工程202427(2): 27-36., articleTitle=基于深度智能视觉的表面缺陷检测研究进展, refAbstract=null), Reference(id=1276896921046356463, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2025, volume=31, issue=3, pageStart=721, pageEnd=745, url=null, language=null, rfNumber=[20], rfOrder=19, authorNames=邓志鹏, 何施茗, 杨根, journalName=计算机集成制造系统, refType=null, unstructuredReference=邓志鹏, 何施茗, 杨根, . 基于深度学习的纹理表面缺陷检测方法综述[J]. 计算机集成制造系统202531(3): 721-745., articleTitle=基于深度学习的纹理表面缺陷检测方法综述, refAbstract=null), Reference(id=1276896921130242544, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2020, volume=20, issue=10, pageStart=2778, pageEnd=null, url=null, language=null, rfNumber=[21], rfOrder=20, authorNames=AZIMI M, ESLAMLOU A D, PEKCAN G, journalName=Sensors, refType=null, unstructuredReference=AZIMI MESLAMLOU A DPEKCAN G. Data-driven structural health monitoring and damage detection through deep learning: state-of-the-art review[J]. Sensors202020(10): 2778., articleTitle=Data-driven structural health monitoring and damage detection through deep learning: state-of-the-art review, refAbstract=null), Reference(id=1276896921251877361, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2020, volume=48, issue=3, pageStart=590, pageEnd=601, url=null, language=null, rfNumber=[22], rfOrder=21, authorNames=刘颖, 刘红燕, 范九伦, journalName=电子学报, refType=null, unstructuredReference=刘颖, 刘红燕, 范九伦, . 基于深度学习的小目标检测研究与应用综述[J]. 电子学报202048(3): 590-601., articleTitle=基于深度学习的小目标检测研究与应用综述, refAbstract=null), Reference(id=1276896921327374834, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=9, issue=3, pageStart=388, pageEnd=406, url=null, language=null, rfNumber=[23], rfOrder=22, authorNames=ZHANG G, LIU Y, LIU J, journalName=Journal of Traffic and Transportation Engineering (English Edition), refType=null, unstructuredReference=ZHANG GLIU YLIU Jet al. Causes and statistical characteristics of bridge failures: a review[J]. Journal of Traffic and Transportation Engineering (English Edition)20229(3): 388-406., articleTitle=Causes and statistical characteristics of bridge failures: a review, refAbstract=null), Reference(id=1276896921386095091, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2010, volume=null, issue=null, pageStart=3275, pageEnd=3282, url=null, language=null, rfNumber=[24], rfOrder=23, authorNames=IMAM B, CHRYSSANTHOPOULOS M K, journalName=null, refType=null, unstructuredReference=IMAM BCHRYSSANTHOPOULOS M K. A review of metallic bridge failure statistics[C]//Bridge Maintenance, Safety and Management: Proceedings of the Fifth International IABMAS Conference. Boca Raton: 2010: 3275-3282., articleTitle=A review of metallic bridge failure statistics, refAbstract=null), Reference(id=1276896921457398260, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=12, issue=3, pageStart=1374, pageEnd=null, url=null, language=null, rfNumber=[25], rfOrder=24, authorNames=HAMISHEBAHAR Y, GUAN H, SO S, journalName=Applied Sciences, refType=null, unstructuredReference=HAMISHEBAHAR YGUAN H, SO S, et al. A comprehensive review of deep learning-based crack detection approaches[J]. Applied Sciences202212(3): 1374., articleTitle=A comprehensive review of deep learning-based crack detection approaches, refAbstract=null), Reference(id=1276896921520312821, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=16, issue=16, pageStart=2910, pageEnd=null, url=null, language=null, rfNumber=[26], rfOrder=25, authorNames=YUAN Q, SHI Y, LI M, journalName=Remote Sensing, refType=null, unstructuredReference=YUAN QSHI YLI M. A review of computer vision-based crack detection methods in civil infrastructure: progress and challenges[J]. Remote Sensing202416(16): 2910., articleTitle=A review of computer vision-based crack detection methods in civil infrastructure: progress and challenges, refAbstract=null), Reference(id=1276896921608393206, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=11, issue=2, pageStart=188, pageEnd=208, url=null, language=null, rfNumber=[27], rfOrder=26, authorNames=CUI C, ZHANG Q, ZHANG D, journalName=Journal of Traffic and Transportation Engineering (English Edition), refType=null, unstructuredReference=CUI CZHANG QZHANG Det al. Monitoring and detection of steel bridge diseases: a review[J]. Journal of Traffic and Transportation Engineering (English Edition)202411(2): 188-208., articleTitle=Monitoring and detection of steel bridge diseases: a review, refAbstract=null), Reference(id=1276896921671307767, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2012, volume=null, issue=null, pageStart=1097, pageEnd=1105, url=null, language=null, rfNumber=[28], rfOrder=27, authorNames=RIZHEVSKY A, SUTSKEVER I, HINTON G E, journalName=null, refType=null, unstructuredReference=RIZHEVSKY ASUTSKEVER IHINTON G E. ImageNet classification with deep convolutional neural networks[C]//Proceedings of the 25th International Conference on Neural Information Processing Systems. New York: 2012: 1097-1105., articleTitle=ImageNet classification with deep convolutional neural networks, refAbstract=null), Reference(id=1276896921750999544, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[29], rfOrder=28, authorNames=SIMONYAN K, ZISSERMAN A, journalName=V6. arXiv, refType=null, unstructuredReference=SIMONYAN KZISSERMAN A. Very deep convolutional networks for large-scale image recognition[PP/OL]. V6. arXiv (2015-04-10) [2026-05-07]. https://doi.org/10.48550/arXiv.1409.1556., articleTitle=Very deep convolutional networks for large-scale image recognition, refAbstract=null), Reference(id=1276896921826497017, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2015, volume=null, issue=null, pageStart=1, pageEnd=9, url=null, language=null, rfNumber=[30], rfOrder=29, authorNames=SZEGEDY C, LIU W, JIA Y, journalName=null, refType=null, unstructuredReference=SZEGEDY CLIU WJIA Yet al. Going deeper with convolutions[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: 2015: 1-9., articleTitle=Going deeper with convolutions, refAbstract=null), Reference(id=1276896921910383098, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=770, pageEnd=778, url=null, language=null, rfNumber=[31], rfOrder=30, authorNames=HE K, ZHANG X, REN S, journalName=null, refType=null, unstructuredReference=HE KZHANG XREN Set al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: 2016: 770-778., articleTitle=Deep residual learning for image recognition, refAbstract=null), Reference(id=1276896922002657787, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=4700, pageEnd=4708, url=null, language=null, rfNumber=[32], rfOrder=31, authorNames=HUANG G, LIU Z, VAN DER MAATEN L, journalName=null, refType=null, unstructuredReference=HUANG GLIU ZVAN DER MAATEN Let al. Densely connected convolutional networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: 2017: 4700-4708., articleTitle=Densely connected convolutional networks, refAbstract=null), Reference(id=1276896922078155260, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=220, pageEnd=234, url=null, language=null, rfNumber=[33], rfOrder=32, authorNames=VANNOCCI M, RITACCO A, CASTELLANO A, journalName=null, refType=null, unstructuredReference=VANNOCCI MRITACCO ACASTELLANO Aet al. Flatness defect detection and classification in hot rolled steel strips using convolutional neural networks[C]//International Work-Conference on Artificial Neural Networks. Cham: Springer International Publishing, 2019: 220-234., articleTitle=Flatness defect detection and classification in hot rolled steel strips using convolutional neural networks, refAbstract=null), Reference(id=1276896922162041341, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2019, volume=18, issue=3, pageStart=653, pageEnd=674, url=null, language=null, rfNumber=[34], rfOrder=33, authorNames=XU Y, BAO Y, CHEN J, journalName=Structural Health Monitoring, refType=null, unstructuredReference=XU YBAO YCHEN Jet al. Surface fatigue crack identification in steel box girder of bridges by a deep fusion convolutional neural network based on consumer-grade camera images[J]. Structural Health Monitoring201918(3): 653-674., articleTitle=Surface fatigue crack identification in steel box girder of bridges by a deep fusion convolutional neural network based on consumer-grade camera images, refAbstract=null), Reference(id=1276896922229150206, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2017, volume=39, issue=6, pageStart=1137, pageEnd=1149, url=null, language=null, rfNumber=[35], rfOrder=34, authorNames=REN S, HE K, GIRSHICK R, journalName=IEEE Transactions on Pattern Analysis and Machine Intelligence, refType=null, unstructuredReference=REN SHE KGIRSHICK Ret al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence201739(6): 1137-1149., articleTitle=Faster R-CNN: towards real-time object detection with region proposal networks, refAbstract=null), Reference(id=1276896922485002751, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=40, issue=2, pageStart=262, pageEnd=269, url=null, language=null, rfNumber=[36], rfOrder=35, authorNames=王海云, 王剑平, 罗富华, journalName=机械科学与技术, refType=null, unstructuredReference=王海云, 王剑平, 罗富华. 融合多层次特征Faster R-CNN的金属板带材表面缺陷检测研究[J]. 机械科学与技术202140(2): 262-269., articleTitle=融合多层次特征Faster R-CNN的金属板带材表面缺陷检测研究, refAbstract=null), Reference(id=1276896922552111616, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2020, volume=49, issue=10, pageStart=362, pageEnd=371, url=null, language=null, rfNumber=[37], rfOrder=36, authorNames=戴学丰, 陈慧, 朱成军, journalName=表面技术, refType=null, unstructuredReference=戴学丰, 陈慧, 朱成军. 基于改进Faster R-CNN的金属工件表面缺陷检测及实现[J]. 表面技术202049(10): 362-371., articleTitle=基于改进Faster R-CNN的金属工件表面缺陷检测及实现, refAbstract=null), Reference(id=1276896922615026177, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=779, pageEnd=788, url=null, language=null, rfNumber=[38], rfOrder=37, authorNames=REDMON J, DIVVALA S, GIRSHICK R, journalName=null, refType=null, unstructuredReference=REDMON JDIVVALA SGIRSHICK Ret al. You only look once: unified, real-time object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: 2016: 779-788., articleTitle=You only look once: unified, real-time object detection, refAbstract=null), Reference(id=1276896922703106562, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=7263, pageEnd=7271, url=null, language=null, rfNumber=[39], rfOrder=38, authorNames=REDMON J, FARHADI A, journalName=null, refType=null, unstructuredReference=REDMON JFARHADI A. YOLO9000: better, faster, stronger[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: 2017: 7263-7271., articleTitle=YOLO9000: better, faster, stronger, refAbstract=null), Reference(id=1276896922795381251, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=7462, pageEnd=7475, url=null, language=null, rfNumber=[40], rfOrder=39, authorNames=WANG C Y, BOCHKOVSKIY A, LIAO H Y M, journalName=null, refType=null, unstructuredReference=WANG C YBOCHKOVSKIY ALIAO H Y M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Vancouver: 2023: 7462-7475., articleTitle=YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors, refAbstract=null), Reference(id=1276896922879267332, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2018, volume=33, issue=9, pageStart=731, pageEnd=747, url=null, language=null, rfNumber=[41], rfOrder=40, authorNames=CHA Y J, CHOI W, SUH G, journalName=Computer-Aided Civil and Infrastructure Engineering, refType=null, unstructuredReference=CHA Y JCHOI W, SUH G, et al. Autonomous structural visual inspection using region-based deep learning for detecting multiple damage types[J]. Computer-Aided Civil and Infrastructure Engineering201833(9): 731-747., articleTitle=Autonomous structural visual inspection using region-based deep learning for detecting multiple damage types, refAbstract=null), Reference(id=1276896922950570501, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2018, volume=10598, issue=null, pageStart=197, pageEnd=204, url=null, language=null, rfNumber=[42], rfOrder=41, authorNames=SUH G, CHA Y J, journalName=null, refType=null, unstructuredReference=SUH G, CHA Y J. Deep Faster R-CNN-based automated detection and localization of multiple types of damage[C]//Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2018. Bellingham: 201810598: 197-204., articleTitle=Deep Faster R-CNN-based automated detection and localization of multiple types of damage, refAbstract=null), Reference(id=1276896923017679366, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2015, volume=null, issue=null, pageStart=3431, pageEnd=3440, url=null, language=null, rfNumber=[43], rfOrder=42, authorNames=LONG J, SHELHAMER E, DARRELL T, journalName=null, refType=null, unstructuredReference=LONG JSHELHAMER EDARRELL T. Fully convolutional networks for semantic segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: 2015: 3431-3440., articleTitle=Fully convolutional networks for semantic segmentation, refAbstract=null), Reference(id=1276896923088982535, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2015, volume=null, issue=null, pageStart=234, pageEnd=241, url=null, language=null, rfNumber=[44], rfOrder=43, authorNames=RONNEBERGER O, FISCHER P, BROX T, journalName=null, refType=null, unstructuredReference=RONNEBERGER OFISCHER PBROX T. U-Net: convolutional networks for biomedical image segmentation[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention. Cham: 2015: 234-241., articleTitle=U-Net: convolutional networks for biomedical image segmentation, refAbstract=null), Reference(id=1276896923151897096, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=801, pageEnd=818, url=null, language=null, rfNumber=[45], rfOrder=44, authorNames=CHEN L C, ZHU Y, PAPANDREOU G, journalName=null, refType=null, unstructuredReference=CHEN L CZHU YPAPANDREOU Get al. Encoder-decoder with atrous separable convolution for semantic image segmentation[C]//Proceedings of the European conference on computer vision (ECCV). Cham: 2018: 801-818., articleTitle=Encoder-decoder with atrous separable convolution for semantic image segmentation, refAbstract=null), Reference(id=1276896923214811657, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=14, issue=18, pageStart=8132, pageEnd=null, url=null, language=null, rfNumber=[46], rfOrder=45, authorNames=JIA X, WANG Y, WANG Z, journalName=Applied Sciences, refType=null, unstructuredReference=JIA XWANG YWANG Z. Fatigue crack detection based on semantic segmentation using DeepLabV3+ for steel girder bridges[J]. Applied Sciences202414(18): 8132., articleTitle=Fatigue crack detection based on semantic segmentation using DeepLabV3+ for steel girder bridges, refAbstract=null), Reference(id=1276896923273531914, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=37, issue=11, pageStart=1468, pageEnd=1487, url=null, language=null, rfNumber=[47], rfOrder=46, authorNames=PAN Y, ZHANG L, journalName=Computer‐Aided Civil and Infrastructure Engineering, refType=null, unstructuredReference=PAN YZHANG L. Dual attention deep learning network for automatic steel surface defect segmentation[J]. Computer‐Aided Civil and Infrastructure Engineering202237(11): 1468-1487., articleTitle=Dual attention deep learning network for automatic steel surface defect segmentation, refAbstract=null), Reference(id=1276896923340640779, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2019, volume=338, issue=null, pageStart=139, pageEnd=153, url=null, language=null, rfNumber=[48], rfOrder=47, authorNames=LIU Y, YAO J, LU X, journalName=Neurocomputing, refType=null, unstructuredReference=LIU YYAO JLU Xet al. DeepCrack: a deep hierarchical feature learning architecture for crack segmentation[J]. Neurocomputing2019338: 139-153., articleTitle=DeepCrack: a deep hierarchical feature learning architecture for crack segmentation, refAbstract=null), Reference(id=1276896923416138252, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=21, issue=12, pageStart=4135, pageEnd=null, url=null, language=null, rfNumber=[49], rfOrder=48, authorNames=DONG C, LI L, YAN J, journalName=Sensors, refType=null, unstructuredReference=DONG CLI LYAN Jet al. Pixel-level fatigue crack segmentation in large-scale images of steel structures using an encoder-decoder network[J]. Sensors202121(12): 4135., articleTitle=Pixel-level fatigue crack segmentation in large-scale images of steel structures using an encoder-decoder network, refAbstract=null), Reference(id=1276896923495830029, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=202, issue=null, pageStart=111805, pageEnd=null, url=null, language=null, rfNumber=[50], rfOrder=49, authorNames=ZHANG C, WAN L, WAN R Q, journalName=Measurement, refType=null, unstructuredReference=ZHANG CWAN LWAN R Qet al. Automated fatigue crack detection in steel box girder of bridges based on ensemble deep neural network[J]. Measurement2022202: 111805., articleTitle=Automated fatigue crack detection in steel box girder of bridges based on ensemble deep neural network, refAbstract=null), Reference(id=1276896923558744590, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2025, volume=170, issue=null, pageStart=105896, pageEnd=null, url=null, language=null, rfNumber=[51], rfOrder=50, authorNames=KOMPANETS A, DUITS R, PAI G, journalName=Automation in Construction, refType=null, unstructuredReference=KOMPANETS ADUITS R, PAI G, et al. Loss function inversion for improved crack segmentation in steel bridges using a CNN framework[J]. Automation in Construction2025170: 105896., articleTitle=Loss function inversion for improved crack segmentation in steel bridges using a CNN framework, refAbstract=null), Reference(id=1276896923617464847, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=44, issue=3, pageStart=29, pageEnd=36, url=null, language=null, rfNumber=[52], rfOrder=51, authorNames=舒江鹏, 李俊, 马亥波, journalName=土木与环境工程学报(中英文), refType=null, unstructuredReference=舒江鹏, 李俊, 马亥波,. 基于特征金字塔网络的超大尺寸图像裂纹识别检测方法[J]. 土木与环境工程学报(中英文)202244(3): 29-36., articleTitle=基于特征金字塔网络的超大尺寸图像裂纹识别检测方法, refAbstract=null), Reference(id=1276896923697156624, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=2117, pageEnd=2125, url=null, language=null, rfNumber=[53], rfOrder=52, authorNames=LIN T Y, DOLLÁR P, GIRSHICK R, journalName=null, refType=null, unstructuredReference=LIN T YDOLLÁR PGIRSHICK Ret al. Feature pyramid networks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017: 2117-2125., articleTitle=Feature pyramid networks for object detection, refAbstract=null), Reference(id=1276896923877511697, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=2021, issue=1, pageStart=5592878, pageEnd=null, url=null, language=null, rfNumber=[54], rfOrder=53, authorNames=ZHAO W, CHEN F, HUANG H, journalName=Computational Intelligence and Neuroscience, refType=null, unstructuredReference=ZHAO WCHEN FHUANG Het al. A new steel defect detection algorithm based on deep learning[J]. Computational Intelligence and Neuroscience20212021(1): 5592878., articleTitle=A new steel defect detection algorithm based on deep learning, refAbstract=null), Reference(id=1276896923940426258, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=176, issue=null, pageStart=109171, pageEnd=null, url=null, language=null, rfNumber=[55], rfOrder=54, authorNames=LI G, LI X, ZHOU J, journalName=Measurement, refType=null, unstructuredReference=LI GLI XZHOU Jet al. Pixel-level bridge crack detection using a deep fusion about recurrent residual convolution and context encoder network[J]. Measurement2021176: 109171., articleTitle=Pixel-level bridge crack detection using a deep fusion about recurrent residual convolution and context encoder network, refAbstract=null), Reference(id=1276896925588787731, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=12, issue=1, pageStart=127, pageEnd=140, url=null, language=null, rfNumber=[56], rfOrder=55, authorNames=QUQA S, MARTAKIS P, MOVSESSIAN A, journalName=Journal of Civil Structural Health Monitoring, refType=null, unstructuredReference=QUQA SMARTAKIS PMOVSESSIAN Aet al. Two-step approach for fatigue crack detection in steel bridges using convolutional neural networks[J]. Journal of Civil Structural Health Monitoring202212(1): 127-140., articleTitle=Two-step approach for fatigue crack detection in steel bridges using convolutional neural networks, refAbstract=null), Reference(id=1276896925668479508, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=1, issue=2, pageStart=37, pageEnd=51, url=null, language=null, rfNumber=[57], rfOrder=56, authorNames=TONG T, LIN J, HUA J, journalName=Maintenance, Reliability and Condition Monitoring, refType=null, unstructuredReference=TONG TLIN JHUA Jet al. Crack identification for bridge condition monitoring using deep convolutional networks trained with a feedback-update strategy[J]. Maintenance, Reliability and Condition Monitoring20211(2): 37-51., articleTitle=Crack identification for bridge condition monitoring using deep convolutional networks trained with a feedback-update strategy, refAbstract=null), Reference(id=1276896925735588373, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=29, issue=1, pageStart=29, pageEnd=39, url=null, language=null, rfNumber=[58], rfOrder=57, authorNames=MENG S Q, GAO Z Y, ZHOU Y, journalName=Smart Structures and Systems, refType=null, unstructuredReference=MENG S QGAO Z YZHOU Yet al. A three-stage deep-learning-based method for crack detection of high-resolution steel box girder image[J]. Smart Structures and Systems202229(1): 29-39., articleTitle=A three-stage deep-learning-based method for crack detection of high-resolution steel box girder image, refAbstract=null), Reference(id=1276896925794308630, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=29, issue=1, pageStart=237, pageEnd=250, url=null, language=null, rfNumber=[59], rfOrder=58, authorNames=ZHAI G H, NARAZAKI Y, WANG S, journalName=Smart Structures and Systems, refType=null, unstructuredReference=ZHAI G HNARAZAKI YWANG Set al. Synthetic data augmentation for pixel-wise steel fatigue crack identification using fully convolutional networks[J]. Smart Structures and Systems202229(1): 237-250., articleTitle=Synthetic data augmentation for pixel-wise steel fatigue crack identification using fully convolutional networks, refAbstract=null), Reference(id=1276896925865611799, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=30, issue=1, pageStart=17, pageEnd=34, url=null, language=null, rfNumber=[60], rfOrder=59, authorNames=TA Q B, DANG N L, KIM Y C, journalName=Smart Structures and Systems, refType=null, unstructuredReference=TA Q BDANG N LKIM Y Cet al. Semantic crack-image identification framework for steel structures using atrous convolution-based Deeplabv3+ Network[J]. Smart Structures and Systems202230(1): 17-34., articleTitle=Semantic crack-image identification framework for steel structures using atrous convolution-based Deeplabv3+ Network, refAbstract=null), Reference(id=1276896925920137752, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=50, issue=8, pageStart=66, pageEnd=72, url=null, language=null, rfNumber=[61], rfOrder=60, authorNames=邓露, 香超, 王维, journalName=华中科技大学学报(自然科学版), refType=null, unstructuredReference=邓露, 香超, 王维,. 基于改进编解码网络的钢箱梁疲劳裂纹分割[J]. 华中科技大学学报(自然科学版)202250(8): 66-72., articleTitle=基于改进编解码网络的钢箱梁疲劳裂纹分割, refAbstract=null), Reference(id=1276896925999829529, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=41, issue=3, pageStart=52, pageEnd=63, url=null, language=null, rfNumber=[62], rfOrder=61, authorNames=朱劲松, 李欢, 王世芳, journalName=长安大学学报(自然科学版), refType=null, unstructuredReference=朱劲松, 李欢, 王世芳. 基于卷积神经网络和迁移学习的钢桥病害识别[J]. 长安大学学报(自然科学版)202141(3): 52-63., articleTitle=基于卷积神经网络和迁移学习的钢桥病害识别, refAbstract=null), Reference(id=1276896926066938394, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2019, volume=102, issue=null, pageStart=217, pageEnd=229, url=null, language=null, rfNumber=[63], rfOrder=62, authorNames=DUNG C V, SEKIYA H, HIRANO S, journalName=Automation in Construction, refType=null, unstructuredReference=DUNG C VSEKIYA HHIRANO Set al. A vision-based method for crack detection in gusset plate welded joints of steel bridges using deep convolutional neural networks[J]. Automation in Construction2019102: 217-229., articleTitle=A vision-based method for crack detection in gusset plate welded joints of steel bridges using deep convolutional neural networks, refAbstract=null), Reference(id=1276896926129852955, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2025, volume=2025, issue=1, pageStart=7487687, pageEnd=null, url=null, language=null, rfNumber=[64], rfOrder=63, authorNames=YU Q Q, WANG J, GU X L, journalName=Structural Control and Health Monitoring, refType=null, unstructuredReference=YU Q QWANG JGU X Let al. An attention-based detection method of fatigue cracks on steel[J]. Structural Control and Health Monitoring20252025(1): 7487687., articleTitle=An attention-based detection method of fatigue cracks on steel, refAbstract=null), Reference(id=1276896926217933340, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=5828, pageEnd=5839, url=null, language=null, rfNumber=[65], rfOrder=64, authorNames=DAI A, CHANG A X, SAVVA M, journalName=null, refType=null, unstructuredReference=DAI ACHANG A XSAVVA Met al. Scannet: Richly-annotated 3d reconstructions of indoor scenes[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: 2017: 5828-5839., articleTitle=Scannet: Richly-annotated 3d reconstructions of indoor scenes, refAbstract=null), Reference(id=1276896926301819421, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=50, issue=null, pageStart=104098, pageEnd=null, url=null, language=null, rfNumber=[66], rfOrder=65, authorNames=HAN Q H, LIU X, XU J, journalName=Journal of Building Engineering, refType=null, unstructuredReference=HAN Q HLIU XXU J. Detection and location of steel structure surface cracks based on unmanned aerial vehicle images[J]. Journal of Building Engineering202250: 104098., articleTitle=Detection and location of steel structure surface cracks based on unmanned aerial vehicle images, refAbstract=null), Reference(id=1276896926373122590, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=38, issue=7, pageStart=849, pageEnd=872, url=null, language=null, rfNumber=[67], rfOrder=66, authorNames=MENG S, GAO Z, ZHOU Y, journalName=Computer‐Aided Civil and Infrastructure Engineering, refType=null, unstructuredReference=MENG SGAO ZZHOU Yet al. Real‐time automatic crack detection method based on drone[J]. Computer‐Aided Civil and Infrastructure Engineering202338(7): 849-872., articleTitle=Real‐time automatic crack detection method based on drone, refAbstract=null), Reference(id=1276896926444425759, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[68], rfOrder=67, authorNames=IANDOLA F N, HAN S, MOSKEWICZ M W, journalName=V4. arXiv, refType=null, unstructuredReference=IANDOLA F NHAN SMOSKEWICZ M Wet al. SqueezeNet: alexNet-level accuracy with 50x fewer parameters and <0.5 MB model size[PP/OL]. V4. arXiv (2016-11-04) [2026-05-07]. https://doi.org/10.48550/arXiv.1602.07360., articleTitle=SqueezeNet: alexNet-level accuracy with 50x fewer parameters and <0.5 MB model size, refAbstract=null), Reference(id=1276896926532506144, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[69], rfOrder=68, authorNames=HOWARD A G, ZHU M, CHEN B, journalName=V1. ArXiv, refType=null, unstructuredReference=HOWARD A GZHU MCHEN Bet al. MobileNets: efficient convolutional neural networks for mobile vision applications[PP/OL]. V1. ArXiv (2017-04-17) [2026-05-07]. https://doi.org/10.48550/arXiv.1704.04861., articleTitle=MobileNets: efficient convolutional neural networks for mobile vision applications, refAbstract=null), Reference(id=1276896926599615009, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=6848, pageEnd=6856, url=null, language=null, rfNumber=[70], rfOrder=69, authorNames=ZHANG X, ZHOU X, LIN M, journalName=null, refType=null, unstructuredReference=ZHANG XZHOU XLIN Met al. ShuffleNet: an extremely efficient convolutional neural network for mobile devices[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Long Beach: 2018: 6848-6856., articleTitle=ShuffleNet: an extremely efficient convolutional neural network for mobile devices, refAbstract=null), Reference(id=1276896926670918178, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=6105, pageEnd=6114, url=null, language=null, rfNumber=[71], rfOrder=70, authorNames=TAN M, LE Q V, journalName=null, refType=null, unstructuredReference=TAN MLE Q V. EfficientNet: rethinking model scaling for convolutional neural networks[C]//Proceedings of the 36th International Conference on Machine Learning. 2019: 6105-6114., articleTitle=EfficientNet: rethinking model scaling for convolutional neural networks, refAbstract=null), Reference(id=1276896926763192867, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=10, issue=null, pageStart=133936, pageEnd=133944, url=null, language=null, rfNumber=[72], rfOrder=71, authorNames=WANG Y, WANG H, XIN Z, journalName=IEEE Access, refType=null, unstructuredReference=WANG YWANG HXIN Z. Efficient detection model of steel strip surface defects based on YOLO-V7[J]. IEEE Access202210: 133936-133944., articleTitle=Efficient detection model of steel strip surface defects based on YOLO-V7, refAbstract=null), Reference(id=1276896926838690340, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2025, volume=218, issue=null, pageStart=114099, pageEnd=null, url=null, language=null, rfNumber=[73], rfOrder=72, authorNames=LU N, WANG K, WANG H, journalName=Thin-Walled Structures, refType=null, unstructuredReference=LU NWANG KWANG Het al. Real-time fatigue crack detection and prediction in steel structures based on an automated digital twin-driven framework[J]. Thin-Walled Structures2025218:114099., articleTitle=Real-time fatigue crack detection and prediction in steel structures based on an automated digital twin-driven framework, refAbstract=null), Reference(id=1276896927014851109, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=54, issue=null, pageStart=101741, pageEnd=null, url=null, language=null, rfNumber=[74], rfOrder=73, authorNames=GUO Z, ZHANG Y, ZHU Q, journalName=Advanced Engineering Informatics, refType=null, unstructuredReference=GUO ZZHANG YZHU Q. CSCP-YOLO: a lightweight and efficient algorithm for real-time steel surface defect detection[J]. Advanced Engineering Informatics202254: 101741., articleTitle=CSCP-YOLO: a lightweight and efficient algorithm for real-time steel surface defect detection, refAbstract=null), Reference(id=1276896927081959974, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2026, volume=25, issue=2, pageStart=1165, pageEnd=1181, url=null, language=null, rfNumber=[75], rfOrder=74, authorNames=LIU R, ZENG W, journalName=Structural Health Monitoring, refType=null, unstructuredReference=LIU RZENG W. Automatic detection of structural defects in tunnel lining via network pruning and knowledge distillation in YOLO[J]. Structural Health Monitoring202625(2): 1165-1181., articleTitle=Automatic detection of structural defects in tunnel lining via network pruning and knowledge distillation in YOLO, refAbstract=null), Reference(id=1276896927165846055, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=158, issue=null, pageStart=105213, pageEnd=null, url=null, language=null, rfNumber=[76], rfOrder=75, authorNames=HUANG H, CAI Y, ZHANG C, journalName=Automation in Construction, refType=null, unstructuredReference=HUANG HCAI YZHANG Cet al. Crack detection of masonry structure based on thermal and visible image fusion and semantic segmentation[J]. Automation in Construction2024158: 105213., articleTitle=Crack detection of masonry structure based on thermal and visible image fusion and semantic segmentation, refAbstract=null), Reference(id=1276896927241343528, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=1, issue=1, pageStart=3, pageEnd=null, url=null, language=null, rfNumber=[77], rfOrder=76, authorNames=ALEXANDER Q G, HOSKERE V, NARAZAKI Y, journalName=AI in Civil Engineering, refType=null, unstructuredReference=ALEXANDER Q GHOSKERE VNARAZAKI Yet al. Fusion of thermal and RGB images for automated deep learning based crack detection in civil infrastructure[J]. AI in Civil Engineering20221(1): 3., articleTitle=Fusion of thermal and RGB images for automated deep learning based crack detection in civil infrastructure, refAbstract=null), Reference(id=1276896927316841001, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=95, issue=null, pageStart=110122, pageEnd=null, url=null, language=null, rfNumber=[78], rfOrder=77, authorNames=WANG P, XIAO J, QIANG X, journalName=Journal of Building Engineering, refType=null, unstructuredReference=WANG PXIAO JQIANG Xet al. An automatic building facade deterioration detection system using infrared-visible image fusion and deep learning[J]. Journal of Building Engineering202495: 110122., articleTitle=An automatic building facade deterioration detection system using infrared-visible image fusion and deep learning, refAbstract=null), Reference(id=1276896927396532778, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=20, issue=3, pageStart=1274, pageEnd=1293, url=null, language=null, rfNumber=[79], rfOrder=78, authorNames=ZHOU S, SONG W, journalName=Structural Health Monitoring, refType=null, unstructuredReference=ZHOU SSONG W. Deep learning-based roadway crack classification with heterogeneous image data fusion[J]. Structural Health Monitoring202120(3): 1274-1293., articleTitle=Deep learning-based roadway crack classification with heterogeneous image data fusion, refAbstract=null), Reference(id=1276896927493001771, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2020, volume=252, issue=null, pageStart=119096, pageEnd=null, url=null, language=null, rfNumber=[80], rfOrder=79, authorNames=PARK S E, EEM S H, JEON H, journalName=Construction and Building Materials, refType=null, unstructuredReference=PARK S EEEM S HJEON H. Concrete crack detection and quantification using deep learning and structured light[J]. Construction and Building Materials2020252: 119096., articleTitle=Concrete crack detection and quantification using deep learning and structured light, refAbstract=null), Reference(id=1276896927568499244, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=160, issue=null, pageStart=105318, pageEnd=null, url=null, language=null, rfNumber=[81], rfOrder=80, authorNames=WANG S, ZHAO X, GAO L, journalName=Automation in Construction, refType=null, unstructuredReference=WANG SZHAO XGAO Let al. Pixel-level crack segmentation and quantification enabled by multi-modality cross-fusion of RGB and depth images[J]. Automation in Construction2024160: 105318., articleTitle=Pixel-level crack segmentation and quantification enabled by multi-modality cross-fusion of RGB and depth images, refAbstract=null), Reference(id=1276896927639802413, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=350, issue=null, pageStart=128868, pageEnd=null, url=null, language=null, rfNumber=[82], rfOrder=81, authorNames=XU W, CUI C, LUO C, journalName=Construction and Building Materials, refType=null, unstructuredReference=XU WCUI CLUO Cet al. Fatigue crack monitoring of steel bridge with coating sensor based on potential difference method[J]. Construction and Building Materials2022350: 128868., articleTitle=Fatigue crack monitoring of steel bridge with coating sensor based on potential difference method, refAbstract=null), Reference(id=1276896927702716974, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2025, volume=73, issue=1, pageStart=189, pageEnd=198, url=null, language=null, rfNumber=[83], rfOrder=82, authorNames=MANU K C, MADHUSHREE C, CHANDINI M S, journalName=Journal of Mines, Metals and Fuels, refType=null, unstructuredReference=MANU K CMADHUSHREE CCHANDINI M Set al. Corrosion in steel structures: a review[J]. Journal of Mines, Metals and Fuels202573(1): 189-198., articleTitle=Corrosion in steel structures: a review, refAbstract=null), Reference(id=1276896927832740399, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2020, volume=89, issue=null, pageStart=115978, pageEnd=null, url=null, language=null, rfNumber=[84], rfOrder=83, authorNames=ANWAR S, LI C, journalName=Signal Processing: Image Communication, refType=null, unstructuredReference=ANWAR SLI C. Diving deeper into underwater image enhancement: a survey[J]. Signal Processing: Image Communication202089: 115978., articleTitle=Diving deeper into underwater image enhancement: a survey, refAbstract=null), Reference(id=1276896927908237872, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=262, pageEnd=266, url=null, language=null, rfNumber=[85], rfOrder=84, authorNames=DUY L D, ANH N T, SON N T, journalName=null, refType=null, unstructuredReference=DUY L DANH N TSON N Tet al. Deep learning in semantic segmentation of rust in images[C]//Proceedings of the 2020 9th International Conference on Software and Computer Applications. Malaysia: 2020: 262-266., articleTitle=Deep learning in semantic segmentation of rust in images, refAbstract=null), Reference(id=1276896927983735345, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=11, pageStart=3630, pageEnd=null, url=null, language=null, rfNumber=[86], rfOrder=85, authorNames=DAS A, DORAFSHAN S, KAABOUCH N, journalName=Sensors, refType=null, unstructuredReference=DAS A, DORAFSHAN SKAABOUCH N. Autonomous image-based corrosion detection in steel structures using deep learning[J]. Sensors202424(11): 3630., articleTitle=Autonomous image-based corrosion detection in steel structures using deep learning, refAbstract=null), Reference(id=1276896928059232818, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2002, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[87], rfOrder=86, authorNames=中国国家标准化管理委员会, journalName=null, refType=null, unstructuredReference=中国国家标准化管理委员会. 金属基体上金属和其他无机覆盖层 经腐蚀试验后的试样和试件的评级:GB/T 6461—2002[S]. 北京: 中国标准出版社, 2002., articleTitle=null, refAbstract=null), Reference(id=1276896928168284723, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[88], rfOrder=87, authorNames=ASTM International, journalName=null, refType=null, unstructuredReference=ASTM International. Standard test method for evaluating degree of rusting on painted steel surfaces:ASTM D610-23[S]. West Conshohocken, PA: ASTM International, 2023., articleTitle=null, refAbstract=null), Reference(id=1276896928277336628, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[89], rfOrder=88, authorNames=International Organization for Standardization, journalName=null, refType=null, unstructuredReference=International Organization for Standardization. Paints and varnishes-evaluation of degradation of coatings-part 3: assessment of degree of rusting:ISO 4628-3∶2016 [S]. Geneva, Switzerland: ISO, 2016., articleTitle=null, refAbstract=null), Reference(id=1276896928377999925, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=56, issue=7, pageStart=713, pageEnd=722, url=null, language=null, rfNumber=[90], rfOrder=89, authorNames=陆廷杰, 刘东海, 齐志龙, journalName=天津大学学报(自然科学与工程技术版), refType=null, unstructuredReference=陆廷杰, 刘东海, 齐志龙. 基于深度学习的水下钢结构锈蚀识别与评价[J]. 天津大学学报(自然科学与工程技术版)202356(7): 713-722., articleTitle=基于深度学习的水下钢结构锈蚀识别与评价, refAbstract=null), Reference(id=1276896930072498742, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=91, pageEnd=99, url=null, language=null, rfNumber=[91], rfOrder=90, authorNames=PETRICCA L, MOSS T, FIGUEROA G, journalName=null, refType=null, unstructuredReference=PETRICCA LMOSS TFIGUEROA Get al. Corrosion detection using AI: a comparison of standard computer vision techniques and deep learning model[C]//Proceedings of the Sixth International Conference on Computer Science, Engineering and Information Technology. Chennai: 2016: 91-99., articleTitle=Corrosion detection using AI: a comparison of standard computer vision techniques and deep learning model, refAbstract=null), Reference(id=1276896930185744951, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2018, volume=17, issue=5, pageStart=1110, pageEnd=1128, url=null, language=null, rfNumber=[92], rfOrder=91, authorNames=ATHA D J, JAHANSHAHI M R, journalName=Structural Health Monitoring, refType=null, unstructuredReference=ATHA D JJAHANSHAHI M R. Evaluation of deep learning approaches based on convolutional neural networks for corrosion detection[J]. Structural Health Monitoring201817(5): 1110-1128., articleTitle=Evaluation of deep learning approaches based on convolutional neural networks for corrosion detection, refAbstract=null), Reference(id=1276896930269631032, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=549, pageEnd=556, url=null, language=null, rfNumber=[93], rfOrder=92, authorNames=HOLM E, TRANSETH A A, KNUDSEN O Ø, journalName=null, refType=null, unstructuredReference=HOLM ETRANSETH A AKNUDSEN O Øet al. Classification of corrosion and coating damages on bridge constructions from images using convolutional neural networks[C]//Twelfth International Conference on Machine Vision (ICMV 2019). Bellingham: 2020: 549-556., articleTitle=Classification of corrosion and coating damages on bridge constructions from images using convolutional neural networks, refAbstract=null), Reference(id=1276896930349322809, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2019, volume=107, issue=null, pageStart=102134, pageEnd=null, url=null, language=null, rfNumber=[94], rfOrder=93, authorNames=BASTIAN B T, JASPREETH N, RANJITH S K, journalName=NDT & E International, refType=null, unstructuredReference=BASTIAN B TJASPREETH NRANJITH S Ket al. Visual inspection and characterization of external corrosion in pipelines using deep neural network[J]. NDT & E International2019107: 102134., articleTitle=Visual inspection and characterization of external corrosion in pipelines using deep neural network, refAbstract=null), Reference(id=1276896930458374714, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=193, issue=null, pageStart=116461, pageEnd=null, url=null, language=null, rfNumber=[95], rfOrder=94, authorNames=FORKAN A R M, KANG Y B, JAYARAMAN P P, journalName=Expert Systems with Applications, refType=null, unstructuredReference=FORKAN A R MKANG Y BJAYARAMAN P Pet al. CorrDetector: a framework for structural corrosion detection from drone images using ensemble deep learning[J]. Expert Systems with Applications2022193: 116461., articleTitle=CorrDetector: a framework for structural corrosion detection from drone images using ensemble deep learning, refAbstract=null), Reference(id=1276896930529677883, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2020, volume=35, issue=10, pageStart=1160, pageEnd=1174, url=null, language=null, rfNumber=[96], rfOrder=95, authorNames=XU J, GUI C, HAN Q, journalName=Computer-Aided Civil and Infrastructure Engineering, refType=null, unstructuredReference=XU JGUI CHAN Q. Recognition of rust grade and rust ratio of steel structures based on ensembled convolutional neural network[J]. Computer-Aided Civil and Infrastructure Engineering202035(10): 1160-1174., articleTitle=Recognition of rust grade and rust ratio of steel structures based on ensembled convolutional neural network, refAbstract=null), Reference(id=1276896930613563964, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=20, issue=6, pageStart=3424, pageEnd=3435, url=null, language=null, rfNumber=[97], rfOrder=96, authorNames=JIN LIM H, HWANG S, KIM H, journalName=Structural Health Monitoring, refType=null, unstructuredReference=JIN LIM HHWANG SKIM Het al. Steel bridge corrosion inspection with combined vision and thermographic images[J]. Structural Health Monitoring202120(6): 3424-3435., articleTitle=Steel bridge corrosion inspection with combined vision and thermographic images, refAbstract=null), Reference(id=1276896930714227261, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=197, pageEnd=204, url=null, language=null, rfNumber=[98], rfOrder=97, authorNames=ANDERSEN R, NALPANTIDIS L, RAVN O, journalName=null, refType=null, unstructuredReference=ANDERSEN RNALPANTIDIS LRAVN Oet al. Investigating deep learning architectures towards autonomous inspection for marine classification[C]//2020 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR). Abu Dhabi: 2020: 197-204., articleTitle=Investigating deep learning architectures towards autonomous inspection for marine classification, refAbstract=null), Reference(id=1276896930785530430, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=16, issue=6, pageStart=1701, pageEnd=1709, url=null, language=null, rfNumber=[99], rfOrder=98, authorNames=ZHOU Q, DING S, FENG Y, journalName=Signal, Image and Video Processing, refType=null, unstructuredReference=ZHOU QDING SFENG Yet al. Corrosion inspection and evaluation of crane metal structure based on UAV vision[J]. Signal, Image and Video Processing202216(6): 1701-1709., articleTitle=Corrosion inspection and evaluation of crane metal structure based on UAV vision, refAbstract=null), Reference(id=1276896930865222207, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=76, issue=null, pageStart=102367, pageEnd=null, url=null, language=null, rfNumber=[100], rfOrder=99, authorNames=JIA Z, FU M, ZHAO X, journalName=Displays, refType=null, unstructuredReference=JIA ZFU MZHAO Xet al. Intelligent identification of metal corrosion based on Corrosion-YOLOv5s[J]. Displays202376: 102367., articleTitle=Intelligent identification of metal corrosion based on Corrosion-YOLOv5s, refAbstract=null), Reference(id=1276896930932331072, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=24, issue=8, pageStart=2911, pageEnd=2923, url=null, language=null, rfNumber=[101], rfOrder=100, authorNames=NABIZADEH E, PARGHI A, journalName=Asian Journal of Civil Engineering, refType=null, unstructuredReference=NABIZADEH EPARGHI A. Automated corrosion detection using deep learning and computer vision[J]. Asian Journal of Civil Engineering202324(8): 2911-2923., articleTitle=Automated corrosion detection using deep learning and computer vision, refAbstract=null), Reference(id=1276896931016217153, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=9, issue=1, pageStart=1, pageEnd=16, url=null, language=null, rfNumber=[102], rfOrder=101, authorNames=AMELI Z, NESHELI S J, LANDIS E N, journalName=Infrastructures, refType=null, unstructuredReference=AMELI ZNESHELI S JLANDIS E N. Deep learning-based steel bridge corrosion segmentation and condition rating using mask RCNN and YOLOv8[J]. Infrastructures20239(1): 1-16., articleTitle=Deep learning-based steel bridge corrosion segmentation and condition rating using mask RCNN and YOLOv8, refAbstract=null), Reference(id=1276896931083326018, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=12, issue=7, pageStart=1090, pageEnd=null, url=null, language=null, rfNumber=[103], rfOrder=102, authorNames=YU Q F, HAN Y D, LIN W G, journalName=Journal of Marine Science and Engineering, refType=null, unstructuredReference=YU Q FHAN Y DLIN W Get al. Detection and analysis of corrosion on coated metal surfaces using enhanced YOLOv5 algorithm for anti-corrosion performance evaluation[J]. Journal of Marine Science and Engineering202412(7): 1090., articleTitle=Detection and analysis of corrosion on coated metal surfaces using enhanced YOLOv5 algorithm for anti-corrosion performance evaluation, refAbstract=null), Reference(id=1276896931150434883, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=220, pageEnd=231, url=null, language=null, rfNumber=[104], rfOrder=103, authorNames=PIRIE C, MORENO-GARCIA C F, journalName=null, refType=null, unstructuredReference=PIRIE CMORENO-GARCIA C F. Image pre-processing and segmentation for real-time subsea corrosion inspection[C]//International Conference on Engineering Applications of Neural Networks. Cham: 2021: 220-231., articleTitle=Image pre-processing and segmentation for real-time subsea corrosion inspection, refAbstract=null), Reference(id=1276896931234320964, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2018, volume=46, issue=12, pageStart=121, pageEnd=127, url=null, language=null, rfNumber=[105], rfOrder=104, authorNames=王达磊, 彭博, 潘玥, journalName=华南理工大学学报(自然科学版), refType=null, unstructuredReference=王达磊, 彭博, 潘玥, . 基于深度神经网络的锈蚀图像分割与定量分析[J]. 华南理工大学学报(自然科学版)201846(12): 121-127., articleTitle=基于深度神经网络的锈蚀图像分割与定量分析, refAbstract=null), Reference(id=1276896931297235525, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=787, pageEnd=792, url=null, language=null, rfNumber=[106], rfOrder=105, authorNames=FONDEVIK S K, STAHL A, TRANSETH A A, journalName=null, refType=null, unstructuredReference=FONDEVIK S KSTAHL ATRANSETH A Aet al. Image segmentation of corrosion damages in industrial inspections[C]//2020 IEEE 32nd International Conference on Tools with Artificial Intelligence (ICTAI). Baltimore: 2020: 787-792., articleTitle=Image segmentation of corrosion damages in industrial inspections, refAbstract=null), Reference(id=1276896931372732998, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=160, pageEnd=169, url=null, language=null, rfNumber=[107], rfOrder=106, authorNames=KATSAMENIS I, PROTOPAPADAKIS E, DOULAMIS A, journalName=null, refType=null, unstructuredReference=KATSAMENIS IPROTOPAPADAKIS EDOULAMIS Aet al. Pixel-level corrosion detection on metal constructions by fusion of deep learning semantic and contour segmentation[C]//International Symposium on Visual Computing. Cham: 2020: 160-169., articleTitle=Pixel-level corrosion detection on metal constructions by fusion of deep learning semantic and contour segmentation, refAbstract=null), Reference(id=1276896931439841863, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=174, issue=null, pageStart=320, pageEnd=327, url=null, language=null, rfNumber=[108], rfOrder=107, authorNames=AKHLAGHI B, MESGHALI H, EHTESHAMI M, journalName=Process Safety and Environmental Protection, refType=null, unstructuredReference=AKHLAGHI BMESGHALI HEHTESHAMI Met al. Predictive deep learning for pitting corrosion modeling in buried transmission pipelines[J]. Process Safety and Environmental Protection2023174: 320-327., articleTitle=Predictive deep learning for pitting corrosion modeling in buried transmission pipelines, refAbstract=null), Reference(id=1276896931511145032, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=14, issue=12, pageStart=1552, pageEnd=null, url=null, language=null, rfNumber=[109], rfOrder=108, authorNames=GAO R P, SHANG W J, ZHAO Y, journalName=Coatings, refType=null, unstructuredReference=GAO R PSHANG W JZHAO Yet al. Research on fusion model method for corrosion damage detection of switch sliding baseplate[J]. Coatings202414(12): 1552., articleTitle=Research on fusion model method for corrosion damage detection of switch sliding baseplate, refAbstract=null), Reference(id=1276896931590836809, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2025, volume=253, issue=null, pageStart=110507, pageEnd=null, url=null, language=null, rfNumber=[110], rfOrder=109, authorNames=ZHANG Z W, LI S L, WANG H J, journalName=Reliability Engineering & System Safety, refType=null, unstructuredReference=ZHANG Z WLI S LWANG H Jet al. A study of neural network-based evaluation methods for pipelines with multiple corrosive regions[J]. Reliability Engineering & System Safety2025253: 110507., articleTitle=A study of neural network-based evaluation methods for pipelines with multiple corrosive regions, refAbstract=null), Reference(id=1276896931662139978, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[111], rfOrder=110, authorNames=秦荣杰, journalName=null, refType=null, unstructuredReference=秦荣杰.基于深度学习Transformer网络的钢板锈蚀类别识别方法研究[D].西安:西安建筑科技大学,2023., articleTitle=基于深度学习Transformer网络的钢板锈蚀类别识别方法研究, refAbstract=null), Reference(id=1276896931737637451, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=11, issue=13, pageStart=6063, pageEnd=null, url=null, language=null, rfNumber=[112], rfOrder=111, authorNames=ALTABEY W A, NOORI M, WANG T, journalName=Applied Sciences, refType=null, unstructuredReference=ALTABEY W ANOORI MWANG Tet al. Deep learning-based crack identification for steel pipelines by extracting features from 3D shadow modeling[J]. Applied Sciences202111(13): 6063., articleTitle=Deep learning-based crack identification for steel pipelines by extracting features from 3D shadow modeling, refAbstract=null), Reference(id=1276896931821523532, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=237, issue=null, pageStart=112334, pageEnd=null, url=null, language=null, rfNumber=[113], rfOrder=112, authorNames=WANG B Q, LIU L A, CHENG X Q, journalName=Corrosion Science, refType=null, unstructuredReference=WANG B QLIU L ACHENG X Qet al. Advanced multi-image segmentation-based machine learning modeling strategy for corrosion prediction and rust layer performance evaluation of weathering steel[J]. Corrosion Science2024237: 112334., articleTitle=Advanced multi-image segmentation-based machine learning modeling strategy for corrosion prediction and rust layer performance evaluation of weathering steel, refAbstract=null), Reference(id=1276896931892826701, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=11, issue=null, pageStart=1277710, pageEnd=null, url=null, language=null, rfNumber=[114], rfOrder=113, authorNames=LI Y P, LI H G, GUAN Y, journalName=Frontiers in Physics, refType=null, unstructuredReference=LI Y PLI H GGUAN Yet al. Dense metal corrosion depth estimation[J]. Frontiers in Physics202311: 1277710., articleTitle=Dense metal corrosion depth estimation, refAbstract=null), Reference(id=1276896931959935566, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=16, issue=null, pageStart=100617, pageEnd=null, url=null, language=null, rfNumber=[115], rfOrder=114, authorNames=SON E Y, JEONG D, OH M J, journalName=International Journal of Naval Architecture and Ocean Engineering, refType=null, unstructuredReference=SON E YJEONG DOH M J. Corrosion area detection and depth prediction using machine learning[J]. International Journal of Naval Architecture and Ocean Engineering202416: 100617., articleTitle=Corrosion area detection and depth prediction using machine learning, refAbstract=null), Reference(id=1276896932052210255, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2025, volume=15, issue=6, pageStart=645, pageEnd=null, url=null, language=null, rfNumber=[116], rfOrder=115, authorNames=ARIAS F, GUEVARA E, JARAMILLO E, journalName=Coatings, refType=null, unstructuredReference=ARIAS FGUEVARA EJARAMILLO Eet al. Automated assessment of marine steel corrosion using visible-near-infrared hyperspectral imaging[J]. Coatings202515(6):645., articleTitle=Automated assessment of marine steel corrosion using visible-near-infrared hyperspectral imaging, refAbstract=null), Reference(id=1276896932123513424, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2025, volume=15, issue=1, pageStart=23894, pageEnd=null, url=null, language=null, rfNumber=[117], rfOrder=116, authorNames=EGODAWELA S, GOSTAR A K, BUDDIKA H A D S, journalName=Scientific Reports, refType=null, unstructuredReference=EGODAWELA SGOSTAR A KBUDDIKA H A D Set al. Metal loss defect detection and depth estimation using multi-spectral image analysis of cooling excited steel specimen with corrosion[J]. Scientific Reports202515(1): 23894., articleTitle=Metal loss defect detection and depth estimation using multi-spectral image analysis of cooling excited steel specimen with corrosion, refAbstract=null), Reference(id=1276896932257731153, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=606, pageEnd=610, url=null, language=null, rfNumber=[118], rfOrder=117, authorNames=LIU L, TAN E, ZHEN Y, journalName=null, refType=null, unstructuredReference=LIU LTAN EZHEN Yet al. AI-facilitated coating corrosion assessment system for productivity enhancement[C]//2018 13th IEEE Conference on Industrial Electronics and Applications (ICIEA). Wuhan: 2018: 606-610., articleTitle=AI-facilitated coating corrosion assessment system for productivity enhancement, refAbstract=null), Reference(id=1276896932345811538, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2019, volume=90, issue=null, pageStart=101823, pageEnd=null, url=null, language=null, rfNumber=[119], rfOrder=118, authorNames=YAO Y, YANG Y, WANG Y, journalName=Applied Ocean Research, refType=null, unstructuredReference=YAO YYANG YWANG Yet al. Artificial intelligence-based hull structural plate corrosion damage detection and recognition using convolutional neural network[J]. Applied Ocean Research201990: 101823., articleTitle=Artificial intelligence-based hull structural plate corrosion damage detection and recognition using convolutional neural network, refAbstract=null), Reference(id=1276896932425503315, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=36, issue=12, pageStart=1585, pageEnd=1599, url=null, language=null, rfNumber=[120], rfOrder=119, authorNames=LUO C, YU L, YAN J, journalName=Computer‐Aided Civil and Infrastructure Engineering, refType=null, unstructuredReference=LUO CYU LYAN Jet al. Autonomous detection of damage to multiple steel surfaces from 360 panoramas using deep neural networks[J]. Computer‐Aided Civil and Infrastructure Engineering202136(12): 1585-1599., articleTitle=Autonomous detection of damage to multiple steel surfaces from 360 panoramas using deep neural networks, refAbstract=null), Reference(id=1276896932492612180, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=14, issue=null, pageStart=8087, pageEnd=8098, url=null, language=null, rfNumber=[121], rfOrder=120, authorNames=YU L J, YANG E F, LUO C, journalName=Journal of Ambient Intelligence and Humanized Computing, refType=null, unstructuredReference=YU L JYANG E FLUO Cet al. AMCD: an accurate deep learning-based metallic corrosion detector for MAV-based real-time visual inspection[J]. Journal of Ambient Intelligence and Humanized Computing202314:8087-8098., articleTitle=AMCD: an accurate deep learning-based metallic corrosion detector for MAV-based real-time visual inspection, refAbstract=null), Reference(id=1276896932559721045, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=null, pageStart=1674, pageEnd=1699, url=null, language=null, rfNumber=[122], rfOrder=121, authorNames=HUANG M, ZHANG J, LI J, journalName=Structural Health Monitoring, refType=null, unstructuredReference=HUANG MZHANG JLI Jet al. Damage identification of steel bridge based on data augmentation and adaptive optimization neural network[J]. Structural Health Monitoring202424: 1674-1699., articleTitle=Damage identification of steel bridge based on data augmentation and adaptive optimization neural network, refAbstract=null), Reference(id=1276896932672967254, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=139, issue=null, pageStart=104299, pageEnd=null, url=null, language=null, rfNumber=[123], rfOrder=122, authorNames=BIANCHI E, HEBDON M, journalName=Automation in Construction, refType=null, unstructuredReference=BIANCHI EHEBDON M. Visual structural inspection datasets[J]. Automation in Construction2022139: 104299., articleTitle=Visual structural inspection datasets, refAbstract=null), Reference(id=1276896932744270423, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2020, volume=14, issue=1, pageStart=1, pageEnd=17, url=null, language=null, rfNumber=[124], rfOrder=123, authorNames=梁俊杰, 韦舰晶, 蒋正锋, journalName=计算机科学与探索, refType=null, unstructuredReference=梁俊杰, 韦舰晶, 蒋正锋. 生成对抗网络GAN综述[J]. 计算机科学与探索202014(1): 1-17., articleTitle=生成对抗网络GAN综述, refAbstract=null), Reference(id=1276896932844933720, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[125], rfOrder=124, authorNames=RADFORD A, METZ L, CHINTALA S, journalName=V2. arXiv, refType=null, unstructuredReference=RADFORD AMETZ LCHINTALA S. Unsupervised representation learning with deep convolutional generative adversarial networks[PP/OL]. V2. arXiv (2016-01-07) [2026-05-07] .https://doi.org/10.48550/arXiv.1511.06434., articleTitle=Unsupervised representation learning with deep convolutional generative adversarial networks, refAbstract=null), Reference(id=1276896932916236889, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=5769, pageEnd=5779, url=null, language=null, rfNumber=[126], rfOrder=125, authorNames=GULRAJANI I, AHMED F, ARJOVSKY M, journalName=null, refType=null, unstructuredReference=GULRAJANI IAHMED FARJOVSKY Met al. Improved training of Wasserstein GANs[C]// Neural Information Processing Systems. Long Beach: 2017: 5769-5779., articleTitle=Improved training of Wasserstein GANs, refAbstract=null), Reference(id=1276896934614930011, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=20, issue=4, pageStart=2069, pageEnd=2087, url=null, language=null, rfNumber=[127], rfOrder=126, authorNames=LEI X, SUN L, XIA Y, journalName=Structural Health Monitoring, refType=null, unstructuredReference=LEI XSUN LXIA Y. Lost data reconstruction for structural health monitoring using deep convolutional generative adversarial networks[J]. Structural Health Monitoring202220(4): 2069-2087., articleTitle=Lost data reconstruction for structural health monitoring using deep convolutional generative adversarial networks, refAbstract=null), Reference(id=1276896934677844572, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2019, volume=34, issue=9, pageStart=755, pageEnd=773, url=null, language=null, rfNumber=[128], rfOrder=127, authorNames=GAO Y, KONG B, MOSALAM K M, journalName=Computer-Aided Civil and Infrastructure Engineering, refType=null, unstructuredReference=GAO YKONG BMOSALAM K M. Deep leaf-bootstrapping generative adversarial network for structural image data augmentation[J]. Computer-Aided Civil and Infrastructure Engineering201934(9): 755-773., articleTitle=Deep leaf-bootstrapping generative adversarial network for structural image data augmentation, refAbstract=null), Reference(id=1276896934786896477, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=36, issue=1, pageStart=47, pageEnd=60, url=null, language=null, rfNumber=[129], rfOrder=128, authorNames=MAEDA H, KASHIYAMA T, SEKIMOTO Y, journalName=Computer-Aided Civil and Infrastructure Engineering, refType=null, unstructuredReference=MAEDA HKASHIYAMA TSEKIMOTO Yet al. Generative adversarial network for road damage detection[J]. Computer-Aided Civil and Infrastructure Engineering202136(1): 47-60., articleTitle=Generative adversarial network for road damage detection, refAbstract=null), Reference(id=1276896934858199646, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=2890, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[130], rfOrder=129, authorNames=XU Y Z, WU H, LIU Y L, journalName=null, refType=null, unstructuredReference=XU Y ZWU HLIU Y Let al. Automated surface defect detection based on CycleGAN model[C]//Journal of Physics: Conference Series. Bristol: 20242890: 012036., articleTitle=Automated surface defect detection based on CycleGAN model, refAbstract=null), Reference(id=1276896934933697119, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2025, volume=15, issue=1, pageStart=12994, pageEnd=null, url=null, language=null, rfNumber=[131], rfOrder=130, authorNames=WANG Y, LIAO X, CUI W, journalName=Scientific Reports, refType=null, unstructuredReference=WANG YLIAO XCUI Wet al. Defending against and generating adversarial examples together with generative adversarial networks[J]. Scientific Reports202515(1): 12994., articleTitle=Defending against and generating adversarial examples together with generative adversarial networks, refAbstract=null), Reference(id=1276896934996611680, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=229, issue=null, pageStart=103649, pageEnd=null, url=null, language=null, rfNumber=[132], rfOrder=131, authorNames=ZHOU H Y, JIANG F, LU H T, journalName=Computer Vision and Image Understanding, refType=null, unstructuredReference=ZHOU H YJIANG FLU H Tet al. SSDA-YOLO: semi-supervised domain adaptive YOLO for cross-domain object detection[J]. Computer Vision and Image Understanding2023229: 103649., articleTitle=SSDA-YOLO: semi-supervised domain adaptive YOLO for cross-domain object detection, refAbstract=null), Reference(id=1276896935101469281, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[133], rfOrder=132, authorNames=牟宗涵, journalName=null, refType=null, unstructuredReference=牟宗涵. 基于无人机图像的铁路桥梁钢结构表面缺陷智能识别方法研究[D]. 北京:北京交通大学, 2023., articleTitle=基于无人机图像的铁路桥梁钢结构表面缺陷智能识别方法研究, refAbstract=null), Reference(id=1276896935168578146, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=44, issue=5, pageStart=34, pageEnd=40, url=null, language=null, rfNumber=[134], rfOrder=133, authorNames=钱企豪, 郑战光, 梁钊, journalName=腐蚀与防护, refType=null, unstructuredReference=钱企豪, 郑战光, 梁钊, . 基于颜色特征的半监督聚类算法在铜片腐蚀等级识别中的应用[J]. 腐蚀与防护202344(5): 34-40., articleTitle=基于颜色特征的半监督聚类算法在铜片腐蚀等级识别中的应用, refAbstract=null), Reference(id=1276896935256658531, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=12, issue=15, pageStart=2348, pageEnd=null, url=null, language=null, rfNumber=[135], rfOrder=134, authorNames=FENG J J, TIAN L F, LI X X, journalName=Mathematics, refType=null, unstructuredReference=FENG J JTIAN L FLI X Xet al. Adaptive adversarial self-training for semi-supervised object detection in complex maritime scenes[J]. Mathematics202412(15): 2348., articleTitle=Adaptive adversarial self-training for semi-supervised object detection in complex maritime scenes, refAbstract=null), Reference(id=1276896935340544612, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2025, volume=24, issue=null, pageStart=2229, pageEnd=2249, url=null, language=null, rfNumber=[136], rfOrder=135, authorNames=WANG S Y, NGUYEN H D, WILSON R, journalName=Structural Health Monitoring, refType=null, unstructuredReference=WANG S YNGUYEN H DWILSON Ret al. Deep CNN-based semi-supervised learning approach for identifying and segmenting corrosion in hydraulic steel and water resources infrastructure[J]. Structural Health Monitoring202524: 2229-2249., articleTitle=Deep CNN-based semi-supervised learning approach for identifying and segmenting corrosion in hydraulic steel and water resources infrastructure, refAbstract=null), Reference(id=1276896935407653477, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=8, issue=3, pageStart=331, pageEnd=368, url=null, language=null, rfNumber=[137], rfOrder=136, authorNames=GUO M H, XU T X, LIU J J, journalName=Computational Visual Media, refType=null, unstructuredReference=GUO M HXU T XLIU J Jet al. Attention mechanisms in computer vision: a survey[J]. Computational Visual Media20228(3): 331-368., articleTitle=Attention mechanisms in computer vision: a survey, refAbstract=null), Reference(id=1276896935483150950, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=14, issue=12, pageStart=3972, pageEnd=null, url=null, language=null, rfNumber=[138], rfOrder=137, authorNames=DUAN Z, HUANG X H, HOU J, journalName=Buildings, refType=null, unstructuredReference=DUAN ZHUANG X HHOU Jet al. Research on intelligent diagnosis of corrosion in the operation and maintenance stage of steel structure engineering based on U-Net attention[J]. Buildings202414(12): 3972., articleTitle=Research on intelligent diagnosis of corrosion in the operation and maintenance stage of steel structure engineering based on U-Net attention, refAbstract=null), Reference(id=1276896935550259815, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=137, issue=null, pageStart=104182, pageEnd=null, url=null, language=null, rfNumber=[139], rfOrder=138, authorNames=KATSAMENIS I, DOULAMIS N, DOULAMIS A, journalName=Automation in Construction, refType=null, unstructuredReference=KATSAMENIS IDOULAMIS NDOULAMIS Aet al. Simultaneous precise localization and classification of metal rust defects for robotic-driven maintenance and prefabrication using residual attention U-Net[J]. Automation in Construction2022137: 104182., articleTitle=Simultaneous precise localization and classification of metal rust defects for robotic-driven maintenance and prefabrication using residual attention U-Net, refAbstract=null), Reference(id=1276896935617368680, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2025, volume=15, issue=null, pageStart=93469, pageEnd=null, url=null, language=null, rfNumber=[140], rfOrder=139, authorNames=MA S B, ZHAO X, WAN L, journalName=Scientific Reports, refType=null, unstructuredReference=MA S BZHAO XWAN Let al. A lightweight algorithm for steel surface defect detection using improved YOLOv8[J]. Scientific Reports202515: 93469., articleTitle=A lightweight algorithm for steel surface defect detection using improved YOLOv8, refAbstract=null), Reference(id=1276896935692866153, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=19, issue=4, pageStart=1932, pageEnd=6203, url=null, language=null, rfNumber=[141], rfOrder=140, authorNames=FU M J, JIA Z T, WU L Z, journalName=PloS One, refType=null, unstructuredReference=FU M JJIA Z TWU L Zet al. Detection and recognition of metal surface corrosion based on CBG-YOLOv5s[J]. PloS One, San Francisco, 202419(4): 1932-6203., articleTitle=Detection and recognition of metal surface corrosion based on CBG-YOLOv5s, refAbstract=null), Reference(id=1276896935755780714, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=16, issue=4, pageStart=1, pageEnd=12, url=null, language=null, rfNumber=[142], rfOrder=141, authorNames=ZHANG G H, LIU S X, NIE S Q, journalName=Symmetry, refType=null, unstructuredReference=ZHANG G HLIU S XNIE S Qet al. YOLO-RDP: lightweight steel defect detection through improved YOLOv7-tiny and model pruning[J]. Symmetry202416(4): 1-12., articleTitle=YOLO-RDP: lightweight steel defect detection through improved YOLOv7-tiny and model pruning, refAbstract=null), Reference(id=1276896935827083883, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=1, pageStart=1, pageEnd=16, url=null, language=null, rfNumber=[143], rfOrder=142, authorNames=TAN L, CHEN X H, YUAN D J, journalName=Structural Control and Health Monitoring, refType=null, unstructuredReference=TAN LCHEN X HYUAN D Jet al. DSNet: a Computer vision-based detection and corrosion segmentation network for corroded bolt detection in tunnel[J]. Structural Control and Health Monitoring2024, 2024(1): 1-16., articleTitle=DSNet: a Computer vision-based detection and corrosion segmentation network for corroded bolt detection in tunnel, refAbstract=null), Reference(id=1276896935889998444, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2025, volume=13, issue=null, pageStart=71673, pageEnd=71687, url=null, language=null, rfNumber=[144], rfOrder=143, authorNames=YU V F, SANTIYUDA G, LIN S W, journalName=IEEE Access, refType=null, unstructuredReference=YU V FSANTIYUDA GLIN S Wet al. Neural network pruning for lightweight metal corrosion image segmentation models[J]. IEEE Access202513: 71673-71687., articleTitle=Neural network pruning for lightweight metal corrosion image segmentation models, refAbstract=null), Reference(id=1276896935957107309, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=11, issue=5, pageStart=1375, pageEnd=1392, url=null, language=null, rfNumber=[145], rfOrder=144, authorNames=HAN Q H, ZHAO N, XU J, journalName=Journal of Civil Structural Health Monitoring, refType=null, unstructuredReference=HAN Q HZHAO NXU J. Recognition and location of steel structure surface corrosion based on unmanned aerial vehicle images[J]. Journal of Civil Structural Health Monitoring202111(5): 1375-1392., articleTitle=Recognition and location of steel structure surface corrosion based on unmanned aerial vehicle images, refAbstract=null), Reference(id=1276896936024216174, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=18, pageStart=6007, pageEnd=null, url=null, language=null, rfNumber=[146], rfOrder=145, authorNames=ELTOUNY K, SAJEDI S, LIANG X, journalName=Sensors, refType=null, unstructuredReference=ELTOUNY KSAJEDI SLIANG X. Dmg2Former-AR: vision transformers with adaptive rescaling for high-resolution structural visual inspection[J]. Sensors202424(18): 6007., articleTitle=Dmg2Former-AR: vision transformers with adaptive rescaling for high-resolution structural visual inspection, refAbstract=null), Reference(id=1276896936103907951, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=147, issue=11, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[147], rfOrder=146, authorNames=YE X W, JIN T, LI Z X, journalName=Journal of Structural Engineering, refType=null, unstructuredReference=YE X WJIN TLI Z Xet al. Structural crack detection from benchmark data sets using pruned fully convolutional networks[J]. Journal of Structural Engineering2021147(11): 04721008., articleTitle=Structural crack detection from benchmark data sets using pruned fully convolutional networks, refAbstract=null), Reference(id=1276896936179405424, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[148], rfOrder=147, authorNames=BIANCHI E, HEBDON M, journalName=Blacksburg, refType=null, unstructuredReference=BIANCHI EHEBDON M. Corrosion condition state semantic segmentation dataset[DS]. Blacksburg, VA, USA: Virginia Tech, 2021., articleTitle=Corrosion condition state semantic segmentation dataset, refAbstract=null), Reference(id=1276896936263291505, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2020, volume=10, issue=5, pageStart=757, pageEnd=773, url=null, language=null, rfNumber=[149], rfOrder=148, authorNames=HOSKERE V, NARAZAKI Y, HOANG T A, journalName=Journal of Civil Structural Health Monitoring, refType=null, unstructuredReference=HOSKERE VNARAZAKI YHOANG T Aet al. MaDnet: multi-task semantic segmentation of multiple types of structural materials and damage in images of civil infrastructure[J]. Journal of Civil Structural Health Monitoring202010(5): 757-773., articleTitle=MaDnet: multi-task semantic segmentation of multiple types of structural materials and damage in images of civil infrastructure, refAbstract=null), Reference(id=1276896936334594674, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2020, volume=16, issue=12, pageStart=7448, pageEnd=7458, url=null, language=null, rfNumber=[150], rfOrder=149, authorNames=DONG H, SONG K, HE J, journalName=IEEE Transactions on Industrial Informatics, refType=null, unstructuredReference=DONG HSONG KHE Jet al. PGA-Net: pyramid feature fusion and global context attention network for automated surface defect detection[J]. IEEE Transactions on Industrial Informatics202016(12): 7448-7458., articleTitle=PGA-Net: pyramid feature fusion and global context attention network for automated surface defect detection, refAbstract=null), Reference(id=1276896936401703539, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2017, volume=17, issue=23, pageStart=7935, pageEnd=7944, url=null, language=null, rfNumber=[151], rfOrder=150, authorNames=GAN J, LI Q, WANG J, journalName=IEEE Sensors Journal, refType=null, unstructuredReference=GAN JLI QWANG Jet al. A hierarchical extractor-based visual rail surface inspection system[J]. IEEE Sensors Journal201717(23): 7935-7944., articleTitle=A hierarchical extractor-based visual rail surface inspection system, refAbstract=null), Reference(id=1276896936468812404, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2014, volume=54, issue=11, pageStart=2598, pageEnd=2607, url=null, language=null, rfNumber=[152], rfOrder=151, authorNames=SONG K C, HU S P, YAN Y H, journalName=ISIJ International, refType=null, unstructuredReference=SONG K CHU S PYAN Y Het al. Surface defect detection method using saliency linear scanning morphology for silicon steel strip under oil pollution interference[J]. ISIJ International201454(11): 2598-2607., articleTitle=Surface defect detection method using saliency linear scanning morphology for silicon steel strip under oil pollution interference, refAbstract=null), Reference(id=1276896936535921269, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2013, volume=null, issue=1, pageStart=429094, pageEnd=null, url=null, language=null, rfNumber=[153], rfOrder=152, authorNames=SONG K, YAN Y H, journalName=Mathematical Problems in Engineering, refType=null, unstructuredReference=SONG KYAN Y H. Micro surface defect detection method for silicon steel strip based on saliency convex active contour model, mathematical problems in engineering[J]. Mathematical Problems in Engineering2013(1): 429094., articleTitle=Micro surface defect detection method for silicon steel strip based on saliency convex active contour model, mathematical problems in engineering, refAbstract=null), Reference(id=1276896936628195958, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=153, issue=null, pageStart=107541, pageEnd=null, url=null, language=null, rfNumber=[154], rfOrder=153, authorNames=ZHANG S, ZHANG Q, GU J, journalName=Mechanical Systems and Signal Processing, refType=null, unstructuredReference=ZHANG SZHANG QGU Jet al. Visual inspection of steel surface defects based on domain adaptation and adaptive convolutional neural network[J]. Mechanical Systems and Signal Processing2021153: 107541., articleTitle=Visual inspection of steel surface defects based on domain adaptation and adaptive convolutional neural network, refAbstract=null), Reference(id=1276896936699499127, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2023, volume=145, issue=null, pageStart=110578, pageEnd=null, url=null, language=null, rfNumber=[155], rfOrder=154, authorNames=HU X, YANG J, JIANG F, journalName=Applied Soft Computing, refType=null, unstructuredReference=HU XYANG JJIANG Fet al. Steel surface defect detection based on self-supervised contrastive representation learning with matching metric[J]. Applied Soft Computing2023145: 110578., articleTitle=Steel surface defect detection based on self-supervised contrastive representation learning with matching metric, refAbstract=null), Reference(id=1276896936779190904, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=1, volume=284, issue=null, pageStart=115917, pageEnd=null, url=null, language=null, rfNumber=[156], rfOrder=155, authorNames=YANG X CA, journalName=Engineering Structures, refType=null, unstructuredReference=YANG X CA1, FAN Y L, BAO Y Q,et al. Task-aware meta-learning paradigm for universal structural damage segmentation using limited images[J]. Engineering Structures, 2023, 284: 115917., articleTitle=FAN Y L, BAO Y Q,, refAbstract=null), Reference(id=1276896936863076985, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2022, volume=33, issue=6, pageStart=2365, pageEnd=2377, url=null, language=null, rfNumber=[157], rfOrder=156, authorNames=LI Y, CHEN Z, ZHA D, journalName=IEEE Transactions on Neural Networks and Learning Systems, refType=null, unstructuredReference=LI YCHEN ZZHA Det al. Automated anomaly detection via curiosity-guided search and self-imitation learning[J]. IEEE Transactions on Neural Networks and Learning Systems202233(6): 2365-2377., articleTitle=Automated anomaly detection via curiosity-guided search and self-imitation learning, refAbstract=null), Reference(id=1276896936955351674, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=6726, pageEnd=6733, url=null, language=null, rfNumber=[158], rfOrder=157, authorNames=RIPPEL O, MERTENS P, MERHOF D, journalName=null, refType=null, unstructuredReference=RIPPEL OMERTENS PMERHOF D. Modeling the distribution of normal data in pre-trained deep features for anomaly detection[C]//2020 25th International Conference on Pattern Recognition (ICPR). Milan: 2021: 6726-6733., articleTitle=Modeling the distribution of normal data in pre-trained deep features for anomaly detection, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1276896909293916583, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, xref=1, ext=[AuthorCompanyExt(id=1276896909302305192, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909293916583, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1Department of Structural Engineering, Tongji University, Shanghai200092, China), AuthorCompanyExt(id=1276896909310693801, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909293916583, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1同济大学建筑工程系,上海200092)]), AuthorCompany(id=1276896909621072298, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, xref=2, ext=[AuthorCompanyExt(id=1276896909633655211, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909621072298, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Performance Evolution and Control for Engineering Structures, Tongji University, Shanghai200092, China), AuthorCompanyExt(id=1276896909646238124, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909621072298, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2同济大学工程结构性能演化与控制教育部重点实验室,上海200092)]), AuthorCompany(id=1276896910027919790, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, xref=3, ext=[AuthorCompanyExt(id=1276896910040502702, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896910027919790, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3State Key Laboratory of Disaster Reduction in Civil Engineering, Tongji University, Shanghai200092, China), AuthorCompanyExt(id=1276896910048891311, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896910027919790, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3同济大学;土木工程防灾减灾全国重点实验室,上海200092)])], figs=[ArticleFig(id=1276896916667503052, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=EN, label=Fig.1, caption=Damage images of steel structures, figureFileSmall=hXqm0jHOsVTOrmVee4DJzw==, figureFileBig=33whRWrml7PzQsYG8D+tcw==, tableContent=null), ArticleFig(id=1276896916751389133, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=CN, label=图1, caption=钢结构损伤, figureFileSmall=hXqm0jHOsVTOrmVee4DJzw==, figureFileBig=33whRWrml7PzQsYG8D+tcw==, tableContent=null), ArticleFig(id=1276896916923355598, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=EN, label=Fig.2, caption=Co-occurrence network of keywords for steel structure damage, figureFileSmall=i/BY7atN0U/KrP73EmESgg==, figureFileBig=mp5Qer3T4i+MZmI2xpwSGQ==, tableContent=null), ArticleFig(id=1276896917003047375, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=CN, label=图2, caption=钢结构损伤关键词共现图, figureFileSmall=i/BY7atN0U/KrP73EmESgg==, figureFileBig=mp5Qer3T4i+MZmI2xpwSGQ==, tableContent=null), ArticleFig(id=1276896917162430928, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=EN, label=Fig.3, caption=Classification, object detection, and segmentation of damage in steel structures, figureFileSmall=ipL2Rjdlj1IukEi9+RjgYw==, figureFileBig=pZB2ZseBsRnGIcaAhmFTsw==, tableContent=null), ArticleFig(id=1276896917229539793, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=CN, label=图3, caption=钢结构损伤分类识别、目标检测、分割检测, figureFileSmall=ipL2Rjdlj1IukEi9+RjgYw==, figureFileBig=pZB2ZseBsRnGIcaAhmFTsw==, tableContent=null), ArticleFig(id=1276896917288260050, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=EN, label=Fig.4, caption=Failure cases of crack detection in steel box girders, figureFileSmall=Vt5jLRkePuTliJkv+wMZLQ==, figureFileBig=asz36jtPrJfclwnhiTYvwg==, tableContent=null), ArticleFig(id=1276896917351174611, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=CN, label=图4, caption=钢箱梁裂纹检测失败案例51, figureFileSmall=Vt5jLRkePuTliJkv+wMZLQ==, figureFileBig=asz36jtPrJfclwnhiTYvwg==, tableContent=null), ArticleFig(id=1276896917422477780, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=EN, label=Table 1, caption=

Evaluation of substrate corrosion grade

, figureFileSmall=null, figureFileBig=null, tableContent=
序号面积锈蚀率η锈蚀等级L
1无腐蚀10
20<η≤0.19
30.1<η≤0.258
40.25<η≤0.57
50.5<η≤1.06
61.0<η≤2.55
72.5<η≤5.04
82.5<η≤103
910<η≤252
1025<η≤501
1150<η0
), ArticleFig(id=1276896917514752469, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=CN, label=表1, caption=

基体腐蚀等级评价

, figureFileSmall=null, figureFileBig=null, tableContent=
序号面积锈蚀率η锈蚀等级L
1无腐蚀10
20<η≤0.19
30.1<η≤0.258
40.25<η≤0.57
50.5<η≤1.06
61.0<η≤2.55
72.5<η≤5.04
82.5<η≤103
910<η≤252
1025<η≤501
1150<η0
), ArticleFig(id=1276896917581861334, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=EN, label=Table 2, caption=

Types of coating damage

, figureFileSmall=null, figureFileBig=null, tableContent=
序号破坏特征缺陷类型
1涂层损坏导致斑点和颜色变化A
2很难看见,或者看不见的涂层腐蚀导致的发暗B
3阳极性覆盖层的腐蚀产物C
4阴极性覆盖层的腐蚀产物D
5表面出现点蚀E
6碎落、起皮、剥落F
7鼓泡G
8开裂H
9龟裂I
10鸡爪状或星状缺陷J
), ArticleFig(id=1276896917636387287, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=CN, label=表2, caption=

涂层破坏类型

, figureFileSmall=null, figureFileBig=null, tableContent=
序号破坏特征缺陷类型
1涂层损坏导致斑点和颜色变化A
2很难看见,或者看不见的涂层腐蚀导致的发暗B
3阳极性覆盖层的腐蚀产物C
4阴极性覆盖层的腐蚀产物D
5表面出现点蚀E
6碎落、起皮、剥落F
7鼓泡G
8开裂H
9龟裂I
10鸡爪状或星状缺陷J
), ArticleFig(id=1276896917728661976, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=EN, label=Table 3, caption=

Comparison of corrosion classification standards for steel structures

, figureFileSmall=null, figureFileBig=null, tableContent=
腐蚀等级腐蚀面积比例(%)
欧洲腐蚀等级标准ASTM D610ISO 4628-3
00≦0.010
10.050.01~0.030.05
20.50.03~0.10.5
310.1~0.31
430.3~1.08
581.0~3.040~50
615~203.0~10.0
745~5010.0~16.0
875~8516.0~33.0
99533.0~50.0
10>50
), ArticleFig(id=1276896917808353753, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=CN, label=表3, caption=

钢结构腐蚀等级分级标准比较

, figureFileSmall=null, figureFileBig=null, tableContent=
腐蚀等级腐蚀面积比例(%)
欧洲腐蚀等级标准ASTM D610ISO 4628-3
00≦0.010
10.050.01~0.030.05
20.50.03~0.10.5
310.1~0.31
430.3~1.08
581.0~3.040~50
615~203.0~10.0
745~5010.0~16.0
875~8516.0~33.0
99533.0~50.0
10>50
), ArticleFig(id=1276896917892239834, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=EN, label=Table 4, caption=

Datasets of surface defects in steel structures

, figureFileSmall=null, figureFileBig=null, tableContent=
数据集名称规模类型数据集来源网址
Cracks in Steel Bridge (CSB)[51]

755张512×512像素级标注

钢箱梁疲劳裂纹https://data.4tu.nl/datasets/6162a9b6-2a20-4600-8207-e9dcd53a264a
Bridge Crack Library(BCL)[147]

11000张256×256像素级标注

钢结构/混凝土裂纹https://doi.org/10.7910/DVN/RURXSH
Corrosion Condition State Classification[148]

440张512×512像素级标注

钢框架结构腐蚀4级腐蚀程度https://github.com/beric7/structural_inspection_main
Illinois-MADNet[149]检测框标注钢结构/混凝土结构损伤(腐蚀、裂纹、表面剥落、变形等)https://sites.google.com/view/illinois-madnet/home
GC10-DET[53]检测框标注金属表面损伤10类https://github.com/lvxiaoming2019/GC10-DET-Metallic-Surface-Defect-Datasets
NEU Surface Defect Database[150]

300张200×200分类/目标检测

热轧带钢缺陷(氧化铁皮、压痕、麻点、裂纹、夹杂、划痕)http://faculty.neu.edu.cn/yunhyan/NEU_surface_defect_database.html
RSDDs[151]2656张检测框标注钢轨表面损伤(裂纹、孔洞、磨损等)http://icn.bjtu.edu.cn/Visint/resources/RSDDs.aspx
Severstal Steel Defect Detection18074张像素级标注钢材表面损伤4类https://www.kaggle.com/c/severstal-steel-defect-detection/data
SLSM[152]1500张检测框标注钢结构纵横裂纹http://faculty.neu.edu.cn/yunhyan/SLSM.html
SCACM[153]2200张像素级标注裂纹(金属1200张,混凝土1000张)http://faculty.neu.edu.cn/yunhyan/SCACM.html
), ArticleFig(id=1276896917967737307, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, language=CN, label=表4, caption=

钢结构表面缺陷数据集

, figureFileSmall=null, figureFileBig=null, tableContent=
数据集名称规模类型数据集来源网址
Cracks in Steel Bridge (CSB)[51]

755张512×512像素级标注

钢箱梁疲劳裂纹https://data.4tu.nl/datasets/6162a9b6-2a20-4600-8207-e9dcd53a264a
Bridge Crack Library(BCL)[147]

11000张256×256像素级标注

钢结构/混凝土裂纹https://doi.org/10.7910/DVN/RURXSH
Corrosion Condition State Classification[148]

440张512×512像素级标注

钢框架结构腐蚀4级腐蚀程度https://github.com/beric7/structural_inspection_main
Illinois-MADNet[149]检测框标注钢结构/混凝土结构损伤(腐蚀、裂纹、表面剥落、变形等)https://sites.google.com/view/illinois-madnet/home
GC10-DET[53]检测框标注金属表面损伤10类https://github.com/lvxiaoming2019/GC10-DET-Metallic-Surface-Defect-Datasets
NEU Surface Defect Database[150]

300张200×200分类/目标检测

热轧带钢缺陷(氧化铁皮、压痕、麻点、裂纹、夹杂、划痕)http://faculty.neu.edu.cn/yunhyan/NEU_surface_defect_database.html
RSDDs[151]2656张检测框标注钢轨表面损伤(裂纹、孔洞、磨损等)http://icn.bjtu.edu.cn/Visint/resources/RSDDs.aspx
Severstal Steel Defect Detection18074张像素级标注钢材表面损伤4类https://www.kaggle.com/c/severstal-steel-defect-detection/data
SLSM[152]1500张检测框标注钢结构纵横裂纹http://faculty.neu.edu.cn/yunhyan/SLSM.html
SCACM[153]2200张像素级标注裂纹(金属1200张,混凝土1000张)http://faculty.neu.edu.cn/yunhyan/SCACM.html
)], attaches=null, journal=Journal(id=1276576551509397524, delFlag=0, nameCn=工业建筑, nameEn=Industrial Construction, nameHistory1=null, nameHistory2=null, issn=1000-8993, eissn=null, cn=11-2068/TU, coden=null, periodic=0, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=KHR9x8UxgSX2gqrJN8gfTg==, journalPrice=null, startedYear=null, abbrevIsoEn=Industrial Construction, journalRemark=null, publicationField=null, createdTime=1782289167526, updatedTime=1784021757656, createdBy=18614031015, updatedBy=13041195026, firstLetterCn=G, firstLetterEn=G, subjectCode=Engineering, subjectName=null, subjectCodeEn=Engineering, subjectNameEn=null, picCn=KHR9x8UxgSX2gqrJN8gfTg==, picEn=glrMkaIg87Oz10PZrZOnSg==, jcr=null, cjcr=null, exts=[JournalExt(id=1283843561321894544, language=CN, name=工业建筑, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1784021757679, updatedTime=1784021757679, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=, submissionAuthorUrl=https://www.scicloudcenter.com/IC, submissionEditorUrl=https://www.scicloudcenter.com/IC, submissionReviewUrl=https://www.scicloudcenter.com/IC, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1283843561376420497, language=EN, name=Industrial Construction, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1784021757692, updatedTime=1784021757692, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=, submissionAuthorUrl=https://www.scicloudcenter.com/IC, submissionEditorUrl=https://www.scicloudcenter.com/IC, submissionReviewUrl=https://www.scicloudcenter.com/IC, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1276577754012160025, websiteList=[Website(id=1276578024992669908, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1276577754012160025, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/gyjz/CN, language=CN, createTime=1782289518831, createBy=18614031015, updateTime=1782290419260, updateBy=18614031015, name=工业建筑-中文, tplId=1146099689490845704, title=工业建筑, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1276581942745231958, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024992669908, code=articleTextType, value=kx, createTime=1782290452896, updateTime=1782290452896, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581942715871827, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024992669908, code=banner, value=null, createTime=1782290452889, updateTime=1782290452889, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581942778786393, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024992669908, code=grayFlag, value=0, createTime=1782290452904, updateTime=1782290452904, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581942703288914, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024992669908, code=logo, value=https://castjournals.cast.org.cn/joweb/gyjz/CN/file/pic?fileId=ObTw7qIsmguDunMFeBGvBA==, createTime=1782290452886, updateTime=1782290452886, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581942795563611, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024992669908, code=minRunFlag, value=0, createTime=1782290452908, updateTime=1782290452908, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581942736843349, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024992669908, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/gyjz/CN/file/pic, createTime=1782290452894, updateTime=1782290452894, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581942787175002, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024992669908, code=silenceFlag, value=0, createTime=1782290452906, updateTime=1782290452906, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581942724260436, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024992669908, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1782290452891, updateTime=1782290452891, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581942757814871, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024992669908, code=themeColor, value=null, createTime=1782290452899, updateTime=1782290452899, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581942766203480, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024992669908, code=themeStyle, value=null, createTime=1782290452901, updateTime=1782290452901, creator=18614031015, updator=18614031015)]), Website(id=1276578025097527513, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1276577754012160025, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/gyjz/EN, language=EN, createTime=1782289518856, createBy=18614031015, updateTime=1782290415485, updateBy=18614031015, name=工业建筑-英文, tplId=1146101810881728533, title=Industrial Construction, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1276581912684655179, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025097527513, code=articleTextType, value=kx, createTime=1782290445729, updateTime=1782290445729, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581912659489352, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025097527513, code=banner, value=null, createTime=1782290445723, updateTime=1782290445723, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581912722403918, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025097527513, code=grayFlag, value=0, createTime=1782290445738, updateTime=1782290445738, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581912634323527, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025097527513, code=logo, value=https://castjournals.cast.org.cn/joweb/gyjz/EN/file/pic?fileId=ObTw7qIsmguDunMFeBGvBA==, createTime=1782290445717, updateTime=1782290445717, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581912739181136, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025097527513, code=minRunFlag, value=0, createTime=1782290445742, updateTime=1782290445742, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581912676266570, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025097527513, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/gyjz/EN/file/pic, createTime=1782290445727, updateTime=1782290445727, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581912730792527, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025097527513, code=silenceFlag, value=0, createTime=1782290445740, updateTime=1782290445740, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581912667877961, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025097527513, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1782290445725, updateTime=1782290445725, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581912693043788, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025097527513, code=themeColor, value=null, createTime=1782290445731, updateTime=1782290445731, creator=18614031015, updator=18614031015), WebsiteProps(id=1276581912709821005, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025097527513, code=themeStyle, value=null, createTime=1782290445735, updateTime=1782290445735, creator=18614031015, updator=18614031015)])], journalTitle=工业建筑, weixinUrl=null, journalUrl=http://gyjz.ic-mag.com/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Industrial Construction, journalPhotoCn=KHR9x8UxgSX2gqrJN8gfTg==, journalPhotoEn=glrMkaIg87Oz10PZrZOnSg==, journalFirstLetter=G, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false, interPubPlatform=, interPubPlatformUrl=null), detailUrlCn=https://castjournals.cast.org.cn/joweb/gyjz/CN/10.3724/j.gyjzG26033109, detailUrlEn=https://castjournals.cast.org.cn/joweb/gyjz/EN/10.3724/j.gyjzG26033109, pdfUrlCn=https://castjournals.cast.org.cn/joweb/gyjz/CN/PDF/10.3724/j.gyjzG26033109, pdfUrlEn=https://castjournals.cast.org.cn/joweb/gyjz/EN/PDF/10.3724/j.gyjzG26033109, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1779206400000, fullTextJson=null, articleText=null, reference=null)
收藏切换
基于计算机视觉的钢结构损伤检测综述
收藏切换
PDF下载
刘逸康 1, 2 , 张铭煊 1, 2 , 余倩倩 1, 3
工业建筑 | 工程诊治与运维数智化 2026,56(5): 215-231
收起
收藏切换
工业建筑 |工程诊治与运维数智化 2026 , 56 (5) : 215 -231
基于计算机视觉的钢结构损伤检测综述
全屏
[Author(id=1276896910116000178, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=2430876@tongji.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1276896910476710326, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, authorId=1276896910116000178, language=EN, stringName=Yikang LIU, firstName=Yikang, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1Department of Structural Engineering, Tongji University, Shanghai200092, China
2Key Laboratory of Performance Evolution and Control for Engineering Structures, Tongji University, Shanghai200092, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1276896912146043319, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, authorId=1276896910116000178, language=CN, stringName=刘逸康, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1同济大学建筑工程系,上海200092
2同济大学工程结构性能演化与控制教育部重点实验室,上海200092, bio={"content":"

刘逸康,硕士研究生,主要从事钢结构性能检测方向研究,

"}, bioImg=null, bioContent=

刘逸康,硕士研究生,主要从事钢结构性能检测方向研究,

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1276896909293916583, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, xref=1, ext=[AuthorCompanyExt(id=1276896909302305192, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909293916583, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1Department of Structural Engineering, Tongji University, Shanghai200092, China), AuthorCompanyExt(id=1276896909310693801, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909293916583, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1同济大学建筑工程系,上海200092)]), AuthorCompany(id=1276896909621072298, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, xref=2, ext=[AuthorCompanyExt(id=1276896909633655211, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909621072298, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Performance Evolution and Control for Engineering Structures, Tongji University, Shanghai200092, China), AuthorCompanyExt(id=1276896909646238124, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909621072298, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2同济大学工程结构性能演化与控制教育部重点实验室,上海200092)])]), Author(id=1276896912238318009, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1276896912540307900, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, authorId=1276896912238318009, language=EN, stringName=Mingxuan ZHANG, firstName=Mingxuan, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1Department of Structural Engineering, Tongji University, Shanghai200092, China
2Key Laboratory of Performance Evolution and Control for Engineering Structures, Tongji University, Shanghai200092, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1276896912615805373, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, authorId=1276896912238318009, language=CN, stringName=张铭煊, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1同济大学建筑工程系,上海200092
2同济大学工程结构性能演化与控制教育部重点实验室,上海200092, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1276896909293916583, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, xref=1, ext=[AuthorCompanyExt(id=1276896909302305192, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909293916583, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1Department of Structural Engineering, Tongji University, Shanghai200092, China), AuthorCompanyExt(id=1276896909310693801, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909293916583, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1同济大学建筑工程系,上海200092)]), AuthorCompany(id=1276896909621072298, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, xref=2, ext=[AuthorCompanyExt(id=1276896909633655211, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909621072298, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Performance Evolution and Control for Engineering Structures, Tongji University, Shanghai200092, China), AuthorCompanyExt(id=1276896909646238124, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909621072298, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2同济大学工程结构性能演化与控制教育部重点实验室,上海200092)])]), Author(id=1276896912691302847, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=qianqian.yu@tongji.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1276896913106538946, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, authorId=1276896912691302847, language=EN, stringName=Qianqian YU, firstName=Qianqian, middleName=null, lastName=YU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 3, address=1Department of Structural Engineering, Tongji University, Shanghai200092, China
3State Key Laboratory of Disaster Reduction in Civil Engineering, Tongji University, Shanghai200092, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1276896913395945923, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, authorId=1276896912691302847, language=CN, stringName=余倩倩, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 3, address=1同济大学建筑工程系,上海200092
3同济大学;土木工程防灾减灾全国重点实验室,上海200092, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1276896909293916583, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, xref=1, ext=[AuthorCompanyExt(id=1276896909302305192, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909293916583, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1Department of Structural Engineering, Tongji University, Shanghai200092, China), AuthorCompanyExt(id=1276896909310693801, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896909293916583, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1同济大学建筑工程系,上海200092)]), AuthorCompany(id=1276896910027919790, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, xref=3, ext=[AuthorCompanyExt(id=1276896910040502702, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896910027919790, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3State Key Laboratory of Disaster Reduction in Civil Engineering, Tongji University, Shanghai200092, China), AuthorCompanyExt(id=1276896910048891311, tenantId=1146029695717560320, journalId=1276577754012160025, articleId=1276896904998949269, companyId=1276896910027919790, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3同济大学;土木工程防灾减灾全国重点实验室,上海200092)])])]
刘逸康1, 2 , 张铭煊1, 2, 余倩倩1, 3
作者信息
  • 1同济大学建筑工程系,上海200092
  • 2同济大学工程结构性能演化与控制教育部重点实验室,上海200092
  • 3同济大学;土木工程防灾减灾全国重点实验室,上海200092
通讯作者:
余倩倩,博士,教授,主要从事结构性能演化与控制方向研究,
A Review of Computer Vision-Based Damage Detection in Steel Structures
Yikang LIU1, 2 , Mingxuan ZHANG1, 2, Qianqian YU1, 3
Affiliations
  • 1Department of Structural Engineering, Tongji University, Shanghai200092, China
  • 2Key Laboratory of Performance Evolution and Control for Engineering Structures, Tongji University, Shanghai200092, China
  • 3State Key Laboratory of Disaster Reduction in Civil Engineering, Tongji University, Shanghai200092, China
出版时间: 2026-05-20 doi: 10.3724/j.gyjzG26033109
文章导航
收藏切换

高效可靠的钢结构健康监测是保证结构安全、延长服役寿命的关键。当前,计算机视觉(Computer Vision, CV)技术因其非接触、效率高和自动化程度高等优势,正逐渐成为钢结构运维检测的重要技术手段。围绕钢结构表面裂纹与腐蚀损伤,系统梳理了计算机视觉技术在钢结构损伤检测中的研究进展,归纳了图像分类、目标检测和图像分割等主要方法,重点总结了现有研究在微小目标识别、复杂背景抑制、小样本训练及工程现场部署等方面的关键优化策略。现有研究表明,计算机视觉技术有效提升了钢结构损伤检测的自动化、智能化和精细化水平,但在数据集规范化、模型抗干扰能力、跨场景泛化能力和轻量化实时推理等方面仍有进一步发展空间。

钢结构  /  损伤检测  /  计算机视觉  /  深度学习

Efficient and reliable structural health monitoring is essential for ensuring the safety and extending the service life of steel structures. Owing to the advantages of non-contact nature, high efficiency, and a high degree of automation, computer vision (CV) has gradually become an important technology for the inspection and maintenance of steel structures. Focusing on surface cracks and corrosion damage of steel structures, this review systematically summarizes the recent research progress in CV-based damage detection and outlines the major approaches, including image classification, object detection, and image segmentation. Particular attention is paid to key optimization strategies for small object detection, robustness under complex backgrounds, few-shot learning, and on-site deployment. Existing studies indicate that CV has significantly improved the automation, intelligence, and precision of damage detection for steel structures. However, further advances are still required in dataset standardization, model robustness to interference, generalization capability across scenarios, and lightweight real-time inference.

steel structure  /  damage detection  /  computer vision  /  deep learning
刘逸康, 张铭煊, 余倩倩. 基于计算机视觉的钢结构损伤检测综述. 工业建筑, 2026 , 56 (5) : 215 -231 . DOI: 10.3724/j.gyjzG26033109
Yikang LIU, Mingxuan ZHANG, Qianqian YU. A Review of Computer Vision-Based Damage Detection in Steel Structures[J]. Industrial Construction, 2026 , 56 (5) : 215 -231 . DOI: 10.3724/j.gyjzG26033109
钢结构因强度高、自重轻、施工便捷等优势,在桥梁、工业建筑及公共建筑等工程领域得到广泛应用1。随着建筑业进入高质量发展阶段,钢结构行业的发展重心正由规模扩张逐步转向绿色化、智能化与工业化并重的发展路径,在城市更新、新型基础设施建设及既有工程运维中的作用日益凸显2。在长期服役过程中,既有钢结构不可避免地受到环境作用与服役荷载的共同影响,进而产生多种形式的性能劣化与局部损伤,如疲劳、腐蚀、螺栓脱落等3-4图1)。传统钢结构损伤检测主要依赖人工目视巡检以及超声、磁粉等常规无损检测技术。这类方法虽然在局部精细检测中具有一定优势,但通常存在作业效率较低、对人员经验依赖较强、检测成本较高以及大范围连续巡检困难等问题,而常规非破坏性检测技术5-6,虽具备一定量化能力,但在大尺度结构快速筛查中成本高昂、操作复杂,难以实现常态化监测,难以充分满足现代大型复杂钢结构高频次、全覆盖的智能化运维需求。
随着人工智能技术和低成本图像采集硬件的快速发展,基于计算机视觉的钢结构损伤检测方法因具有非接触、高效率和自动化程度高等特点,正逐渐成为钢结构工程运维中的重要研究方向和关键技术手段7。该方法早期主要是数字图像处理技术(IPT)8,包括形态学处理9、阈值分割10、二值化11、直方图均衡12、边缘检测13等,对于理想光照和纹理下的损伤图像处理具有一定的适用性,但其精度取决于选取特征的质量,在实际工程应用中,裂纹、腐蚀等小目标往往受限于复杂背景干扰、特征选取困难等问题。
经过多年发展,深度学习(Deep Learning, DL)算法能够从输入图像中自动提取适用于该任务的特征,无需依赖经验和人为设计特征,在特征提取方面比图像处理方法更具鲁棒性,现已被广泛应用于土木工程领域中的损伤识别。在Web of Science对“计算机视觉”“钢结构”“损伤”进行近5年文献的搜索,对96篇论文进行关键词共现图绘制,如图2所示。早期研究多遵循工程实施顺序14或围绕具体损伤类型15展开,之后研究视角转向方法学分类,如依据数据标注范式16或技术原理类别17构建框架,并深入剖析核心网络结构的技术演进脉络与共性挑战18。近期研究进一步体现出细分化与交叉融合趋势,在检测任务维度上细化为分类、定位、分割19,并针对纹理干扰等具体特征进行专述20,部分将视觉检测的功能从静态缺陷识别拓展至动态参数测量与多模态感知21,从而构建起更为系统且面向实际工程约束的知识体系。
随着Transformer架构、扩散模型、自监督预训练等新技术的兴起以及边缘计算设备、无人机自主巡航、5G传输等硬件平台的进步,实际钢结构损伤检测中面临的小目标、小样本检测22以及现场实时推理难的问题迎来了新的算法突破与技术革新。
钢结构表面裂纹是其服役期间最常见的损伤形式之一,通常由疲劳问题引起,高发于焊缝、切口、连接细节等应力集中区域。传统的人工目视检测存在检测效率低、主观性强等局限,难以满足重大工程海量巡检与高可靠性要求23-24。近年来,计算机视觉技术与深度学习方法不断发展,已能够针对钢结构疲劳高发区域进行结构表面裂纹的识别、定位与定量表征25-27
基于计算机视觉和深度学习的裂纹检测方法主要是用大量标注的图像训练深度学习模型进行裂纹检测,主要包括三类任务,即图像分类、目标检测和图像分割,如图3所示。
1)图像分类算法能够自动对输入图像中的裂纹、腐蚀和螺栓缺失等损伤进行分类,当前主流网络以卷积神经网络(CNN)为核心,如AlexNet28、VGGNet29、ResNet30、GoogLeNet31和DenseNet32等,通常模型相对简单,计算速度较快。Vannocci等33在热轧钢带缺陷场景下的对比研究表明,CNN模型对缺陷的分类性能显著优于传统机器学习方法,测试集准确率可达90%;而针对钢桥检测场景,Xu等34提出一种基于图像分块的深度融合分类策略,将消费级相机采集的高分辨率图像切分为多尺度子图并进行判定,有效规避了全局视角的特征丢失。
2)目标检测算法则通过边界框(Bounding Box)实现了裂纹的空间定位。现有方法主要包括以R-CNN系列为代表的两阶段检测框架35-37和以YOLO系列为代表的一阶段检测框架38-40,通常两阶段算法更侧重精度,而一阶段算法则更侧重效率。Cha等41将Faster R-CNN用于五类结构损伤的同步检测,五类损伤的平均精度(mAP)达到87.8%;Suh等42使用的Deep Faster R-CNN模型在多类型损伤自动检测与定位中取得了77.97%的平均精度。
3)相比前两者,图像分割算法实现了像素级裂纹表征,不仅能精确提取裂纹的边界与细部形态,还能直接服务于裂纹长度、宽度等关键几何参数的计算。当前主流网络多采用编码器-解码器结构,如FCN43、U-Net44和Deeplabv3+45-47等模型及其改进模型。Liu等48提出的DeepCrack能够以端到端的方法预测逐像素的裂纹分割,在包含537张裂纹图像的自建数据集上平均交并比(mIoU)达到85.9%;类似地,Dong等49也针对钢结构大幅面图像提出了基于改进U-Net的裂纹分割框架,在测试集上mIoU达到65.06%,优于FCN网络的57.74%。
这三类算法并非完全独立,而是在不同工程需求下各有适用性,如分类、目标检测算法适用于海量巡检图像的快速筛查与标记,而分割算法能提供更高精度的裂纹几何尺寸与细部形态,因此实际钢结构表面裂纹检测往往根据任务需求选择不同的算法。
随着深度学习的发展,更多研究关注到钢结构相比其他结构或建筑的特殊性,相关研究的核心问题逐渐集中于复杂背景下微小裂纹的准确检测、小样本条件下的模型训练以及面向现场巡检的实时部署等方面。
不同于混凝土裂纹,钢结构裂纹尺寸更加微小、形态细长,一般在图像中的像素占比也很小,因此属于典型小目标检测问题22。此外,真实裂纹图像中往往包含焊缝、划痕和阴影等复杂背景干扰,在形态和纹理上容易与真实裂纹混淆,因此在模型判断时容易产生误检[图4(g)图4(h)]和漏检[图4(h)图4(i)50-51
当前,有许多研究对深度学习模型进行优化以增强其对小目标的感知能力。一种优化思路是在模型中集成金字塔型的采样方法,提取图像在不同尺度下的特征图,将所有特征图同时用于模型后续的检测头,以此增强模型对小目标的检测能力52。需要注意的是,该方法在提升小裂纹检测能力的同时,也可能使模型对噪声更加敏感,在降低漏检率的同时误检率增加,如引入特征金字塔(FPN)53改进后的Faster R-CNN模型对细裂纹的平均精度均值(mAP)仅为50.1%54。除FPN外,也有研究通过残差卷积和上下文编码增强细裂纹特征提取能力,提出的R2CE-Net裂纹检测准确率达到98.62%、mIoU达到80.93%55
另一种优化思路则是将高分辨率的裂纹图像划分为多张低分辨率的子图,在局部尺度上提高裂纹显著性,再通过后续融合恢复全图结果5056-57。Meng等58将高分辨率钢箱梁图像先划分为子图,再进行裂纹区域定位和边缘细化,测试集召回率(Recall)达到0.95;Zhang等50采用集成式框架,将前端判别、像素级分割与后续推理结合起来,以提高复杂背景下钢箱梁裂纹的识别可靠性。
总之,合理引入多尺度采样与图像切分方法,可以一定程度上缓解微小裂纹像素占比低、背景干扰大等特征问题,实现复杂工程环境中裂纹检测精度的提升。
除裂纹尺寸小及实际背景复杂的特点外,当前钢结构裂纹相关的公开数据集较少,而深度学习模型极大依赖于训练数据,小样本可能无法完全发挥模型的性能,甚至可能产生负面影响,因此许多研究尝试对数据集构建和模型训练过程进行优化,进而增大数据集体量,或使得模型在小数据集上也能训练得到较好的结果,从而增强裂纹检测的精度。
在数据集构建时,除旋转、平移、错切、反转、缩放、添加噪声和颜色变换等基础数据增强方法外,也可以使用虚拟合成数据构建数据集,更加省力高效。如创建具有疲劳裂纹的钢结构表面的随机纹理并映射至三维模型中,通过改变照明条件和相机角度来合成图像以增强真实数据集,研究结果表明,相比于仅使用真实数据,采用包含真实图像和虚拟合成图像的混合数据集训练模型,裂纹检测准确率从49%提升到62%,交并比(IoU)从35%提升到40%59。另外,当数据集中包含更多的复杂背景时,模型对于背景的抗干扰能力也会增强。有研究表明,在训练集中针对性地引入含有焊缝、划痕等干扰信息的背景切片(Patch),可大幅增强模型对现实场景的鲁棒性并显著降低假阳性率5160-61
除采用上述方法扩展真实数据集外,迁移学习也是一种针对小样本学习的有效方法4762。迁移学习是指在已采用大型数据集预先训练的模型基础上,再针对新问题小样本数据进行二次训练的方法。由于深度学习模型浅层部分主要检测边缘、纹理等特征,而这些特征对于类似场景是通用的,因此通过迁移学习可以直接利用浅层的权重参数,仅对深层部分重新训练,从而节省训练时间,且使得模型对于小样本也能有较好的性能。Dung等63采用了在ImageNet数据集上预训练的VGG16网络进行钢桥裂纹分类检测,结果表明迁移学习相比于从零训练裂纹检测的准确率从90%提升到97%;Yu等64提出的基于视频流特征点跟踪的钢表面裂纹检测方法,其用于特征匹配的改进LoFTR模型也采用了类似的预训练和微调策略,预训练阶段使用ScanNet数据集65,并用自建的钢板裂纹数据集进行了微调,使模型有效学习并适应目标领域的特性,有效提升了钢表面图像的匹配性能。
综上所述,针对钢结构表面裂纹检测面临的小样本问题,通过不同数据增强方法扩充原始数据集及基于大型通用数据集预训练模型进行微调与迁移均是增强裂纹检测精度的可行方案。
在实际工程部署中,钢结构表面裂纹检测模型的实时推理能力主要受两方面制约:1)现场图像通常分辨率较高而裂纹占比很小,直接进行整图分割计算代价较高。2)无人机或移动终端的算力、存储和续航能力均较为有限,难以长期运行参数量大、推理链路复杂的深度模型。因此,面向现场的模型设计更强调在有限算力条件下兼顾检测速度、稳定性与量化能力。当前提高检测效率的技术路径主要包括粗到细联动检测和网络轻量化重设计等。
粗到细的联动检测流程是提升现场效率最直接的手段,其核心是充分利用较快的分类器和检测器筛查候选区域,再在局部尺寸上用分割模型进行精细分割与尺寸计算。Han等66先利用YOLOv3粗略提取候选区域后,联合多个Deeplabv3+模型进行像素级精细提取,并辅以全景拼接技术直观重构了裂纹形貌。Meng等67进一步提出基于无人机的实时自动裂纹检测方法,采用轻量分类、轻量分割、高精度分割和裂纹宽度测量相结合的多阶段流程,本质也是通过分级推理降低整图精细识别的计算负担。该方法既保留了局部细节,又避免了整图高精度推理带来的高时延,较适用于钢结构裂纹的现场检测。
此外,轻量化的模型设计也是提高推理速度的方法之一。早期的轻量化更多体现在以SqueezeNet68、MobileNet69、EfficientNet70和ShuffleNet71等轻量骨干替换传统卷积主干,以降低参数量和内存占用。近年来则进一步扩展到对特征融合路径72和检测头的重设计,Lu等73将DeepLabv3+主干替换为MobileNetv3,并对上下文特征提取模块进行重设计,在兼顾识别精度的同时参数量、浮点运算和训练时间分别减少了91.18%、73.18%和35.55%。类似地,改进的CSCP-YOLO模型将参数量大幅压缩至14.3 MB,同时在高端算力下达到了186.2每秒帧率(FPS)的高频实时检测,但mAP略微下降到79.1%74。因此,在钢结构表面裂纹检测中,通过轻量骨干与关键模块协同优化可以有效提升实时推理效率。
现阶段,粗到细联合检测及网络轻量化重设计已在钢桥裂纹与钢结构表面裂纹研究中展现出较好的工程适用性。此外模型剪枝、知识蒸馏等模型压缩方法也可压缩模型规模75
钢结构表面裂纹的计算机视觉检测方法现已形成由分类识别、目标定位到像素级分割与几何量化的基本技术路径,但从工程应用角度看,现有方法仍面临裂纹尺寸微小、实际图像背景复杂、小样本训练和实时部署算力限制等挑战。
需要指出的是,当前计算机视觉方法主要针对结构表面可见裂纹,通常难以检测防腐或防火涂层下的近表面裂纹或缺陷。未来通过引入其他模态的图像或数据,如热成像76-78、激光成像79、结构光80或深度图81等,利用不同模态之间的互补信息提升裂纹分割与量化性能是更具潜力的研究方向。此外,通过纳米涂层传感器与电位差监测来跟踪疲劳裂纹萌生与扩展也是涂层下近表面裂纹检测的方向之一82
因此,未来钢结构疲劳裂纹检测的发展方向将集中于早期微小裂纹的准确识别、复杂背景下的稳健检测、小样本条件下的有效训练,以及面向现场应用的实时推理与多模态协同感知。
随着我国基础设施建设进入存量时代,大量服役中的钢结构设施面临长期环境侵蚀、疲劳累积和材料退化等问题,其中腐蚀作为最普遍的劣化形式之一,严重威胁结构安全性和耐久性83。尤其在沿海、高湿度或工业污染区域,钢材表面氧化反应加速,形成局部点蚀、均匀腐蚀乃至结构性截面削弱,若未能及时识别与干预,可能导致突发性断裂事故。已有综述集中于特定检测场景84、算法模型85,Das等86在2024年对钢结构图像腐蚀检测进行了系统评述,通过融合多模态数据构建量化评估体系,致力于检测后的评估与工程整合。基于此,本章将聚焦于腐蚀等级的精细化划分与面向实际应用的算法优化。
钢结构构件表面通常有防腐防火涂层保护,因此仅针对涂装前钢材的腐蚀等级分类标准并不适用于钢结构质量检测中的腐蚀等级分类。我国的GB/T 6461—2002 《金属基体上金属和无机覆盖层 经腐蚀试验后的试样和试件的评级》87中对于基体的腐蚀等级和覆盖层的破坏类型评价如表1表2所示。
欧洲、美国及国际标准针对涂层钢结构的腐蚀评价,均以腐蚀面积比例为核心划分依据,但在等级设置上各有不同,如表3所示。欧洲标准将腐蚀程度分为10级;美国ASTM D610标准88将腐蚀类型区分为点蚀、普通腐蚀与针蚀,并进一步按面积比例划分出11个腐蚀等级;ISO 4628-3标准89则将其分为6级。这三种分类体系虽然划分严密,但在工程实践的具体应用中仍存在明显不足,尤其是对腐蚀面积比例在0%~10%区间内的过细分级,增加了腐蚀图像自动识别的难度,已超出当前常见深度学习算法能够稳定达到的识别精度。现有算法多基于标准简化调整,设计不同场景下钢结构腐蚀的分级90
钢结构腐蚀检测的核心目标是获取腐蚀的相关特征参数,包含腐蚀类型、面积、三维信息等。腐蚀面积直观反映钢结构表面的腐蚀区域,是腐蚀等级划分的核心依据;腐蚀三维信息决定构件的结构承载能力,直接关系到钢结构的安全服役性能。下文将结合具体技术路径梳理其发展现状。
对于图像采集设备采集的图像,既需要判断该图像中有无腐蚀,又需要对图像中存在腐蚀的程度进行分级,以便采取不同的维修措施。图像分类问题,即判断有无腐蚀,Petricca等91比较传统图像处理和深度学习在腐蚀分类的性能,结果表明深度学习方法在实际应用中表现更好。Atha等92比较不同CNN模型的腐蚀检测性能,并探究了网络层数、输入图像色彩空间以及滑动窗口大小对分类准确率的影响。对于腐蚀等级评估,即自动识别图像中的腐蚀严重程度,其核心依赖腐蚀面积的量化分析。Holm等93比较使用了AlexNet、VGG16、ResNet和GoogLeNet的性能,结果表明,VGG-16总体表现最佳,在损伤检测类别中,AlexNet表现最佳,准确率为99.14%。上述结果表明,将CNN运用到腐蚀图像分类任务中,可以得到较理想的性能94-95。Xu等96为进一步提升模型的性能,将多个CNN模型的识别结果进行结合,发现单个CNN分类器的精度最高达90%,通过集成学习可以将腐蚀等级分类精度提高到93%。
目前研究的CNN分类器适用于特定环境下的腐蚀构件,准确率基本可以达到使用要求,是一种快速、准确的腐蚀等级评估方法。但仅对图像进行分类已无法满足腐蚀检测的要求,需要确定腐蚀位置,并对腐蚀区域进行定性或者定量描述。基于深度学习的目标检测方法,大致可以分为两类,即单阶段(One-Stage)检测器和二阶段(Two-Stage)检测器。最具代表性的是Faster RCNN97-98和YOLO99系列,它们通过预定义不同尺寸形状的anchor来检测不同尺寸形状的目标区域,从而实现目标的定位和分类。Jia等100开发了用于检测公共设施腐蚀的Corrosion-YOLOv5s模型,它使用金属腐蚀图像数据集表现出出色的检测精度。Nabizadeh等101训练了YOLOv3、YOLOv5s、YOLOv7模型,并将其在混凝土腐蚀图像上进行测试,发现YOLOv5在混凝土腐蚀检测方面表现最好。Ameli等102构建了一个带有注释的钢桥腐蚀开源数据集,通过训练,验证了Mask RCNN和YOLOv8算法在腐蚀分割和分级方面的性能良好。Yu等103提出的YOLOv5-GOLD-NWD模型在涂层金属表面腐蚀检测中实现了78%的检测精度,较传统YOLOv5模型提升了4%。
在钢结构腐蚀检测的工程实践中,若腐蚀沿构件对角线方向呈细长形态分布,矩形边界框往往包含大量背景区域,导致腐蚀面积估算误差显著。为从根本上解决边界框冗余引起的误判问题,需采用像素级分割方法,分为语义分割与实例分割两类算法。语义分割算法可将图像中所有像素按腐蚀、钢材基体、背景等语义属性进行分类,实现腐蚀区域与非腐蚀区域的全局像素级区分,实例分割算法则在此基础上进一步区分不同腐蚀个体104。针对分割算法的工程落地,研究者开展了大量实践。王达磊等105基于U-Net架构设计了深度神经网络,对提取到的腐蚀区域进行了定量分析,但模型主要在自建数据集上有效。Pirie等104对CNN、YOLO和Mask RCNN的性能进行比较,发现Mask RCNN针对水下腐蚀分割图像表现最优。Fondevik等106基于自建数据集训练Mask R-CNN和PSPNet网络,结果发现在红色背景下效果较差,数据集规模不够。Duy等85把具有DenseNet和PSPNet层的全卷积网络与U-Net、E-Net进行比较,发现实际应用中同时包含U-Net和FCN的性能更好。Katsamenis等107将RGB图像输入经过FCN、U-Net和Mask RCNN三种语义分割网络,通过形态学运算,得到高置信度和低置信度的区域,运用分水岭算法得到最终的准确边界。整体来看,深度学习腐蚀检测精度受数据集规模和分割数据集标注困难的限制,同时腐蚀的形状不规则、区域不连续,标注主要依赖于标注者的主观性,会导致标注质量不稳定且耗时耗力。
腐蚀的深度检测旨在通过技术手段获取场景中目标的三维空间深度信息,是计算机视觉领域实现从二维感知到三维理解的核心任务之一。传统深度检测方法,如结构光法、立体匹配法,受限于手工特征设计,在复杂场景下鲁棒性不足,而深度学习凭借其强大的特征自动学习能力,显著突破了传统方法的瓶颈,成为当前腐蚀深度检测研究的主流方向。Akhlaghi等108针对埋地管道点蚀深度预测的多环境特征耦合问题,提出两类深度学习模型,该模型通过反馈连接记忆土壤氯化物含量与点蚀深度的历史关联,在高盐土壤场景的误差<5%,证明深度学习在复杂环境特征建模上的优势。Gao等109针对铁路滑床板的锈蚀深度检测,提出VGG和ResNet50的融合模型,在72 h中性盐雾试验(等效自然环境3a)样本上,深度预测相对误差平均值仅4.08%,显著优于超声检测,且在现场粉尘干扰下仍保持较小的误差。Zhang等110提出基于CNN的区域关联模型,突破了传统深度学习孤立区域建模的局限,通过区域交互特征提取,更贴合工程中管道多锈蚀共存的实际场景。
目前对于仅依靠RGB可见光图像的深度学习模型较少。秦荣杰111将跨尺度注意力Transformer用于钢板锈蚀检测,利用单反相机收集钢板不同锈蚀周期的锈蚀形貌照片数据,以及对锈蚀钢板进行形貌扫描确定其锈蚀深度,并对钢板锈蚀产物进行X射线衍射(XRD)成分分析,得出钢板锈蚀深度与表观形貌特征的关联关系。基于实验室数据构建60000张4个深度梯度(0.01~0.65 mm)的均匀锈蚀图像,发现模型锈蚀深度对应4个类型的识别精度达96.86%。但模型未针对深度特征优化,仅沿用通用分类网络结构,同时数据库基于理想盐雾环境,泛化性不足。Altabey等112针对钢管道锈蚀深度检测提出3D阴影建模与CNN结合的方法,利用锈蚀深度低于周围区域产生的阴影特征,通过双向投影构建阴影图,用CNN提取阴影特征反推深度。该方法通过迭代最近点算法(ICP)排除光照干扰,大幅提高精度,但其过于依赖光照条件,且未考虑锈蚀产物对阴影的影响。仅依靠图片信息的核心不足源于二维视觉信息与三维深度物理量的本质矛盾,二者关联性弱且易受干扰,同时数据集稀缺,标注成本较高,即使训练后的模型也因缺乏可解释性难以满足工程信任需求,环境适应性差。
单纯RGB图像更适合面积与等级评定,而要实现定量腐蚀深度的估计,常需引入多光谱、高光谱或X射线等具有更强物理相关性的成像,并结合回归或图像修补框架。典型方法113-117已能在试验条件下达到高精度连续深度预测或厚度损失预测,但向复杂工程现场推广仍需解决环境干扰、标定与真值获取等问题。
目前针对钢结构腐蚀检测的视觉深度研究仍较为匮乏,相关检测技术的精度尚未达到工程应用需求,因此当前基于计算机视觉的钢结构腐蚀检测研究仍以腐蚀分类、面积识别和分级为核心重点。
在腐蚀检测中,为处理数据集图像数量不足的问题,多采用基于数据扩充的方法。小样本学习的根本问题是数据缺乏,若利用已有数据生成一批新数据,可以直接解决这一问题。一般的数据增强方法是对输入图片进行翻转、旋转、平移、添加噪声或者随机裁剪96等操作。然而腐蚀损伤没有方向,也没有任何规定的尺寸、形状或绝对颜色,该方法处理的腐蚀图片,无法生成具有新特征的图片。另一种常用的方法是迁移学习(Transfer Learning),在大数据集上对模型预训练,然后在目标数据集上微调。Liu等118提出基于VGG19的Faster-RCNN网络,利用迁移学习实现了对海上结构压载舱中不同类型的涂层损坏和腐蚀的分类。许多学者在ImageNet119-120、PASCAL VOC2007和PASCAL VOC2012121等标准数据集上初始化权重,利用迁移学习将大型数据集上预训练学到的低层次特征知识迁移到小样本的腐蚀数据集上,从而解决样本数量不足的问题。然而,使用迁移学习时,大型数据集和任务数据集需要足够相关,使底层特征例如颜色特征、纹理特征、形状特征等可以保留用于任务目标检测122。目前大型的公开数据集与腐蚀相关的图片极少123,即使应用了迁移学习,仍然需要数千张输入图片作为目标数据集。
生成对抗网络(GAN)自2014年被提出以来,为应对深度学习中的小样本问题提供了一种创新思路124。其核心思想是通过生成器与判别器的对抗博弈进行学习,使生成器能够合成与真实数据高度相似的图像。然而,经典GAN存在训练不稳定、收敛慢等问题,常导致生成失败或样本质量较低。为此,研究者们相继提出了一系列改进模型。Radford等125提出的深度卷积生成对抗网络(DCGAN)通过引入卷积结构取代全连接层,有效提升了训练的稳定性和收敛速度。针对训练困难与模式崩溃问题,Gulrajani等126进一步提出了沃瑟斯坦生成对抗网络(WGAN)及其改进版本,从损失函数层面增强了训练的稳定性与生成样本的多样性。GAN在土木工程中的应用已有一些研究,Lei等127使用GAN重建了结构健康监测的丢失数据;Gao等128使用GAN进行损伤数据集扩充,并评估了生成图像的质量;Maeda等129使用GAN生成公共道路表面坑洞图像。Xu等130提出基于双CycleGAN的表面缺陷检测方法,核心创新在于构建两阶段框架,CycleGAN1通过无监督图像转换生成无缺陷和有缺陷的配对样本,CycleGAN2输出多轮迭代图像后,通过缺陷像素差异累积与阈值分割生成二值化缺陷图,全程无需人工标注缺陷特征。Wang等131提出DG-GAN双向框架,编码器无需梯度信息即可生成对抗样本,且框架无需修改现有分类器结构,可灵活适配各类损伤检测模型。在Caltech-101等数据集上的验证表明,其平均准确率超88%。
GAN类方法通过无监督样本生成、对抗干扰优化,为腐蚀小样本检测及复杂环境适配提供新思路。而半监督学习132与半自动标注133技术,从样本利用效率与标注成本控制角度,缓解了腐蚀检测中标注样本少且定位标注成本高的核心矛盾。钱企豪等134针对铜片腐蚀等级识别,提出基于颜色特征的半监督k-means算法,通过核主成分分析(KPCA)降维处理颜色直方图特征,以标准比色卡特征作为初始聚类中心,无需大量标注样本即可实现84%的识别精度。Feng等135在复杂海洋场景的半监督目标检测研究中,提出自适应对抗自训练框架,仅用2%标注样本结合全量未标注样本的检测性能,即可超越5%标注样本的全监督模型。Wang等136结合Mean Teacher模型,在仅有67%标注数据的情况下,实现90.0%的平均精度和87.1%的交并比(IoU),即使在仅33%标签可用时,性能也优于全监督模型。这进一步验证了半监督学习在降低腐蚀检测标注成本中的潜力。
CNN与Transformer架构在图像语义分割、目标检测等任务中展现出卓越性能。但在实际工程部署中,仍面临两大核心挑战,一是如何提升模型对小目标腐蚀、边缘模糊区域及复杂背景干扰下的特征提取能力;二是如何在嵌入式设备或无人机平台等资源受限环境中实现高效推理。为此,研究者们将注意力机制137引入主流网络框架,并探索轻量化策略以降低参数量与计算开销。
许多学者尝试将注意力机制嵌入编码器-解码器结构,以增强关键特征的表达权重。一种典型路径是在U-Net基础上集成通道-空间联合注意力模块。Duan等138提出一种改进型U-Net Attention架构,在跳跃连接处引入注意力门控机制,使解码器能够有选择性地融合来自编码器各层级的特征图,抑制无关背景响应,用于复杂背景下钢构件腐蚀等级的智能诊断。试验结果显示,该方法在真实工程现场采集的数据集上实现了94.1%的分割准确率,显著优于原始U-Net。Katsamenis等139构建SPLAC U-Net模型,在最后一层附加残差注意力模块,用于锈迹定位与分级任务,相比仅使用颜色阈值的传统方法,其F1和IoU指标均有明显提升。除了U-Net变体,Ma等140展示了如何将MPCA(MultiPath Coordinate Attention)整合进YOLOv8n的颈部结构,从而强化对坐标位置敏感的特征响应。MPCA的设计理念在于同时建模水平与垂直方向的空间依赖关系,相较于SE(Squeeze and Excitation Module)或CBAM(Convolutional Block Attention Module)等通用注意力模块,更适合处理具有规则几何形态的钢板表面损伤。试验证明,该改进使mAP提升了1.2%,同时能维持较低的计算负载。Fu等141开发CBG-YOLOv5s模型,通过在C3模块中嵌入CBAM,增强通道与空间维度的关注能力。此外,还设计一种轻量化的C3Ghost组件,进一步减少参数数量。目前大部分研究均未直接比较不同类型注意力机制的效果,而是默认选用某种特定结构。
轻量化部署的需求促使研究者探索更极致的模型压缩策略。资源约束主要体现在边缘设备的内存与算力限制以及无人机巡检系统对功耗与实时性的要求。单纯增加网络深度或宽度不再可行,目前主流的轻量化手段可分为三类,主干网络替换、结构重参数化与模型剪枝。Ma等140采用GhostNet作为YOLOv8的主干,借助线性变换生成冗余特征图,大幅削减浮点运算次数。Zhang等142提出YOLO-RD模型以RexNet作为主干,引入GSConv与VOV-GSCSP模块重构特征融合路径,模型在NEU-DET和GC10-DET两个公开数据集上分别取得3.7%和3.5%的mAP增益,参数量下降达40%。该方法在极端压缩情况下可能出现特征退化现象,特别是在处理低对比度目标时表现不稳定。Tan等143关注DSNet网络内部结构的优化重组,采用共享主干的双分支架构,分别执行螺栓检测与像素级腐蚀分割任务,二者输出取交集后生成检测结果,避免独立训练两个模型带来的额外开销。模型剪枝属于后训练压缩技术,Yu等144系统评估了五种分割模型在两种腐蚀图像数据集(NEA、SSCS)上的剪枝效果,涵盖线性衰减、渐进式与运动剪枝三种策略。研究表明,当稀疏度达到90%时,多数模型在SSCS数据集上的IoU下降不超过10%,而在噪声更少的NEA数据集上降幅仅为5%左右,表明在合理控制剪枝强度的前提下,可实现模型体积的显著压缩而不致严重损失性能。
目前越来越多的学者强调算法的实际落地能力。Han等145提出一种两阶段定位法,先利用超像素分割方法中的简单线性迭代聚类方法(SLIC)预处理无人机图像,再结合飞行轨迹信息精确定位腐蚀区域相对位置。Eltouny等146开发的Dmg2Former-AR模型结合视觉Transformer与拉普拉斯金字塔缩放网络,专门应对高分辨率检测图像带来的计算压力。Katsamenis等139的SPLAC U-Net服务于自动化维修机器人,要求模型既能精确定界又能识别腐蚀等级,以便规划打磨或喷涂动作。Duan等138则面向运维阶段的大规模结构评估,提出滑动窗口采样与Inception v3分类结合的流水线流程,便于批量处理航拍影像。尽管已有大量工作聚焦于单一维度优化,强调精度提升或追求模型压缩,但兼顾高精度与低延迟的系统性解决方案尚不成熟。
数据集是深度学习研究的基石,但当前钢结构损伤检测领域常用的数据集普遍存在背景单一、缺陷种类有限的问题,难以真实反映复杂多变的工业现场。实际上,生产中的图像往往来自不同成像条件、拍摄角度,构建贴近真实场景的数据集,不仅有助于开发更具实用性的算法,为算法落地部署提供关键基础,还能为无人机巡检、异常溯源等延伸任务提供必要的评估基准。
数据集的匮乏也直接加剧了小样本问题,这已成为深度学习在钢结构损伤检测等领域推广的瓶颈。深度学习方法依赖大量数据进行训练,但在实际应用中,可用损伤样本数量不足,现有公开数据集中真实多损伤、多工况的钢结构数据仍然有限123,如表4所示,钢结构裂纹和腐蚀的公开数据集很少。现有数据集还存在制作不规范、不统一的问题,导致模型在训练过程中容易出现过拟合,难以实际部署应用。当前解决小样本问题的思路主要围绕增加有效样本、降低算法对样本数量依赖两个方面。
在增加有效样本方面,传统数据扩充方法通过简单图像变换提升样本数量,但无法解决样本多样性不足的问题;GAN及其改进模型可合成高相似度无监督腐蚀样本,弥补真实样本匮乏短板并降低标注成本,部分改进模型125130实现无人工标注突破,但生成样本的真实性与适配性仍需优化。在降低样本依赖方面,迁移学习119-122通过通用数据集预训练迁移特征,缓解小样本压力,但受特征差异限制仍需进行目标样本微调;半监督学习与半自动标注技术可挖掘未标注样本信息,降低标注成本并保障性能,但其复杂场景稳定性仍需验证。此外,各类方法落地需依托统一的数据集制作标准,解决数据集不规范问题,为方法优化与工程化应用提供支撑。
钢结构服役环境的复杂性导致损伤识别难度显著提升。钢结构表面的裂纹与划痕、锈迹与表面附着物在视觉纹理、灰度特征上存在高度重叠,损伤特征混淆,常规模型难以实现精准区分,易出现误判、漏判现象。同时环境干扰显著,因光照不均、阴影遮挡、雨水附着、海洋盐雾侵蚀等现象严重破坏钢结构损伤区域的特征完整性,模型特征提取易失效。另外,传统损伤标注依赖人工完成,裂纹、腐蚀、螺栓脱落等小目标的标注耗时耗力、效率低下,易因标注人员的经验差异产生主观误差,影响模型训练的准确性与一致性。
在此背景下,单纯依赖单一数据源或传统模型优化已难以满足实际需求,当前的主要热点在于拓展多源数据融合技术。通过融合视觉图像、振动信号、红外热成像、机电阻抗(EMI)等多维度数据,实现优势互补86。视觉图像可捕捉表面损伤的直观特征,振动信号能反映结构内部力学性能的变化,红外热成像可识别钢结构涂层下腐蚀、裂纹等隐蔽损伤76-78,机电阻抗信号可实现腐蚀程度的定量表征。结合环境自适应预处理算法可减少复杂环境对特征提取的干扰,进一步提升模型在实际场景中的识别稳健性,突破单源数据的应用局限。
现有深度学习模型通常在特定结构类型与环境条件下训练,导致其特征提取能力呈现较高的场景依赖性,泛化与迁移能力有限。当应用于不同结构类型或不同服役环境时,因损伤特征分布存在差异,模型识别精度往往下降趋势显著。同时,跨结构、跨环境的特征迁移难度大,缺乏有效缓解特征分布偏移的技术手段,已成为制约该技术规模化应用的关键122
为提升模型在多样化场景下的适应能力,现有研究采用了多种训练策略:1)采用迁移学习策略,在大规模通用数据集上预训练模型119-121,结合目标场景的少量标注数据进行微调,以增强模型对新结构与新环境的适配能力。2)引入领域自适应方法,利用领域对抗训练、风格迁移等方法减小跨场景的特征分布差异,实现不同光照、纹理风格条件下钢结构损伤的跨域识别154。3)结合对比学习155、元学习156等技术,进一步增强模型对损伤特征与场景变化的自适应与泛化能力,从而推动检测技术在多类型钢结构、多服役环境中的规模化应用。
当前主流的深度学习模型如CNN、Transformer及混合架构等,通常结构复杂、参数量庞大,虽有助于提升钢结构损伤的识别精度,但也导致算力消耗显著增加。这使得模型对硬件要求高,难以部署至嵌入式设备或边缘终端等常见工业场景,难以满足现场实时检测与在线监测的需求。同时高昂的算力成本也制约了该技术在中小型工程中的推广应用。针对上述问题,模型加速技术是目前重要的改进方向。在算法层面,可在设计时引入轻量化思想,采用分组卷积、深度可分离卷积减少计算量,或直接使用轻量级网络68-71进行特征提取。在硬件层面,可借助GPU、FPGA、DSP等专用加速芯片进一步提升推理效率,从而在保证精度的前提下满足实时性要求。
同时,现有检测技术与工程运维设备的集成度较低,未实现检测流程的自动化、智能化升级,导致技术落地成本高、实用性不强。可聚焦工业化集成优化,重点推动检测技术与无人机、巡检机器人等自动化设备的深度融合。
随着深度学习技术的不断发展,钢结构损伤检测正逐步从依赖人工巡检、单一传感器检测的传统模式,向智能化、自动化与高精度的方向演进。目前,以CNN、Transformer及其混合架构为核心的技术路线已在视觉损伤识别等领域取得显著进展,初步实现了损伤的存在性判断、区域定位与程度分级,为钢结构健康监测提供了新的技术路径。然而,面对工程实际需求,现有技术仍面临一系列突出瓶颈,制约其规模化落地。展望未来,为进一步推动计算机视觉在钢结构损伤识别领域落地,以下几个方向值得重点关注。
1)模型架构设计与优化。当前钢结构损伤检测模型多依赖手工设计的网络拓扑结构,超参数选择易受主观经验影响,缺乏针对工程场景的针对性优化,难以实现性能与效率的较优平衡。未来可探索引入神经架构搜索(Neural Architecture Search, NAS)等自动机器学习技术,面向钢结构损伤检测任务自主挖掘最优网络架构,弥补人工设计的不足,为模型轻量化与检测效率提升提供新的技术路径。相关研究已证实,基于自动机器学习构建的模型在性能上优于传统人工设计的网络157-158,为该方向的探索提供了可行性支撑。目前NAS在无监督与半监督视觉任务中的应用仍有待拓展,这也为后续面向少标注腐蚀检测场景的模型优化提供研究缺口。
2)多模态信息融合。钢结构损伤种类繁杂,单一成像方式往往难以全面捕捉复杂背景下的各类特征,利用不同成像条件的差异形成信息互补是提升检测能力的重要思路。通过融合来自振动、图像、红外等多源异构数据的特征,模型能够应对更复杂的真实场景,增强对损伤的表征与区分能力。同时推动多模态系统与边缘计算、实时在线监测平台结合,是实现技术工程化落地的关键发展方向。
3)模型可解释性。尽管基于深度学习的方法性能卓越,但其决策过程通常缺乏透明性,许多无监督方法仍依赖经验性假设。对模型可解释性的研究不仅有助于理解其工作机制、推动方法创新,在实际落地中对于建立人机互信、辅助人工决策也至关重要。因此,提升模型的可解释性,是推动技术落地不可或缺的一环。
4)工业化部署技术。当前,模型轻量化与边缘计算的融合应用仍存在短板,需针对钢结构损伤识别任务的特殊性,优化知识蒸馏、网络剪枝、量化压缩等轻量化技术,在保障检测精度的前提下,降低模型参数量与计算开销,适配工程现场便携式巡检终端、无人机机载处理器等边缘设备的算力约束。此外,还需解决多设备数据格式统一、通信协议兼容、模型与工程监测平台接口衔接等问题,完善模型与设备的常态化校准、故障排查、版本更新运维机制,推动技术从实验室原型向标准化、可推广的钢结构损伤实地检测部署转型。

参考文献 引证文献
排序方式:
[1]
袁宇峰. 我国钢结构产业发展现状及其用钢需求调查[J]. 冶金管理2024(8): 10-13.
[2]
中国建筑金属结构协会建筑钢结构行业可持续发展研究课题组. 中国建筑钢结构行业发展报告(2023—2024年度)[J]. 建筑2025(7): 62-68.
[3]
幸坤涛, 赵晓青, 郭小华, . 工业建筑钢结构疲劳损伤检测、评估及加固关键技术研究[C]//第十二届中国钢铁年会论文集. 北京: 2019.
[4]
何润, 周世康, 张琦超, . 桥梁钢结构的腐蚀与防护技术研究进展[J]. 钢铁研究学报202537(5): 539-556.
[5]
WU R KZHANG HYANG R Zet al. Nondestructive testing for corrosion evaluation of metal under coating[J]. Journal of Sensors2021,2021(1): 1-16.
[6]
VASAGAR VHASSAN M KABDULLAH A Met al. Non-destructive techniques for corrosion detection: a review[J]. Corrosion Engineering, Science and Technology, 202459(1): 56-85.
[7]
陈飞圻, 薛江, 逯鹏, . 基于机器视觉的钢结构工程运维关键技术研究现状[J]. 工业建筑202555(7): 131-142.
[8]
朱洪洲, 谭祺琦, 范世平, . 基于图像技术的沥青混合料细观结构研究进展[J]. 重庆交通大学学报 (自然科学版)202140(10): 97-110.
[9]
宋伟, 左丹, 邓邦飞, . 高压输电线防震锤锈蚀缺陷检测[J]. 仪器仪表学报201637(增刊1): 113-117.
[10]
XU YLI HLI Set al. 3-D modelling and statistical properties of surface pits of corroded wire based on image processing technique[J]. Corrosion Science2016111: 275-287.
[11]
刘涛, 艾军, 张丽芳, . 基于图像处理技术的钢箱梁防腐涂层寿命预测实验研究[J]. 中国腐蚀与防护学报201333(5): 407-412.
[12]
刘淼, 薛建军, 王玲, . 基于数字图像分析的碳钢腐蚀等级评定方法[J]. 腐蚀与防护201334(11): 997-1000.
[13]
SHI TKONG J YWANG X Det al. Improved Sobel algorithm for defect detection of rail surfaces with enhanced efficiency and accuracy[J]. Journal of Central South University201623(11): 2867-2875.
[14]
姚志东, 卢佳祁, 熊梦雅, . 基于计算机视觉的钢结构表面缺陷智能识别研究综述[J]. 建筑结构202353(24): 126-135.
[15]
逯鹏, 赵天淞, 王剑, . 基于计算机视觉的钢结构表面损伤识别与健康监测综述[J]. 工业建筑202252(10): 22-27.
[16]
罗东亮, 蔡雨萱, 杨子豪, . 工业缺陷检测深度学习方法综述[J]. 中国科学:信息科学202252(6): 1002-1039.
[17]
杨泽青, 张明轩, 陈英姝, . 基于机器视觉的表面缺陷检测方法研究进展[J]. 现代制造工程2023(4): 143-156.
[18]
程锦锋, 方贵盛, 高惠芳. 表面缺陷检测的机器视觉技术研究进展[J]. 计算机应用研究202340(4): 967-977.
[19]
高艺平, 王浩, 李新宇, . 基于深度智能视觉的表面缺陷检测研究进展[J]. 工业工程202427(2): 27-36.
[20]
邓志鹏, 何施茗, 杨根, . 基于深度学习的纹理表面缺陷检测方法综述[J]. 计算机集成制造系统202531(3): 721-745.
[21]
AZIMI MESLAMLOU A DPEKCAN G. Data-driven structural health monitoring and damage detection through deep learning: state-of-the-art review[J]. Sensors202020(10): 2778.
[22]
刘颖, 刘红燕, 范九伦, . 基于深度学习的小目标检测研究与应用综述[J]. 电子学报202048(3): 590-601.
[23]
ZHANG GLIU YLIU Jet al. Causes and statistical characteristics of bridge failures: a review[J]. Journal of Traffic and Transportation Engineering (English Edition)20229(3): 388-406.
[24]
IMAM BCHRYSSANTHOPOULOS M K. A review of metallic bridge failure statistics[C]//Bridge Maintenance, Safety and Management: Proceedings of the Fifth International IABMAS Conference. Boca Raton: 2010: 3275-3282.
[25]
HAMISHEBAHAR YGUAN H, SO S, et al. A comprehensive review of deep learning-based crack detection approaches[J]. Applied Sciences202212(3): 1374.
[26]
YUAN QSHI YLI M. A review of computer vision-based crack detection methods in civil infrastructure: progress and challenges[J]. Remote Sensing202416(16): 2910.
[27]
CUI CZHANG QZHANG Det al. Monitoring and detection of steel bridge diseases: a review[J]. Journal of Traffic and Transportation Engineering (English Edition)202411(2): 188-208.
[28]
RIZHEVSKY ASUTSKEVER IHINTON G E. ImageNet classification with deep convolutional neural networks[C]//Proceedings of the 25th International Conference on Neural Information Processing Systems. New York: 2012: 1097-1105.
[29]
SIMONYAN KZISSERMAN A. Very deep convolutional networks for large-scale image recognition[PP/OL]. V6. arXiv (2015-04-10) [2026-05-07]. https://doi.org/10.48550/arXiv.1409.1556.
[30]
SZEGEDY CLIU WJIA Yet al. Going deeper with convolutions[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: 2015: 1-9.
[31]
HE KZHANG XREN Set al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: 2016: 770-778.
[32]
HUANG GLIU ZVAN DER MAATEN Let al. Densely connected convolutional networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: 2017: 4700-4708.
[33]
VANNOCCI MRITACCO ACASTELLANO Aet al. Flatness defect detection and classification in hot rolled steel strips using convolutional neural networks[C]//International Work-Conference on Artificial Neural Networks. Cham: Springer International Publishing, 2019: 220-234.
[34]
XU YBAO YCHEN Jet al. Surface fatigue crack identification in steel box girder of bridges by a deep fusion convolutional neural network based on consumer-grade camera images[J]. Structural Health Monitoring201918(3): 653-674.
[35]
REN SHE KGIRSHICK Ret al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence201739(6): 1137-1149.
[36]
王海云, 王剑平, 罗富华. 融合多层次特征Faster R-CNN的金属板带材表面缺陷检测研究[J]. 机械科学与技术202140(2): 262-269.
[37]
戴学丰, 陈慧, 朱成军. 基于改进Faster R-CNN的金属工件表面缺陷检测及实现[J]. 表面技术202049(10): 362-371.
[38]
REDMON JDIVVALA SGIRSHICK Ret al. You only look once: unified, real-time object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: 2016: 779-788.
[39]
REDMON JFARHADI A. YOLO9000: better, faster, stronger[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: 2017: 7263-7271.
[40]
WANG C YBOCHKOVSKIY ALIAO H Y M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Vancouver: 2023: 7462-7475.
[41]
CHA Y JCHOI W, SUH G, et al. Autonomous structural visual inspection using region-based deep learning for detecting multiple damage types[J]. Computer-Aided Civil and Infrastructure Engineering201833(9): 731-747.
[42]
SUH G, CHA Y J. Deep Faster R-CNN-based automated detection and localization of multiple types of damage[C]//Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2018. Bellingham: 201810598: 197-204.
[43]
LONG JSHELHAMER EDARRELL T. Fully convolutional networks for semantic segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: 2015: 3431-3440.
[44]
RONNEBERGER OFISCHER PBROX T. U-Net: convolutional networks for biomedical image segmentation[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention. Cham: 2015: 234-241.
[45]
CHEN L CZHU YPAPANDREOU Get al. Encoder-decoder with atrous separable convolution for semantic image segmentation[C]//Proceedings of the European conference on computer vision (ECCV). Cham: 2018: 801-818.
[46]
JIA XWANG YWANG Z. Fatigue crack detection based on semantic segmentation using DeepLabV3+ for steel girder bridges[J]. Applied Sciences202414(18): 8132.
[47]
PAN YZHANG L. Dual attention deep learning network for automatic steel surface defect segmentation[J]. Computer‐Aided Civil and Infrastructure Engineering202237(11): 1468-1487.
[48]
LIU YYAO JLU Xet al. DeepCrack: a deep hierarchical feature learning architecture for crack segmentation[J]. Neurocomputing2019338: 139-153.
[49]
DONG CLI LYAN Jet al. Pixel-level fatigue crack segmentation in large-scale images of steel structures using an encoder-decoder network[J]. Sensors202121(12): 4135.
[50]
ZHANG CWAN LWAN R Qet al. Automated fatigue crack detection in steel box girder of bridges based on ensemble deep neural network[J]. Measurement2022202: 111805.
[51]
KOMPANETS ADUITS R, PAI G, et al. Loss function inversion for improved crack segmentation in steel bridges using a CNN framework[J]. Automation in Construction2025170: 105896.
[52]
舒江鹏, 李俊, 马亥波,. 基于特征金字塔网络的超大尺寸图像裂纹识别检测方法[J]. 土木与环境工程学报(中英文)202244(3): 29-36.
[53]
LIN T YDOLLÁR PGIRSHICK Ret al. Feature pyramid networks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017: 2117-2125.
[54]
ZHAO WCHEN FHUANG Het al. A new steel defect detection algorithm based on deep learning[J]. Computational Intelligence and Neuroscience20212021(1): 5592878.
[55]
LI GLI XZHOU Jet al. Pixel-level bridge crack detection using a deep fusion about recurrent residual convolution and context encoder network[J]. Measurement2021176: 109171.
[56]
QUQA SMARTAKIS PMOVSESSIAN Aet al. Two-step approach for fatigue crack detection in steel bridges using convolutional neural networks[J]. Journal of Civil Structural Health Monitoring202212(1): 127-140.
[57]
TONG TLIN JHUA Jet al. Crack identification for bridge condition monitoring using deep convolutional networks trained with a feedback-update strategy[J]. Maintenance, Reliability and Condition Monitoring20211(2): 37-51.
[58]
MENG S QGAO Z YZHOU Yet al. A three-stage deep-learning-based method for crack detection of high-resolution steel box girder image[J]. Smart Structures and Systems202229(1): 29-39.
[59]
ZHAI G HNARAZAKI YWANG Set al. Synthetic data augmentation for pixel-wise steel fatigue crack identification using fully convolutional networks[J]. Smart Structures and Systems202229(1): 237-250.
[60]
TA Q BDANG N LKIM Y Cet al. Semantic crack-image identification framework for steel structures using atrous convolution-based Deeplabv3+ Network[J]. Smart Structures and Systems202230(1): 17-34.
[61]
邓露, 香超, 王维,. 基于改进编解码网络的钢箱梁疲劳裂纹分割[J]. 华中科技大学学报(自然科学版)202250(8): 66-72.
[62]
朱劲松, 李欢, 王世芳. 基于卷积神经网络和迁移学习的钢桥病害识别[J]. 长安大学学报(自然科学版)202141(3): 52-63.
[63]
DUNG C VSEKIYA HHIRANO Set al. A vision-based method for crack detection in gusset plate welded joints of steel bridges using deep convolutional neural networks[J]. Automation in Construction2019102: 217-229.
[64]
YU Q QWANG JGU X Let al. An attention-based detection method of fatigue cracks on steel[J]. Structural Control and Health Monitoring20252025(1): 7487687.
[65]
DAI ACHANG A XSAVVA Met al. Scannet: Richly-annotated 3d reconstructions of indoor scenes[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: 2017: 5828-5839.
[66]
HAN Q HLIU XXU J. Detection and location of steel structure surface cracks based on unmanned aerial vehicle images[J]. Journal of Building Engineering202250: 104098.
[67]
MENG SGAO ZZHOU Yet al. Real‐time automatic crack detection method based on drone[J]. Computer‐Aided Civil and Infrastructure Engineering202338(7): 849-872.
[68]
IANDOLA F NHAN SMOSKEWICZ M Wet al. SqueezeNet: alexNet-level accuracy with 50x fewer parameters and <0.5 MB model size[PP/OL]. V4. arXiv (2016-11-04) [2026-05-07]. https://doi.org/10.48550/arXiv.1602.07360.
[69]
HOWARD A GZHU MCHEN Bet al. MobileNets: efficient convolutional neural networks for mobile vision applications[PP/OL]. V1. ArXiv (2017-04-17) [2026-05-07]. https://doi.org/10.48550/arXiv.1704.04861.
[70]
ZHANG XZHOU XLIN Met al. ShuffleNet: an extremely efficient convolutional neural network for mobile devices[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Long Beach: 2018: 6848-6856.
[71]
TAN MLE Q V. EfficientNet: rethinking model scaling for convolutional neural networks[C]//Proceedings of the 36th International Conference on Machine Learning. 2019: 6105-6114.
[72]
WANG YWANG HXIN Z. Efficient detection model of steel strip surface defects based on YOLO-V7[J]. IEEE Access202210: 133936-133944.
[73]
LU NWANG KWANG Het al. Real-time fatigue crack detection and prediction in steel structures based on an automated digital twin-driven framework[J]. Thin-Walled Structures2025218:114099.
[74]
GUO ZZHANG YZHU Q. CSCP-YOLO: a lightweight and efficient algorithm for real-time steel surface defect detection[J]. Advanced Engineering Informatics202254: 101741.
[75]
LIU RZENG W. Automatic detection of structural defects in tunnel lining via network pruning and knowledge distillation in YOLO[J]. Structural Health Monitoring202625(2): 1165-1181.
[76]
HUANG HCAI YZHANG Cet al. Crack detection of masonry structure based on thermal and visible image fusion and semantic segmentation[J]. Automation in Construction2024158: 105213.
[77]
ALEXANDER Q GHOSKERE VNARAZAKI Yet al. Fusion of thermal and RGB images for automated deep learning based crack detection in civil infrastructure[J]. AI in Civil Engineering20221(1): 3.
[78]
WANG PXIAO JQIANG Xet al. An automatic building facade deterioration detection system using infrared-visible image fusion and deep learning[J]. Journal of Building Engineering202495: 110122.
[79]
ZHOU SSONG W. Deep learning-based roadway crack classification with heterogeneous image data fusion[J]. Structural Health Monitoring202120(3): 1274-1293.
[80]
PARK S EEEM S HJEON H. Concrete crack detection and quantification using deep learning and structured light[J]. Construction and Building Materials2020252: 119096.
[81]
WANG SZHAO XGAO Let al. Pixel-level crack segmentation and quantification enabled by multi-modality cross-fusion of RGB and depth images[J]. Automation in Construction2024160: 105318.
[82]
XU WCUI CLUO Cet al. Fatigue crack monitoring of steel bridge with coating sensor based on potential difference method[J]. Construction and Building Materials2022350: 128868.
[83]
MANU K CMADHUSHREE CCHANDINI M Set al. Corrosion in steel structures: a review[J]. Journal of Mines, Metals and Fuels202573(1): 189-198.
[84]
ANWAR SLI C. Diving deeper into underwater image enhancement: a survey[J]. Signal Processing: Image Communication202089: 115978.
[85]
DUY L DANH N TSON N Tet al. Deep learning in semantic segmentation of rust in images[C]//Proceedings of the 2020 9th International Conference on Software and Computer Applications. Malaysia: 2020: 262-266.
[86]
DAS A, DORAFSHAN SKAABOUCH N. Autonomous image-based corrosion detection in steel structures using deep learning[J]. Sensors202424(11): 3630.
[87]
中国国家标准化管理委员会. 金属基体上金属和其他无机覆盖层 经腐蚀试验后的试样和试件的评级:GB/T 6461—2002[S]. 北京: 中国标准出版社, 2002.
[88]
ASTM International. Standard test method for evaluating degree of rusting on painted steel surfaces:ASTM D610-23[S]. West Conshohocken, PA: ASTM International, 2023.
[89]
International Organization for Standardization. Paints and varnishes-evaluation of degradation of coatings-part 3: assessment of degree of rusting:ISO 4628-3∶2016 [S]. Geneva, Switzerland: ISO, 2016.
[90]
陆廷杰, 刘东海, 齐志龙. 基于深度学习的水下钢结构锈蚀识别与评价[J]. 天津大学学报(自然科学与工程技术版)202356(7): 713-722.
[91]
PETRICCA LMOSS TFIGUEROA Get al. Corrosion detection using AI: a comparison of standard computer vision techniques and deep learning model[C]//Proceedings of the Sixth International Conference on Computer Science, Engineering and Information Technology. Chennai: 2016: 91-99.
[92]
ATHA D JJAHANSHAHI M R. Evaluation of deep learning approaches based on convolutional neural networks for corrosion detection[J]. Structural Health Monitoring201817(5): 1110-1128.
[93]
HOLM ETRANSETH A AKNUDSEN O Øet al. Classification of corrosion and coating damages on bridge constructions from images using convolutional neural networks[C]//Twelfth International Conference on Machine Vision (ICMV 2019). Bellingham: 2020: 549-556.
[94]
BASTIAN B TJASPREETH NRANJITH S Ket al. Visual inspection and characterization of external corrosion in pipelines using deep neural network[J]. NDT & E International2019107: 102134.
[95]
FORKAN A R MKANG Y BJAYARAMAN P Pet al. CorrDetector: a framework for structural corrosion detection from drone images using ensemble deep learning[J]. Expert Systems with Applications2022193: 116461.
[96]
XU JGUI CHAN Q. Recognition of rust grade and rust ratio of steel structures based on ensembled convolutional neural network[J]. Computer-Aided Civil and Infrastructure Engineering202035(10): 1160-1174.
[97]
JIN LIM HHWANG SKIM Het al. Steel bridge corrosion inspection with combined vision and thermographic images[J]. Structural Health Monitoring202120(6): 3424-3435.
[98]
ANDERSEN RNALPANTIDIS LRAVN Oet al. Investigating deep learning architectures towards autonomous inspection for marine classification[C]//2020 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR). Abu Dhabi: 2020: 197-204.
[99]
ZHOU QDING SFENG Yet al. Corrosion inspection and evaluation of crane metal structure based on UAV vision[J]. Signal, Image and Video Processing202216(6): 1701-1709.
[100]
JIA ZFU MZHAO Xet al. Intelligent identification of metal corrosion based on Corrosion-YOLOv5s[J]. Displays202376: 102367.
[101]
NABIZADEH EPARGHI A. Automated corrosion detection using deep learning and computer vision[J]. Asian Journal of Civil Engineering202324(8): 2911-2923.
[102]
AMELI ZNESHELI S JLANDIS E N. Deep learning-based steel bridge corrosion segmentation and condition rating using mask RCNN and YOLOv8[J]. Infrastructures20239(1): 1-16.
[103]
YU Q FHAN Y DLIN W Get al. Detection and analysis of corrosion on coated metal surfaces using enhanced YOLOv5 algorithm for anti-corrosion performance evaluation[J]. Journal of Marine Science and Engineering202412(7): 1090.
[104]
PIRIE CMORENO-GARCIA C F. Image pre-processing and segmentation for real-time subsea corrosion inspection[C]//International Conference on Engineering Applications of Neural Networks. Cham: 2021: 220-231.
[105]
王达磊, 彭博, 潘玥, . 基于深度神经网络的锈蚀图像分割与定量分析[J]. 华南理工大学学报(自然科学版)201846(12): 121-127.
[106]
FONDEVIK S KSTAHL ATRANSETH A Aet al. Image segmentation of corrosion damages in industrial inspections[C]//2020 IEEE 32nd International Conference on Tools with Artificial Intelligence (ICTAI). Baltimore: 2020: 787-792.
[107]
KATSAMENIS IPROTOPAPADAKIS EDOULAMIS Aet al. Pixel-level corrosion detection on metal constructions by fusion of deep learning semantic and contour segmentation[C]//International Symposium on Visual Computing. Cham: 2020: 160-169.
[108]
AKHLAGHI BMESGHALI HEHTESHAMI Met al. Predictive deep learning for pitting corrosion modeling in buried transmission pipelines[J]. Process Safety and Environmental Protection2023174: 320-327.
[109]
GAO R PSHANG W JZHAO Yet al. Research on fusion model method for corrosion damage detection of switch sliding baseplate[J]. Coatings202414(12): 1552.
[110]
ZHANG Z WLI S LWANG H Jet al. A study of neural network-based evaluation methods for pipelines with multiple corrosive regions[J]. Reliability Engineering & System Safety2025253: 110507.
[111]
秦荣杰.基于深度学习Transformer网络的钢板锈蚀类别识别方法研究[D].西安:西安建筑科技大学,2023.
[112]
ALTABEY W ANOORI MWANG Tet al. Deep learning-based crack identification for steel pipelines by extracting features from 3D shadow modeling[J]. Applied Sciences202111(13): 6063.
[113]
WANG B QLIU L ACHENG X Qet al. Advanced multi-image segmentation-based machine learning modeling strategy for corrosion prediction and rust layer performance evaluation of weathering steel[J]. Corrosion Science2024237: 112334.
[114]
LI Y PLI H GGUAN Yet al. Dense metal corrosion depth estimation[J]. Frontiers in Physics202311: 1277710.
[115]
SON E YJEONG DOH M J. Corrosion area detection and depth prediction using machine learning[J]. International Journal of Naval Architecture and Ocean Engineering202416: 100617.
[116]
ARIAS FGUEVARA EJARAMILLO Eet al. Automated assessment of marine steel corrosion using visible-near-infrared hyperspectral imaging[J]. Coatings202515(6):645.
[117]
EGODAWELA SGOSTAR A KBUDDIKA H A D Set al. Metal loss defect detection and depth estimation using multi-spectral image analysis of cooling excited steel specimen with corrosion[J]. Scientific Reports202515(1): 23894.
[118]
LIU LTAN EZHEN Yet al. AI-facilitated coating corrosion assessment system for productivity enhancement[C]//2018 13th IEEE Conference on Industrial Electronics and Applications (ICIEA). Wuhan: 2018: 606-610.
[119]
YAO YYANG YWANG Yet al. Artificial intelligence-based hull structural plate corrosion damage detection and recognition using convolutional neural network[J]. Applied Ocean Research201990: 101823.
[120]
LUO CYU LYAN Jet al. Autonomous detection of damage to multiple steel surfaces from 360 panoramas using deep neural networks[J]. Computer‐Aided Civil and Infrastructure Engineering202136(12): 1585-1599.
[121]
YU L JYANG E FLUO Cet al. AMCD: an accurate deep learning-based metallic corrosion detector for MAV-based real-time visual inspection[J]. Journal of Ambient Intelligence and Humanized Computing202314:8087-8098.
[122]
HUANG MZHANG JLI Jet al. Damage identification of steel bridge based on data augmentation and adaptive optimization neural network[J]. Structural Health Monitoring202424: 1674-1699.
[123]
BIANCHI EHEBDON M. Visual structural inspection datasets[J]. Automation in Construction2022139: 104299.
[124]
梁俊杰, 韦舰晶, 蒋正锋. 生成对抗网络GAN综述[J]. 计算机科学与探索202014(1): 1-17.
[125]
RADFORD AMETZ LCHINTALA S. Unsupervised representation learning with deep convolutional generative adversarial networks[PP/OL]. V2. arXiv (2016-01-07) [2026-05-07] .https://doi.org/10.48550/arXiv.1511.06434.
[126]
GULRAJANI IAHMED FARJOVSKY Met al. Improved training of Wasserstein GANs[C]// Neural Information Processing Systems. Long Beach: 2017: 5769-5779.
[127]
LEI XSUN LXIA Y. Lost data reconstruction for structural health monitoring using deep convolutional generative adversarial networks[J]. Structural Health Monitoring202220(4): 2069-2087.
[128]
GAO YKONG BMOSALAM K M. Deep leaf-bootstrapping generative adversarial network for structural image data augmentation[J]. Computer-Aided Civil and Infrastructure Engineering201934(9): 755-773.
[129]
MAEDA HKASHIYAMA TSEKIMOTO Yet al. Generative adversarial network for road damage detection[J]. Computer-Aided Civil and Infrastructure Engineering202136(1): 47-60.
[130]
XU Y ZWU HLIU Y Let al. Automated surface defect detection based on CycleGAN model[C]//Journal of Physics: Conference Series. Bristol: 20242890: 012036.
[131]
WANG YLIAO XCUI Wet al. Defending against and generating adversarial examples together with generative adversarial networks[J]. Scientific Reports202515(1): 12994.
[132]
ZHOU H YJIANG FLU H Tet al. SSDA-YOLO: semi-supervised domain adaptive YOLO for cross-domain object detection[J]. Computer Vision and Image Understanding2023229: 103649.
[133]
牟宗涵. 基于无人机图像的铁路桥梁钢结构表面缺陷智能识别方法研究[D]. 北京:北京交通大学, 2023.
[134]
钱企豪, 郑战光, 梁钊, . 基于颜色特征的半监督聚类算法在铜片腐蚀等级识别中的应用[J]. 腐蚀与防护202344(5): 34-40.
[135]
FENG J JTIAN L FLI X Xet al. Adaptive adversarial self-training for semi-supervised object detection in complex maritime scenes[J]. Mathematics202412(15): 2348.
[136]
WANG S YNGUYEN H DWILSON Ret al. Deep CNN-based semi-supervised learning approach for identifying and segmenting corrosion in hydraulic steel and water resources infrastructure[J]. Structural Health Monitoring202524: 2229-2249.
[137]
GUO M HXU T XLIU J Jet al. Attention mechanisms in computer vision: a survey[J]. Computational Visual Media20228(3): 331-368.
[138]
DUAN ZHUANG X HHOU Jet al. Research on intelligent diagnosis of corrosion in the operation and maintenance stage of steel structure engineering based on U-Net attention[J]. Buildings202414(12): 3972.
[139]
KATSAMENIS IDOULAMIS NDOULAMIS Aet al. Simultaneous precise localization and classification of metal rust defects for robotic-driven maintenance and prefabrication using residual attention U-Net[J]. Automation in Construction2022137: 104182.
[140]
MA S BZHAO XWAN Let al. A lightweight algorithm for steel surface defect detection using improved YOLOv8[J]. Scientific Reports202515: 93469.
[141]
FU M JJIA Z TWU L Zet al. Detection and recognition of metal surface corrosion based on CBG-YOLOv5s[J]. PloS One, San Francisco, 202419(4): 1932-6203.
[142]
ZHANG G HLIU S XNIE S Qet al. YOLO-RDP: lightweight steel defect detection through improved YOLOv7-tiny and model pruning[J]. Symmetry202416(4): 1-12.
[143]
TAN LCHEN X HYUAN D Jet al. DSNet: a Computer vision-based detection and corrosion segmentation network for corroded bolt detection in tunnel[J]. Structural Control and Health Monitoring2024, 2024(1): 1-16.
[144]
YU V FSANTIYUDA GLIN S Wet al. Neural network pruning for lightweight metal corrosion image segmentation models[J]. IEEE Access202513: 71673-71687.
[145]
HAN Q HZHAO NXU J. Recognition and location of steel structure surface corrosion based on unmanned aerial vehicle images[J]. Journal of Civil Structural Health Monitoring202111(5): 1375-1392.
[146]
ELTOUNY KSAJEDI SLIANG X. Dmg2Former-AR: vision transformers with adaptive rescaling for high-resolution structural visual inspection[J]. Sensors202424(18): 6007.
[147]
YE X WJIN TLI Z Xet al. Structural crack detection from benchmark data sets using pruned fully convolutional networks[J]. Journal of Structural Engineering2021147(11): 04721008.
[148]
BIANCHI EHEBDON M. Corrosion condition state semantic segmentation dataset[DS]. Blacksburg, VA, USA: Virginia Tech, 2021.
[149]
HOSKERE VNARAZAKI YHOANG T Aet al. MaDnet: multi-task semantic segmentation of multiple types of structural materials and damage in images of civil infrastructure[J]. Journal of Civil Structural Health Monitoring202010(5): 757-773.
[150]
DONG HSONG KHE Jet al. PGA-Net: pyramid feature fusion and global context attention network for automated surface defect detection[J]. IEEE Transactions on Industrial Informatics202016(12): 7448-7458.
[151]
GAN JLI QWANG Jet al. A hierarchical extractor-based visual rail surface inspection system[J]. IEEE Sensors Journal201717(23): 7935-7944.
[152]
SONG K CHU S PYAN Y Het al. Surface defect detection method using saliency linear scanning morphology for silicon steel strip under oil pollution interference[J]. ISIJ International201454(11): 2598-2607.
[153]
SONG KYAN Y H. Micro surface defect detection method for silicon steel strip based on saliency convex active contour model, mathematical problems in engineering[J]. Mathematical Problems in Engineering2013(1): 429094.
[154]
ZHANG SZHANG QGU Jet al. Visual inspection of steel surface defects based on domain adaptation and adaptive convolutional neural network[J]. Mechanical Systems and Signal Processing2021153: 107541.
[155]
HU XYANG JJIANG Fet al. Steel surface defect detection based on self-supervised contrastive representation learning with matching metric[J]. Applied Soft Computing2023145: 110578.
[156]
YANG X CA1, FAN Y L, BAO Y Q,et al. Task-aware meta-learning paradigm for universal structural damage segmentation using limited images[J]. Engineering Structures, 2023, 284: 115917.
[157]
LI YCHEN ZZHA Det al. Automated anomaly detection via curiosity-guided search and self-imitation learning[J]. IEEE Transactions on Neural Networks and Learning Systems202233(6): 2365-2377.
[158]
RIPPEL OMERTENS PMERHOF D. Modeling the distribution of normal data in pre-trained deep features for anomaly detection[C]//2020 25th International Conference on Pattern Recognition (ICPR). Milan: 2021: 6726-6733.
2026年第56卷第5期
PDF下载
179
76
引用本文
BibTeX
文章信息
doi: 10.3724/j.gyjzG26033109
  • 接收时间:2026-03-31
  • 首发时间:2026-06-25
  • 出版时间:2026-05-20
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2026-03-31
基金
作者信息
    1同济大学建筑工程系,上海200092
    2同济大学工程结构性能演化与控制教育部重点实验室,上海200092
    3同济大学;土木工程防灾减灾全国重点实验室,上海200092

通讯作者:

余倩倩,博士,教授,主要从事结构性能演化与控制方向研究,
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/gyjz/CN/10.3724/j.gyjzG26033109
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数

Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)

Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏