Article(id=1322960951045137026, tenantId=1146029695717560320, journalId=1249024232475115590, issueId=1322960856543273127, articleNumber=null, orderNo=null, doi=10.11834/jig.250458, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1760198400000, receivedDateStr=2025-10-12, revisedDate=1764086400000, revisedDateStr=2025-11-26, acceptedDate=null, acceptedDateStr=null, onlineDate=1793348070155, onlineDateStr=2026-10-30, pubDate=1786809600000, pubDateStr=2026-08-16, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1793348070155, onlineIssueDateStr=2026-10-30, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1793348070155, creator=13701087609, updateTime=1793348070155, updator=13701087609, issue=Issue{id=1322960856543273127, tenantId=1146029695717560320, journalId=1249024232475115590, year='2026', volume='31', issue='8', pageStart='2687', pageEnd='3074', issueExtLink='null', onlineDate='null', pubDate='1786809600000', pubDateStr='2026-08-16', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1793348047624, creator='13701087609', updateTime=1793348605344, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext={EN=IssueExt(id=1322963195874378645, tenantId=1146029695717560320, journalId=1249024232475115590, issueId=1322960856543273127, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1322963195874378646, tenantId=1146029695717560320, journalId=1249024232475115590, issueId=1322960856543273127, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null, downloadFileDto=null}, startPage=2844, endPage=2867, ext={EN=ArticleExt(id=1322960952878047875, articleId=1322960951045137026, tenantId=1146029695717560320, journalId=1249024232475115590, language=EN, title=Review of damage detection methods for wind turbine blades, columnId=1249044010699002654, journalTitle=Journal of Image and Graphics, columnName=Review, runingTitle=null, highlight=null, articleAbstract=

With the accelerating transition of the global energy landscape toward renewables,wind power has been widely promoted as a vital component of clean energy. Alongside the rapid expansion of wind turbine installed capacity,the structural health of wind turbine blades (WTBs),which are among the critical power-generating components,directly influences the safety and efficiency of wind power systems. In recent years,unmanned aerial vehicles (UAVs) have been equipped with high-resolution imaging devices for blade image acquisition. By combining advanced image processing and object detection techniques,they have achieved automated,noncontact defect detection,which not only enhances detection efficiency and coverage but also reduces the risks and costs of traditional manual inspections. Therefore,this approach has become a mainstream methodology in the industry and a focal point of research. However,images captured by UAVs often suffer from poor quality (primarily manifested as noise,motion blur,and low resolution),posing critical challenges to defect detection. Moreover,defect features,such as erosion marks,stains,or manufacturing traces,frequently exhibit high visual similarity to blade surface textures and present multiscale and diverse morphological characteristics. This visual ambiguity further complicates accurate localization and classification. In addition,the lack of standardized annotation protocols and the heavy reliance on manual labeling hinder the construction of high-quality datasets. These issues collectively constrain the detection accuracy and generalization performance of deep learning-based object detection models. Existing survey literature on WTB defect detection has primarily emphasized nondestructive testing methods,with a noticeable scarcity of comprehensive reviews dedicated exclusively to image-based detection approaches. Current surveys in this field focus `on architectural advancements within deep learning models,including the integration of lightweight modules,attention mechanisms,and multiscale feature extraction techniques. Although these architectural improvements have enhanced detection accuracy,they often overlook essential aspects,including image preprocessing methods and learning strategies,both of which play a crucial role in determining overall model performance. In this review,we systematically incorporate perspectives on image preprocessing and learning strategies to analyze defect detection methods for WTBs. The main contributions of this review are as follows. 1) From the perspective of defect manifestation in WTBs,defects can be broadly categorized into internal and surface types. We systematically present and analyze the causes,formation mechanisms,and visual characteristics of these defect categories with respect to their structural properties. We also summarize the most frequently observed defect types in datasets. These datasets provide valuable data support for the design and optimization of subsequent defect detection models. 2) This review summarizes the characteristics of defects in detection tasks across various power system components,such as foreign object recognition on transmission lines and insulator defect detection. Through a comparative analysis,this review demonstrates that WTB defect detection uniquely encompasses all the aforementioned challenges and variabilities. Consequently,WTB defect detection necessitates simultaneously addressing a spectrum of challenges,including poor image quality,multiscale defect manifestations,class imbalance,lack of standardized annotation protocols,and the need for lightweight deployable models. Such a problem-driven perspective provides a foundation for inspiring future methodological advancements. 3) We conduct an in-depth investigation into the key components of deep learning-based image-driven WTB defect detection. Three critical aspects,namely,image preprocessing,model architecture,and dominant learning strategies,are presented. With respect to image preprocessing,task-driven methodologies are categorized into image quality enhancement,geometric correction,and structural feature extraction. Each of them is examined for its critical role in improving the saliency of defects and bolstering the robustness of detection models. At the architectural level,the enhancement strategies across mainstream detection frameworks,including two-stage,one-stage,and Transformer-based models,are presented. In terms of learning strategies,structured analyses of the practical effectiveness and potential of approaches,including transfer learning,ensemble learning,cascade strategies,and unsupervised/semisupervised learning,are presented. These strategies are evaluated in terms of their capability to address limited data availability,class imbalance,and label uncertainty in defect detections. 4) This review also provides an overview of publicly available datasets released in recent years,examining their data types,annotation protocols,and defect categories,thereby offering guidance for researchers in selecting appropriate research sources. A summary of commonly used evaluation metrics is provided,underscoring the necessity of image quality assessment in diverse application scenarios. 5) This review outlines critical challenges in image-based WTB defect detection,including imbalanced defect distributions,heavy reliance on manual annotations,limited model generalization under environmental variations,insufficient data modality diversity,and lack of innovation in model architectures. In light of these challenges,several promising research directions are proposed; these include the effective integration of multimodal data,creative design of model structures,and in-depth adoption of weakly or self-supervised learning strategies,with the aim of inspiring and guiding future advancements in WTB defect detection.

, authors=Jinghong Zhang1, 2, Pan Su2, 3, *, Qingyuan Zhu2, 4, Chaogang Zhang2, 4, Bing Li1, 2, Wangbin Cao1, 5, authorsList=Jinghong Zhang, Pan Su, Qingyuan Zhu, Chaogang Zhang, Bing Li, Wangbin Cao, authorCompany=null, correspAuthors=Pan Su, 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=1322960962004853393, articleId=1322960951045137026, tenantId=1146029695717560320, journalId=1249024232475115590, language=CN, title=风力发电机叶片缺陷目标检测研究综述, columnId=1249044010992603937, journalTitle=中国图象图形学报, columnName=综述, runingTitle=null, highlight=null, articleAbstract=

全球能源转型推动了风力发电机的大规模部署,基于无人机图像采集的风机叶片(wind turbine black,WTB)缺陷目标检测技术已成为研究热点。风力发电机叶片的无人机图像存在质量低、缺陷与背景纹理相似、缺陷特征多尺度等问题,且标注缺乏统一标准并高度依赖人工,这些因素显著增加了缺陷检测的难度。当前关于风力发电机叶片缺陷检测的综述主要聚焦于无损检测方法,针对图像检测技术的系统性综述仍然不足,且多集中于模型架构的优化,而忽略了图像预处理和学习策略对检测性能的重要影响。因此,本文将图像预处理和学习策略纳入视角,系统综述了风力发电机叶片缺陷目标检测的研究进展:1)详细列举叶片缺陷类型及特征。2)与电力系统其他检测任务进行对比研究并总结主要问题。3)从图像预处理、网络改进和学习策略3个维度总结现有风力发电机叶片缺陷目标检测方法。在图像预处理方面,基于任务导向将现有方法归纳为3类:图像质量增强、几何校正以及结构特征提取;在网络优化方面,沿目标检测模型发展脉络,梳理了各类架构的改进策略;最后,总结了该领域常用的学习策略。4)对近年来数据集与常用评估指标进行汇总。5)对风机叶片缺陷目标检测方法面临的挑战及未来研究方向进行总结和展望。

, authors=张弪弘1, 2, 苏攀2, 3, *, 朱清源2, 4, 张超刚2, 4, 李冰1, 2, 曹旺斌1, 5, authorsList=张弪弘, 苏攀, 朱清源, 张超刚, 李冰, 曹旺斌, authorCompany=null, correspAuthors=苏攀, authorNote=

张弪弘,女,硕士研究生,主要研究方向为新能源与目标检测。E-mail:

苏攀,通信作者,男,副教授,主要研究方向为机器学习与医学图像处理。E-mail:

朱清源,男,硕士研究生,主要研究方向为人工智能与新能源。E-mail:

张超刚,男,硕士研究生,主要研究方向为基于轻量化模型的风机叶片缺陷检测。E-mail:

李冰,男,副教授,主要研究方向为机器学习与计算机视觉。E-mail:

曹旺斌,男,副教授,主要研究方向为人工智能与新能源。E-mail:

, correspAuthorsNote=
苏攀,通信作者,男,副教授,主要研究方向为机器学习与医学图像处理。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=EUVhCEh2u25rkr2PGybBBA==, magXml=pgWY6yfiDAUuw5zToucS+g==, pdfUrl=null, pdf=h6QGOYBB73NwtsoyMEiqww==, pdfFileSize=2334300, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=76Xkm2q+QfLthlax0eB85g==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=T4qE9IIZJCZwjvBLB0yGdg==, mapNumber=null, fund=null)}, authors=[Author(id=1322960962684330658, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=17395714117@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1322960962772411045, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960962684330658, language=EN, stringName=Jinghong Zhang, firstName=Jinghong, middleName=null, lastName=Zhang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 Yanzhao Electric Power Laboratory of North China Electric Power University,Baoding 071003,China
2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1322960962852102822, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960962684330658, language=CN, stringName=张弪弘, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 华北电力大学燕赵电力实验室,保定 071003
2 华北电力大学控制与计算机工程学院,保定 071003, bio={"content":"

张弪弘,女,硕士研究生,主要研究方向为新能源与目标检测。E-mail:

"}, bioImg=null, bioContent=

张弪弘,女,硕士研究生,主要研究方向为新能源与目标检测。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1322960962290066066, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=1, ext=[AuthorCompanyExt(id=1322960962298454675, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962290066066, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Yanzhao Electric Power Laboratory of North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962306843284, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962290066066, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 华北电力大学燕赵电力实验室,保定 071003)]), AuthorCompany(id=1322960962361369237, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=2, ext=[AuthorCompanyExt(id=1322960962369757846, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962378146455, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 华北电力大学控制与计算机工程学院,保定 071003)])]), Author(id=1322960962910823080, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=supan@ncepu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1322960962986320555, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960962910823080, language=EN, stringName=Pan Su, firstName=Pan, middleName=null, lastName=Su, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, 3, *, address=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China
3 Hebei Key Laboratory of Knowledge Computing for Energy and Power,Baoding 071003,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1322960963066012332, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960962910823080, language=CN, stringName=苏攀, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, 3, *, address=2 华北电力大学控制与计算机工程学院,保定 071003
3 河北省能源电力知识计算重点实验室,保定 071003, bio={"content":"

苏攀,通信作者,男,副教授,主要研究方向为机器学习与医学图像处理。E-mail:

"}, bioImg=null, bioContent=

苏攀,通信作者,男,副教授,主要研究方向为机器学习与医学图像处理。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1322960962361369237, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=2, ext=[AuthorCompanyExt(id=1322960962369757846, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962378146455, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 华北电力大学控制与计算机工程学院,保定 071003)]), AuthorCompany(id=1322960962445255320, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=3, ext=[AuthorCompanyExt(id=1322960962453643929, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962445255320, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 Hebei Key Laboratory of Knowledge Computing for Energy and Power,Baoding 071003,China), AuthorCompanyExt(id=1322960962462032538, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962445255320, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 河北省能源电力知识计算重点实验室,保定 071003)])]), Author(id=1322960963128926894, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=18957139414@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1322960963225395889, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960963128926894, language=EN, stringName=Qingyuan Zhu, firstName=Qingyuan, middleName=null, lastName=Zhu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, 4, address=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China
4 Engineering Research Center of Intelligent Computing for Complex Energy Systems,Ministry of Education,Baoding 071003,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1322960963296699058, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960963128926894, language=CN, stringName=朱清源, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, 4, address=2 华北电力大学控制与计算机工程学院,保定 071003
4 复杂能源系统智能计算教育部工程研究中心,保定 071003, bio={"content":"

朱清源,男,硕士研究生,主要研究方向为人工智能与新能源。E-mail:

"}, bioImg=null, bioContent=

朱清源,男,硕士研究生,主要研究方向为人工智能与新能源。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1322960962361369237, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=2, ext=[AuthorCompanyExt(id=1322960962369757846, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962378146455, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 华北电力大学控制与计算机工程学院,保定 071003)]), AuthorCompany(id=1322960962520752795, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=4, ext=[AuthorCompanyExt(id=1322960962529141404, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962520752795, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4 Engineering Research Center of Intelligent Computing for Complex Energy Systems,Ministry of Education,Baoding 071003,China), AuthorCompanyExt(id=1322960962537530013, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962520752795, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4 复杂能源系统智能计算教育部工程研究中心,保定 071003)])]), Author(id=1322960963388973748, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=18231204373@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1322960963456082615, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960963388973748, language=EN, stringName=Chaogang Zhang, firstName=Chaogang, middleName=null, lastName=Zhang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, 4, address=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China
4 Engineering Research Center of Intelligent Computing for Complex Energy Systems,Ministry of Education,Baoding 071003,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1322960963506414264, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960963388973748, language=CN, stringName=张超刚, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, 4, address=2 华北电力大学控制与计算机工程学院,保定 071003
4 复杂能源系统智能计算教育部工程研究中心,保定 071003, bio={"content":"

张超刚,男,硕士研究生,主要研究方向为基于轻量化模型的风机叶片缺陷检测。E-mail:

"}, bioImg=null, bioContent=

张超刚,男,硕士研究生,主要研究方向为基于轻量化模型的风机叶片缺陷检测。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1322960962361369237, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=2, ext=[AuthorCompanyExt(id=1322960962369757846, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962378146455, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 华北电力大学控制与计算机工程学院,保定 071003)]), AuthorCompany(id=1322960962520752795, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=4, ext=[AuthorCompanyExt(id=1322960962529141404, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962520752795, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4 Engineering Research Center of Intelligent Computing for Complex Energy Systems,Ministry of Education,Baoding 071003,China), AuthorCompanyExt(id=1322960962537530013, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962520752795, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4 复杂能源系统智能计算教育部工程研究中心,保定 071003)])]), Author(id=1322960963560940218, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=li_bing_hb@126.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1322960963653214909, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960963560940218, language=EN, stringName=Bing Li, firstName=Bing, middleName=null, lastName=Li, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 Yanzhao Electric Power Laboratory of North China Electric Power University,Baoding 071003,China
2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1322960963716129470, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960963560940218, language=CN, stringName=李冰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 华北电力大学燕赵电力实验室,保定 071003
2 华北电力大学控制与计算机工程学院,保定 071003, bio={"content":"

李冰,男,副教授,主要研究方向为机器学习与计算机视觉。E-mail:

"}, bioImg=null, bioContent=

李冰,男,副教授,主要研究方向为机器学习与计算机视觉。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1322960962290066066, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=1, ext=[AuthorCompanyExt(id=1322960962298454675, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962290066066, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Yanzhao Electric Power Laboratory of North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962306843284, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962290066066, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 华北电力大学燕赵电力实验室,保定 071003)]), AuthorCompany(id=1322960962361369237, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=2, ext=[AuthorCompanyExt(id=1322960962369757846, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962378146455, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 华北电力大学控制与计算机工程学院,保定 071003)])]), Author(id=1322960963812598464, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, orderNo=5, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=wbin.cao@ncepu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1322960963892290243, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960963812598464, language=EN, stringName=Wangbin Cao, firstName=Wangbin, middleName=null, lastName=Cao, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 5, address=1 Yanzhao Electric Power Laboratory of North China Electric Power University,Baoding 071003,China
5 Hebei Key Laboratory of Power Internet of Things Technology,North China Electric Power University,Baoding 071003,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1322960963946816196, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960963812598464, language=CN, stringName=曹旺斌, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 5, address=1 华北电力大学燕赵电力实验室,保定 071003
5 河北省电力物联网技术重点实验室,保定 071003, bio={"content":"

曹旺斌,男,副教授,主要研究方向为人工智能与新能源。E-mail:

"}, bioImg=null, bioContent=

曹旺斌,男,副教授,主要研究方向为人工智能与新能源。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1322960962290066066, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=1, ext=[AuthorCompanyExt(id=1322960962298454675, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962290066066, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Yanzhao Electric Power Laboratory of North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962306843284, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962290066066, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 华北电力大学燕赵电力实验室,保定 071003)]), AuthorCompany(id=1322960962596250270, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=5, ext=[AuthorCompanyExt(id=1322960962604638879, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962596250270, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=5 Hebei Key Laboratory of Power Internet of Things Technology,North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962608833184, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962596250270, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=5 河北省电力物联网技术重点实验室,保定 071003)])])], keywords=[Keyword(id=1322960964043285189, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, orderNo=1, keyword=wind turbine blade (WTB) defect detection), Keyword(id=1322960964106199750, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, orderNo=2, keyword=UAV imagery), Keyword(id=1322960964202668743, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, orderNo=3, keyword=image preprocessing), Keyword(id=1322960964265583304, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, orderNo=4, keyword=object detection), Keyword(id=1322960964328497865, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, orderNo=5, keyword=learning strategies), Keyword(id=1322960964391412426, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, orderNo=1, keyword=风机叶片缺陷检测), Keyword(id=1322960966069134027, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, orderNo=2, keyword=无人机图像), Keyword(id=1322960966157214412, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, orderNo=3, keyword=图像预处理), Keyword(id=1322960966224323277, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, orderNo=4, keyword=目标检测), Keyword(id=1322960966291432142, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, orderNo=5, keyword=学习策略)], refs=[Reference(id=1322960968858346221, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024b, volume=17, issue=5, pageStart=null, pageEnd=982, url=null, language=null, rfNumber=null, rfOrder=0, authorNames=Altice B, Nazario E, Davis M, Shekaramiz M, Moon T K, Masoum M A S, journalName=Energies, refType=null, unstructuredReference=Altice B,Nazario E,Davis M,Shekaramiz M,Moon T K and Masoum M A S. 2024b. Anomaly detection on small wind turbine blades using deep learning algorithms. Energies,17(5): #982 [DOI: 10.3390/en17050982], articleTitle=Anomaly detection on small wind turbine blades using deep learning algorithms, refAbstract=null), Reference(id=1322960970531873518, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=459, pageEnd=468, url=null, language=null, rfNumber=null, rfOrder=1, authorNames=Alvela Nieto M T, Gelbhardt H, Ohlendorf J H, Thoben K D, journalName=null, refType=null, unstructuredReference=Alvela Nieto M T,Gelbhardt H,Ohlendorf J H and Thoben K D. 2023. Detecting ice on wind turbine rotor blades: towards deep transfer learning for image data//Proceedings of the 6th International Conference on System-Integrated Intelligence. Genova,Italy: Springer: 459-468 [DOI: 10.1007/978-3-031-16281-7_54], articleTitle=Detecting ice on wind turbine rotor blades: towards deep transfer learning for image data, refAbstract=null), Reference(id=1322960970615759599, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=https://arxiv.org/pdf/2310.04719.pdf, language=null, rfNumber=null, rfOrder=2, authorNames=Amrollahi Biyouki S and Hwangbo H, journalName=null, refType=null, unstructuredReference=Amrollahi Biyouki S and Hwangbo H. 2023. A comprehensive survey on deep neural image deblurring [EB/OL]. [2025-10-12]., articleTitle=A comprehensive survey on deep neural image deblurring, refAbstract=null), Reference(id=1322960970699645680, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=null, pageEnd=null, url=https://arxiv.org/pdf /2112.00556. pdf, language=null, rfNumber=null, rfOrder=3, authorNames=Barker J W, Bhowmik N, Breckon T P, journalName=null, refType=null, unstructuredReference=Barker J W,Bhowmik N and Breckon T P. 2021. Semi-supervised surface anomaly detection of composite wind turbine blades from drone imagery [EB/OL]. [2025-10-12]., articleTitle=Semi-supervised surface anomaly detection of composite wind turbine blades from drone imagery, refAbstract=null), Reference(id=1322960970758365937, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2025, volume=25, issue=2, pageStart=725, pageEnd=740, url=null, language=null, rfNumber=null, rfOrder=4, authorNames=Cao J F, Peng B, Gao M Z, Hao H C, Guo J H, Liu X L, journalName=Journal of Failure Analysis and Prevention, refType=null, unstructuredReference=Cao J F,Peng B,Gao M Z,Hao H C,Guo J H,Liu X L,et al. 2025. Detecting small damage on wind turbine surfaces using an improved YOLO in drone-captured scenes. Journal of Failure Analysis and Prevention,25(2): 725-740 [DOI: 10.1007/s11668-025-02130-w], articleTitle=Detecting small damage on wind turbine surfaces using an improved YOLO in drone-captured scenes, refAbstract=null), Reference(id=1322960970833863410, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=15, issue=12, pageStart=262, pageEnd=270, url=null, language=null, rfNumber=null, rfOrder=5, authorNames=Cao Z, He R D, Zhang S F, Qi Z T, Li S, Liu T, journalName=International Journal of Advanced Computer Science and Applications, refType=null, unstructuredReference=Cao Z,He R D,Zhang S F,Qi Z T,Li S,Liu T,et al. 2024. Integrating local channel attention and focused feature modulation for wind turbine blade defect detection. International Journal of Advanced Computer Science and Applications,15(12): 262-270 [DOI: 10.14569/IJACSA.2024.0151228], articleTitle=Integrating local channel attention and focused feature modulation for wind turbine blade defect detection, refAbstract=null), Reference(id=1322960970905166579, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2018, volume=17, issue=5, pageStart=1225, pageEnd=1244, url=null, language=null, rfNumber=null, rfOrder=6, authorNames=Cawley P, journalName=Structural Health Monitoring, refType=null, unstructuredReference=Cawley P. 2018. Structural health monitoring: closing the gap between research and industrial deployment. Structural Health Monitoring,17(5): 1225-1244 [DOI: 10.1177/1475921717750047], articleTitle=Structural health monitoring: closing the gap between research and industrial deployment, refAbstract=null), Reference(id=1322960970972275444, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2020, volume=942, issue=null, pageStart=null, pageEnd=012001, url=null, language=null, rfNumber=null, rfOrder=7, authorNames=Chen X, Eder M A, journalName=IOP Conference Series: Materials Science and Engineering, refType=null, unstructuredReference=Chen X and Eder M A. 2020. A critical review of damage and failure of composite wind turbine blade structures. IOP Conference Series: Materials Science and Engineering,942: #012001 [DOI: 10.1088/1757-899X/942/1/012001], articleTitle=A critical review of damage and failure of composite wind turbine blade structures, refAbstract=null), Reference(id=1322960971051967221, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=6, issue=null, pageStart=null, pageEnd=200126, url=null, language=null, rfNumber=null, rfOrder=8, authorNames=Dai Z J, journalName=Systems and Soft Computing, refType=null, unstructuredReference=Dai Z J. 2024. Image acquisition technology for unmanned aerial vehicles based on YOLO — Illustrated by the case of wind turbine blade inspection. Systems and Soft Computing,6: #200126 [DOI: 10.1016/j.sasc.2024.200126], articleTitle=Image acquisition technology for unmanned aerial vehicles based on YOLO — Illustrated by the case of wind turbine blade inspection, refAbstract=null), Reference(id=1322960971127464694, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=1, pageEnd=6, url=null, language=null, rfNumber=null, rfOrder=9, authorNames=Dhiman H S, Nizami T K, journalName=null, refType=null, unstructuredReference=Dhiman H S and Nizami T K. 2024. Wind turbine blade erosion detection using visual inspection and transfer learning//Proceedings of 2024 International Conference on Control,Automation and Diagnosis. Paris,France: IEEE: 1-6 [DOI: 10.1109/ICCAD60883.2024.10553885], articleTitle=Wind turbine blade erosion detection using visual inspection and transfer learning, refAbstract=null), Reference(id=1322960971194573559, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=17, issue=5, pageStart=2333, pageEnd=2341, url=null, language=null, rfNumber=null, rfOrder=10, authorNames=Diaz P M, Tittus P, journalName=Signal,Image and Video Processing, refType=null, unstructuredReference=Diaz P M and Tittus P. 2023. Fast detection of wind turbine blade damage using Cascade Mask R-DSCNN-aided drone inspection analysis. Signal,Image and Video Processing,17(5): 2333-2341 [DOI: 10.1007/s11760-022-02450-6], articleTitle=Fast detection of wind turbine blade damage using Cascade Mask R-DSCNN-aided drone inspection analysis, refAbstract=null), Reference(id=1322960971261682424, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=34, issue=2, pageStart=19, pageEnd=27, url=null, language=null, rfNumber=null, rfOrder=11, authorNames=Dong C, Zhang C, Li J, Liu J, journalName=Journal of Computers, refType=null, unstructuredReference=Dong C,Zhang C,Li J and Liu J. 2023. Surface defect recognition of wind turbine blades based on improved YOLOX-X model. Journal of Computers,34(2): 19-27 [DOI: 10.53106/199115992023043402002], articleTitle=Surface defect recognition of wind turbine blades based on improved YOLOX-X model, refAbstract=null), Reference(id=1322960971345568505, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=217, issue=null, pageStart=null, pageEnd=113031, url=null, language=null, rfNumber=null, rfOrder=12, authorNames=Du Y, Wu H K, Garcia Cava D, journalName=Measurement, refType=null, unstructuredReference=Du Y,Wu H K and Garcia Cava D. 2023. A motion-blurred restoration method for surface damage detection of wind turbine blades. Measurement,217: #113031 [DOI: 10.1016/j.measurement.2023.113031], articleTitle=A motion-blurred restoration method for surface damage detection of wind turbine blades, refAbstract=null), Reference(id=1322960971395900154, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2020, volume=141, issue=null, pageStart=null, pageEnd=106445, url=null, language=null, rfNumber=null, rfOrder=13, authorNames=Du Y, Zhou S X, Jing X J, Peng Y P, Wu H K, Kwok N, journalName=Mechanical Systems and Signal Processing, refType=null, unstructuredReference=Du Y,Zhou S X,Jing X J,Peng Y P,Wu H K and Kwok N. 2020. Damage detection techniques for wind turbine blades: a review. Mechanical Systems and Signal Processing,141: #106445 [DOI: 10.1016/j.ymssp.2019.106445], articleTitle=Damage detection techniques for wind turbine blades: a review, refAbstract=null), Reference(id=1322960971483980539, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=131, issue=null, pageStart=null, pageEnd=107836, url=null, language=null, rfNumber=null, rfOrder=14, authorNames=Dwivedi D, Babu K V S M, Yemula P K, Chakraborty P, Pal M, journalName=Engineering Applications of Artificial Intelligence, refType=null, unstructuredReference=Dwivedi D,Babu K V S M,Yemula P K,Chakraborty P and Pal M. 2024. Identification of surface defects on solar PV panels and wind turbine blades using attention based deep learning model. Engineering Applications of Artificial Intelligence,131: #107836 [DOI: 10.1016/j.engappai.2023.107836], articleTitle=Identification of surface defects on solar PV panels and wind turbine blades using attention based deep learning model, refAbstract=null), Reference(id=1322960971567866620, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=12, issue=null, pageStart=135985, pageEnd=135998, url=null, language=null, rfNumber=null, rfOrder=15, authorNames=Fu Z J, Zhang F, Ren X Y, Hao B, Zhang X Y, Yin C L, journalName=IEEE Access, refType=null, unstructuredReference=Fu Z J,Zhang F,Ren X Y,Hao B,Zhang X Y,Yin C L,et al. 2024. LE-YOLO: lightweight and efficient detection model for wind turbine blade defects based on improved YOLO. IEEE Access,12: 135985-135998 [DOI: 10.1109/ACCESS.2024.3463391], articleTitle=LE-YOLO: lightweight and efficient detection model for wind turbine blade defects based on improved YOLO, refAbstract=null), Reference(id=1322960971647558397, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2025, volume=34, issue=2, pageStart=null, pageEnd=023002, url=null, language=null, rfNumber=null, rfOrder=16, authorNames=Gao R X, Jin H Q, Wang T F, Li X Y, Liu Q P, journalName=Journal of Electronic Imaging, refType=null, unstructuredReference=Gao R X,Jin H Q,Wang T F,Li X Y and Liu Q P. 2025. Wind turbine equipment motion blur image restoration algorithm optimized by diffusion model denoising diffusion probabilistic model strategy. Journal of Electronic Imaging,34(2): #023002 [DOI: 10.1117/1.JEI.34.2.023002], articleTitle=Wind turbine equipment motion blur image restoration algorithm optimized by diffusion model denoising diffusion probabilistic model strategy, refAbstract=null), Reference(id=1322960971760804606, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=15, issue=4, pageStart=null, pageEnd=043305, url=null, language=null, rfNumber=null, rfOrder=17, authorNames=Gao R X, Ma Y F, Wang T F, journalName=Journal of Renewable and Sustainable Energy, refType=null, unstructuredReference=Gao R X,Ma Y F and Wang T F. 2023. Early stage damage detection of wind turbine blades based on UAV images and deep learning. Journal of Renewable and Sustainable Energy,15(4): #043305 [DOI: 10.1063/5.0157624], articleTitle=Early stage damage detection of wind turbine blades based on UAV images and deep learning, refAbstract=null), Reference(id=1322960971827913471, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2020, volume=161, issue=null, pageStart=998, pageEnd=1010, url=null, language=null, rfNumber=null, rfOrder=18, authorNames=García M􀅡rquez F P and Peco Chacón A M, journalName=Renewable Energy, refType=null, unstructuredReference=García M􀅡rquez F P and Peco Chacón A M. 2020. A review of non-destructive testing on wind turbines blades. Renewable Energy,161: 998-1010 [DOI: 10.1016/j.renene.2020.07.145], articleTitle=A review of non-destructive testing on wind turbines blades, refAbstract=null), Reference(id=1322960971890828032, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2016, volume=1, issue=null, pageStart=50, pageEnd=57, url=null, language=null, rfNumber=null, rfOrder=19, authorNames=Gholizadeh S, journalName=Procedia Structural Integrity, refType=null, unstructuredReference=Gholizadeh S. 2016. A review of non-destructive testing methods of composite materials. Procedia Structural Integrity,1: 50-57 [DOI: 10.1016/j.prostr.2016.02.008], articleTitle=A review of non-destructive testing methods of composite materials, refAbstract=null), Reference(id=1322960971957936897, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2014, volume=580, issue=587, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=null, rfOrder=20, authorNames=Girshick R, Donahue J, Darrell T, Malik J, journalName=In Proceedings of the IEEE conference on computer vision and pattern recognition, refType=null, unstructuredReference=Girshick R,Donahue J,Darrell T and Malik J,2014. Rich feature hierarchies for accurate object detection and semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition: 580-587 [DOI: 10.1109/CVPR.2014.81], articleTitle=Rich feature hierarchies for accurate object detection and semantic segmentation, refAbstract=null), Reference(id=1322960972037628674, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2025, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=null, rfOrder=21, authorNames=Global Wind Energy Council, journalName=GWEC Global Wind Report 2025, refType=null, unstructuredReference=Global Wind Energy Council. 2025. GWEC Global Wind Report 2025. Brussels: Global Wind Energy Council, articleTitle=null, refAbstract=null), Reference(id=1322960972108931843, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=290, pageEnd=295, url=null, language=null, rfNumber=null, rfOrder=22, authorNames=Gohar I, See J, Halimi A, Yew W K, journalName=null, refType=null, unstructuredReference=Gohar I,See J,Halimi A and Yew W K. 2023. Automatic defect detection in wind turbine blade images: model benchmarks and re-annotations//Proceedings of 2023 IEEE International Conference on Multimedia and Expo Workshops. Brisbane,Australia: IEEE: 290-295 [DOI: 10.1109/ICMEW59549.2023.00056], articleTitle=Automatic defect detection in wind turbine blade images: model benchmarks and re-annotations, refAbstract=null), Reference(id=1322960972176040708, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2019, volume=22, issue=5, pageStart=698, pageEnd=711, url=null, language=null, rfNumber=null, rfOrder=23, authorNames=Gómez Muñoz C Q, García Marquez F P, Hernandez Crespo B, Makaya K, journalName=Wind Energy, refType=null, unstructuredReference=Gómez Muñoz C Q,García Marquez F P,Hernandez Crespo B and Makaya K. 2019. Structural health monitoring for delamination detection and location in wind turbine blades employing guided waves. Wind Energy,22(5): 698-711 [DOI: 10.1002/we.2316], articleTitle=Structural health monitoring for delamination detection and location in wind turbine blades employing guided waves, refAbstract=null), Reference(id=1322960972264121093, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=null, rfOrder=24, authorNames=Gonzalez R C, Woods R E, journalName=Pearson, refType=null, unstructuredReference=Gonzalez R C and Woods R E. 2018 . Digital image processing (4th ed.). Pearson., articleTitle=Digital image processing (4th ed, refAbstract=null), Reference(id=1322960972335424262, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2021, volume=174, issue=null, pageStart=122, pageEnd=133, url=null, language=null, rfNumber=null, rfOrder=25, authorNames=Guo J H, Liu C, Cao J F, Jiang D X, journalName=Renewable Energy, refType=null, unstructuredReference=Guo J H,Liu C,Cao J F and Jiang D X. 2021. Damage identification of wind turbine blades with deep convolutional neural networks. Renewable Energy,174: 122-133 [DOI: 10.1016/j.renene.2021.04.040], articleTitle=Damage identification of wind turbine blades with deep convolutional neural networks, refAbstract=null), Reference(id=1322960972394144519, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2022, volume=182, issue=null, pageStart=1, pageEnd=16, url=null, language=null, rfNumber=null, rfOrder=26, authorNames=Hacıefendioğlu K, Başağa H B, Yavuz Z and Karimi M T, journalName=Renewable Energy, refType=null, unstructuredReference=Hacıefendioğlu K,Başağa H B,Yavuz Z and Karimi M T. 2022. Intelligent ice detection on wind turbine blades using semantic segmentation and class activation map approaches based on deep learning method. Renewable Energy,182: 1-16 [DOI: 10.1016/j.renene.2021.10.025], articleTitle=Intelligent ice detection on wind turbine blades using semantic segmentation and class activation map approaches based on deep learning method, refAbstract=null), Reference(id=1322960972461253384, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2025, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=null, rfOrder=27, authorNames=Han Y, Liu T M, Li K, journalName=Structural Health Monitoring, refType=null, unstructuredReference=Han Y,Liu T M and Li K. 2025. Fault diagnosis of wind turbine blades under wide-weather multi-operating conditions based on multi-modal information fusion and deep learning. Structural Health Monitoring [DOI: 10.1177/14759217251333761], articleTitle=Fault diagnosis of wind turbine blades under wide-weather multi-operating conditions based on multi-modal information fusion and deep learning, refAbstract=null), Reference(id=1322960972524167945, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=770, pageEnd=778, url=null, language=null, rfNumber=null, rfOrder=28, authorNames=He K M, Zhang X Y, Ren S Q, Sun J, journalName=null, refType=null, unstructuredReference=He K M,Zhang X Y,Ren S Q and Sun J. 2016. Deep residual learning for image recognition//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas,USA: IEEE: 770-778 [DOI: 10.1109/CVPR.2016.90], articleTitle=Deep residual learning for image recognition, refAbstract=null), Reference(id=1322960972612248330, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=219, issue=null, pageStart=null, pageEnd=111592, url=null, language=null, rfNumber=null, rfOrder=29, authorNames=He Y, Niu X B, Hao C P, Li Y F, Kang L, Wang Y, journalName=Mechanical Systems and Signal Processing, refType=null, unstructuredReference=He Y,Niu X B,Hao C P,Li Y F,Kang L and Wang Y. 2024. An adaptive detection approach for multi-scale defects on wind turbine blade surface. Mechanical Systems and Signal Processing,219: #111592 [DOI: 10.1016/j.ymssp.2024.111592], articleTitle=An adaptive detection approach for multi-scale defects on wind turbine blade surface, refAbstract=null), Reference(id=1322960972683551499, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=7, issue=3, pageStart=284, pageEnd=292, url=null, language=null, rfNumber=null, rfOrder=30, authorNames=Huang C, Chen M H, Wang L, journalName=Global Energy Interconnection, refType=null, unstructuredReference=Huang C,Chen M H and Wang L. 2024. Semi-supervised surface defect detection of wind turbine blades with YOLOv4. Global Energy Interconnection,7(3): 284-292 [DOI: 10.1016/j.gloei.2024.06.010], articleTitle=Semi-supervised surface defect detection of wind turbine blades with YOLOv4, refAbstract=null), Reference(id=1322960972746466060, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=1996, volume=7, issue=2, pageStart=278, pageEnd=289, url=null, language=null, rfNumber=null, rfOrder=31, authorNames=Hwang J N, You S S, Lay S R, Jou I C, journalName=IEEE Transactions on Neural Networks, refType=null, unstructuredReference=Hwang J N,You S S,Lay S R and Jou I C. 1996. The cascade-correlation learning: a projection pursuit learning perspective. IEEE Transactions on Neural Networks,7(2): 278-289 [DOI: 10.1109/72.485631], articleTitle=The cascade-correlation learning: a projection pursuit learning perspective, refAbstract=null), Reference(id=1322960972813574925, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2025, volume=63, issue=null, pageStart=null, pageEnd=4700612, url=null, language=null, rfNumber=null, rfOrder=32, authorNames=Ji L P, Liu F, Zhu D Q, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=Ji L P,Liu F and Zhu D Q. 2025. CMFCANet: a wind turbine blade damage recognition network based on color-multiscale dual-feature fusion and channel attention mechanism. IEEE Transactions on Geoscience and Remote Sensing,63: #4700612 [DOI: 10.1109/TGRS.2024.3525192], articleTitle=CMFCANet: a wind turbine blade damage recognition network based on color-multiscale dual-feature fusion and channel attention mechanism, refAbstract=null), Reference(id=1322960972897461006, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=6, issue=3, pageStart=63, pageEnd=71, url=null, language=null, rfNumber=null, rfOrder=33, authorNames=Ji L P, Wu S L, Nie T, Yang W W, Yang J Y, journalName=Flight Control and Detection, refType=null, unstructuredReference=Ji L P,Wu S L,Nie T,Yang W W and Yang J Y. 2023. Defect detection of wind turbine blade based on improved SSD. Flight Control and Detection,6(3): 63-71, articleTitle=Defect detection of wind turbine blade based on improved SSD, refAbstract=null), Reference(id=1322960972960375567, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=6, issue=3, pageStart=63, pageEnd=71, url=null, language=null, rfNumber=null, rfOrder=34, authorNames=季利鹏, 吴世龙, 聂涛, 杨文威, 杨迦迤, journalName=飞控与探测, refType=null, unstructuredReference=季利鹏,吴世龙,聂涛,杨文威,杨迦迤. 2023. 基于改进SSD的风机叶片表面缺陷检测方法. 飞控与探测,6(3): 63-71 [DOI: 10.20249/j.cnki.2096-5974.2023.03.008], articleTitle=基于改进SSD的风机叶片表面缺陷检测方法, refAbstract=null), Reference(id=1322960973023290128, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2025, volume=21, issue=1, pageStart=762, pageEnd=770, url=null, language=null, rfNumber=null, rfOrder=35, authorNames=Jia X D, Chen X, journalName=IEEE Transactions on Industrial Informatics, refType=null, unstructuredReference=Jia X D and Chen X. 2025. Unsupervised wind turbine blade damage detection with memory-aided denoising reconstruction. IEEE Transactions on Industrial Informatics,21(1): 762-770 [DOI: 10.1109/TII.2024.3459612], articleTitle=Unsupervised wind turbine blade damage detection with memory-aided denoising reconstruction, refAbstract=null), Reference(id=1322960973094593297, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=null, pageEnd=null, url=https://arxiv.org/pdf /2201.05867.pdf, language=null, rfNumber=null, rfOrder=36, authorNames=Jiang J G, Shu Y, Wang J M, Long M S, journalName=null, refType=null, unstructuredReference=Jiang J G,Shu Y,Wang J M and Long M S. 2022. Transferability in deep learning: a survey [EB/OL]. [2025-10-12]., articleTitle=Transferability in deep learning: a survey, refAbstract=null), Reference(id=1322960973161702162, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2021, volume=14, issue=18, pageStart=null, pageEnd=5974, url=null, language=null, rfNumber=null, rfOrder=37, authorNames=Katsaprakakis D A, Papadakis N, Ntintakis I, journalName=Energies, refType=null, unstructuredReference=Katsaprakakis D A,Papadakis N and Ntintakis I. 2021. A comprehensive analysis of wind turbine blade damage. Energies,14(18): #5974 [DOI: 10.3390/en14185974], articleTitle=A comprehensive analysis of wind turbine blade damage, refAbstract=null), Reference(id=1322960973228811027, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=21, issue=4, pageStart=771, pageEnd=786, url=null, language=null, rfNumber=null, rfOrder=38, authorNames=Khan P W, Byun Y C, journalName=International Journal of Green Energy, refType=null, unstructuredReference=Khan P W and Byun Y C. 2023. A review of machine learning techniques for wind turbine’s fault detection,diagnosis,and prognosis. International Journal of Green Energy,21(4): 771-786 [DOI: 10.1080/15435075.2023.2217901], articleTitle=A review of machine learning techniques for wind turbine’s fault detection,diagnosis,and prognosis, refAbstract=null), Reference(id=1322960973308502804, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=83, issue=40, pageStart=87801, pageEnd=87902, url=null, language=null, rfNumber=null, rfOrder=39, authorNames=Kumar D, Pandey R C, Mishra A K, journalName=Multimedia Tools and Applications, refType=null, unstructuredReference=Kumar D,Pandey R C and Mishra A K. 2024. A review of image features extraction techniques and their applications in image forensic. Multimedia Tools and Applications,83(40): 87801-87902 [DOI: 10.1007/s11042-023-17950-x], articleTitle=A review of image features extraction techniques and their applications in image forensic, refAbstract=null), Reference(id=1322960975032361749, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=8878, pageEnd=8887, url=null, language=null, rfNumber=null, rfOrder=40, authorNames=Kupyn O, Martyniuk T, Wu J R, Wang Z Y, journalName=null, refType=null, unstructuredReference=Kupyn O,Martyniuk T,Wu J R and Wang Z Y. 2019. DeblurGAN-V2: deblurring (orders-of-magnitude) faster and better//Proceedings of 2019 IEEE/CVF International Conference on Computer Vision. Seoul,Korea (South): IEEE: 8878-8887 [DOI: 10.1109/ICCV.2019.00897], articleTitle=DeblurGAN-V2: deblurring (orders-of-magnitude) faster and better, refAbstract=null), Reference(id=1322960975153996566, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=17, issue=8, pageStart=null, pageEnd=1796, url=null, language=null, rfNumber=null, rfOrder=41, authorNames=Lei M W, Wang X F, Wang M H, Cheng Y T, journalName=Energies, refType=null, unstructuredReference=Lei M W,Wang X F,Wang M H and Cheng Y T. 2024. Improved YOLOv5 based on multi-strategy integration for multi-category wind turbine surface defect detection. Energies,17(8): #1796 [DOI: 10.3390/en17081796], articleTitle=Improved YOLOv5 based on multi-strategy integration for multi-category wind turbine surface defect detection, refAbstract=null), Reference(id=1322960975267242775, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=47, issue=13, pageStart=89, pageEnd=99, url=null, language=null, rfNumber=null, rfOrder=42, authorNames=Li B, Zhang Y M, Wei L T, Wang Y, Zhai Y J, journalName=null, refType=null, unstructuredReference=Li B,Zhang Y M,Wei L T,Wang Y and Zhai Y J. 2024. Fan blade defect detection algorithm based on improved YOLOv8. Electronic Measurement Technology,47(13): 89-99, articleTitle=Fan blade defect detection algorithm based on improved YOLOv8, refAbstract=null), Reference(id=1322960975388877592, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=47, issue=13, pageStart=89, pageEnd=99, url=null, language=null, rfNumber=null, rfOrder=43, authorNames=李冰, 张易牧, 魏乐涛, 王月, 翟永杰, journalName=电子测量技术, refType=null, unstructuredReference=李冰,张易牧,魏乐涛,王月,翟永杰. 2024. 基于改进YOLOv8的风机桨叶缺陷检测算法. 电子测量技术,47(13): 89-99 [DOI: 10.19651/j.cnki.emt.2415882], articleTitle=基于改进YOLOv8的风机桨叶缺陷检测算法, refAbstract=null), Reference(id=1322960975699256089, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2021, volume=48, issue=2, pageStart=73, pageEnd=79, url=null, language=null, rfNumber=null, rfOrder=44, authorNames=Li J, Guo P, journalName=Journal of North China Electric Power University, refType=null, unstructuredReference=Li J and Guo P. 2021. ARGAN-based surface shadow preprocessing and transfer learning fault identification of wind turbine blade. Journal of North China Electric Power University,48(2): 73-79, articleTitle=ARGAN-based surface shadow preprocessing and transfer learning fault identification of wind turbine blade, refAbstract=null), Reference(id=1322960975770559258, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2021, volume=48, issue=2, pageStart=73, pageEnd=79, url=null, language=null, rfNumber=null, rfOrder=45, authorNames=李健, 郭鹏, journalName=华北电力大学学报, refType=null, unstructuredReference=李健,郭鹏. 2021. 基于ARGAN表面阴影预处理与迁移学习风电机组叶片故障识别. 华北电力大学学报,48(2): 73-79 [DOI: 10.3969/j.issn.1007-2691.2021.02.09], articleTitle=基于ARGAN表面阴影预处理与迁移学习风电机组叶片故障识别, refAbstract=null), Reference(id=1322960975850251035, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=174, pageEnd=178, url=null, language=null, rfNumber=null, rfOrder=46, authorNames=Li W B, Hu B, Song C, Zhao F, Ma H L, Wang Y J, journalName=null, refType=null, unstructuredReference=Li W B,Hu B,Song C,Zhao F,Ma H L and Wang Y J. 2020. An image stitching method for blades of wind turbine based on background removal preprocessing//Proceedings of the 5th International Conference on Communication,Image and Signal Processing. Chengdu,China: IEEE: 174-178 [DOI: 10.1109/CCISP51026.2020.9273502], articleTitle=An image stitching method for blades of wind turbine based on background removal preprocessing, refAbstract=null), Reference(id=1322960975908971292, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=179, issue=null, pageStart=null, pageEnd=111381, url=null, language=null, rfNumber=null, rfOrder=47, authorNames=Li W R, Pan Z H, Zhu Q X, Du Y F, journalName=null, refType=null, unstructuredReference=Li W R,Pan Z H,Zhu Q X and Du Y F. 2024. Wind turbine blade defect detection and measurement technology based on improved SegFormer and pixel matching. Optics and Laser Technology,179: #111381 [DOI: 10.1016/j.optlastec.2024.111381], articleTitle=Wind turbine blade defect detection and measurement technology based on improved SegFormer and pixel matching, refAbstract=null), Reference(id=1322960975976080157, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2020, volume=37, issue=8, pageStart=2241, pageEnd=2247, url=null, language=null, rfNumber=null, rfOrder=48, authorNames=Li X Y, Long S P, Zhu J, journalName=Application Research of Computers, refType=null, unstructuredReference=Li X Y,Long S P and Zhu J. 2020. Survey of few-shot learning based on deep neural network. Application Research of Computers,37(8): 2241-2247, articleTitle=Survey of few-shot learning based on deep neural network, refAbstract=null), Reference(id=1322960976059966238, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2020, volume=37, issue=8, pageStart=2241, pageEnd=2247, url=null, language=null, rfNumber=null, rfOrder=49, authorNames=李新叶, 龙慎鹏, 朱婧, journalName=计算机应用研究, refType=null, unstructuredReference=李新叶,龙慎鹏,朱婧. 2020. 基于深度神经网络的少样本学习综述. 计算机应用研究,37(8): 2241-2247 [DOI: 10.19734/j.issn.1001-3695.2019.03.0036], articleTitle=基于深度神经网络的少样本学习综述, refAbstract=null), Reference(id=1322960976131269407, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2019, volume=34, issue=4, pageStart=673, pageEnd=688, url=null, language=null, rfNumber=null, rfOrder=50, authorNames=Li Y X, Chai Y, Hu Y Q, Yin H P, journalName=Control and Decision, refType=null, unstructuredReference=Li Y X,Chai Y,Hu Y Q and Yin H P. 2019. Review of imbalanced data classification methods. Control and Decision,34(4): 673-688, articleTitle=Review of imbalanced data classification methods, refAbstract=null), Reference(id=1322960976215155488, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2019, volume=34, issue=4, pageStart=673, pageEnd=688, url=null, language=null, rfNumber=null, rfOrder=51, authorNames=李艳霞, 柴毅, 胡友强, 尹宏鹏, journalName=控制与决策, refType=null, unstructuredReference=李艳霞,柴毅,胡友强,尹宏鹏. 2019. 不平衡数据分类方法综述. 控制与决策,34(4): 673-688 [DOI: 10.13195/j.kzyjc.2018.0865], articleTitle=不平衡数据分类方法综述, refAbstract=null), Reference(id=1322960976307430177, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=8, issue=9, pageStart=null, pageEnd=436, url=null, language=null, rfNumber=null, rfOrder=52, authorNames=Liao Y K, Lyu M Y, Huang M Y, Qu M W, Zou K H, Chen L, journalName=Drones, refType=null, unstructuredReference=Liao Y K,Lyu M Y,Huang M Y,Qu M W,Zou K H,Chen L,et al. 2024. An improved YOLOv7 model for surface damage detection on wind turbine blades based on low-quality UAV images. Drones,8(9): #436 [DOI: 10.3390/drones8090436], articleTitle=An improved YOLOv7 model for surface damage detection on wind turbine blades based on low-quality UAV images, refAbstract=null), Reference(id=1322960976391316258, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2025, volume=40, issue=9, pageStart=2897, pageEnd=2916, url=null, language=null, rfNumber=null, rfOrder=53, authorNames=Liu C Y, Wu Y Q, Liu J J, journalName=null, refType=null, unstructuredReference=Liu C Y,Wu Y Q and Liu J J. 2025. Research progress of deep learning methods for insulator defect detection in UAV based aerial images. Transactions of China Electrotechnical Society,40(9): 2897-2916, articleTitle=Research progress of deep learning methods for insulator defect detection in UAV based aerial images, refAbstract=null), Reference(id=1322960976462619427, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2025, volume=40, issue=9, pageStart=2897, pageEnd=2916, url=null, language=null, rfNumber=null, rfOrder=54, authorNames=刘传洋, 吴一全, 刘景景, journalName=电工技术学报, refType=null, unstructuredReference=刘传洋,吴一全,刘景景. 2025. 无人机航拍图像中绝缘子缺陷检测的深度学习方法研究进展. 电工技术学报,40(9): 2897-2916 [DOI: 10.19595/j.cnki.1000-6753.tces.240610], articleTitle=无人机航拍图像中绝缘子缺陷检测的深度学习方法研究进展, refAbstract=null), Reference(id=1322960976542311204, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024a, volume=154, issue=null, pageStart=null, pageEnd=111364, url=null, language=null, rfNumber=null, rfOrder=55, authorNames=Liu L Z, Li P R, Wang D H, Zhu S Z, journalName=Applied Soft Computing, refType=null, unstructuredReference=Liu L Z,Li P R,Wang D H and Zhu S Z. 2024a. A wind turbine damage detection algorithm designed based on YOLOv8. Applied Soft Computing,154: #111364 [DOI: 10.1016/j.asoc.2024.111364], articleTitle=A wind turbine damage detection algorithm designed based on YOLOv8, refAbstract=null), Reference(id=1322960976638780197, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024b, volume=138, issue=null, pageStart=null, pageEnd=109422, url=null, language=null, rfNumber=null, rfOrder=56, authorNames=Liu Y H, Zheng Y Q, Wei T, Li Y B, journalName=Engineering Applications of Artificial Intelligence, refType=null, unstructuredReference=Liu Y H,Zheng Y Q,Wei T and Li Y B. 2024b. Lightweight algorithm based on you only look once version 5 for multiple class defect detection on wind turbine blade surfaces. Engineering Applications of Artificial Intelligence,138: #109422 [DOI: 10.1016/j.engappai.2024.109422], articleTitle=Lightweight algorithm based on you only look once version 5 for multiple class defect detection on wind turbine blade surfaces, refAbstract=null), Reference(id=1322960976705889062, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=59, issue=null, pageStart=null, pageEnd=102292, url=null, language=null, rfNumber=null, rfOrder=57, authorNames=Liu Y H, Zheng Y Q, Shao Z F, Wei T, Cui T C, Xu R, journalName=Advanced Engineering Informatics, refType=null, unstructuredReference=Liu Y H,Zheng Y Q,Shao Z F,Wei T,Cui T C and Xu R. 2024. Defect detection of the surface of wind turbine blades combining attention mechanism. Advanced Engineering Informatics,59: #102292 [DOI: 10.1016/j.aei.2023.102292], articleTitle=Defect detection of the surface of wind turbine blades combining attention mechanism, refAbstract=null), Reference(id=1322960976802358055, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=27, issue=1, pageStart=43, pageEnd=51, url=null, language=null, rfNumber=null, rfOrder=58, authorNames=Liu Y J, Wang Z, Wu X L, Fang F, Saqlain A S, journalName=IEEE Internet Computing, refType=null, unstructuredReference=Liu Y J,Wang Z,Wu X L,Fang F and Saqlain A S. 2023. Cloud-edge-end cooperative detection of wind turbine blade surface damage based on lightweight deep learning network. IEEE Internet Computing,27(1): 43-51 [DOI: 10.1109/MIC.2022.3175935], articleTitle=Cloud-edge-end cooperative detection of wind turbine blade surface damage based on lightweight deep learning network, refAbstract=null), Reference(id=1322960976873661224, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=#129510, pageEnd=null, url=null, language=null, rfNumber=null, rfOrder=59, authorNames=Lu J L, Gao Q B, Zhou K, journalName=Proceedings of SPIE 12951, refType=null, unstructuredReference=Lu J L,Gao Q B and Zhou K. 2024. Inspection of wind turbine blades using image deblurring and deep learning segmentation//Proceedings of SPIE 12951,Health Monitoring of Structural and Biological Systems XVIII. Long Beach,USA: SPIE: #129510L [DOI: 10.1117/12.3009721], articleTitle=Inspection of wind turbine blades using image deblurring and deep learning segmentation, refAbstract=null), Reference(id=1322960976944964393, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2020, volume=20, issue=6, pageStart=3070, pageEnd=3077, url=null, language=null, rfNumber=null, rfOrder=60, authorNames=Lu L, Ichimura S, Yamagishi A, Rokunohe T, journalName=IEEE Sensors Journal, refType=null, unstructuredReference=Lu L,Ichimura S,Yamagishi A and Rokunohe T. 2020. Oil film detection under solar irradiation and image processing. IEEE Sensors Journal,20(6): 3070-3077 [DOI: 10.1109/JSEN.2019.2955088], articleTitle=Oil film detection under solar irradiation and image processing, refAbstract=null), Reference(id=1322960977020461868, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2018, volume=38, issue=8, pageStart=#0815002, pageEnd=null, url=null, language=null, rfNumber=null, rfOrder=61, authorNames=Lu R S, Wu A, Zhang T D, Wang Y H, journalName=Acta Optica Sinica, refType=null, unstructuredReference=Lu R S,Wu A,Zhang T D and Wang Y H. 2018. Review on automated optical (visual) inspection and its applications in defect detection. Acta Optica Sinica,38(8): #0815002, articleTitle=Review on automated optical (visual) inspection and its applications in defect detection, refAbstract=null), Reference(id=1322960977100153645, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2018, volume=38, issue=8, pageStart=null, pageEnd=0815002, url=null, language=null, rfNumber=null, rfOrder=62, authorNames=卢荣胜, 吴昂, 张腾达, 王永红, journalName=光学学报, refType=null, unstructuredReference=卢荣胜,吴昂,张腾达,王永红. 2018. 自动光学(视觉)检测技术及其在缺陷检测中的应用综述. 光学学报,38(8): #0815002 [DOI: 10.3788/AOS201838.0815002], articleTitle=自动光学(视觉)检测技术及其在缺陷检测中的应用综述, refAbstract=null), Reference(id=1322960977171456814, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2022, volume=10, issue=null, pageStart=123378, pageEnd=123386, url=null, language=null, rfNumber=null, rfOrder=63, authorNames=Lyu L, Yao Z Y, Wang E M, Ren X, Pang R, Wang H, journalName=IEEE Access, refType=null, unstructuredReference=Lyu L,Yao Z Y,Wang E M,Ren X,Pang R,Wang H,et al. 2022. Efficient and accurate damage detector for wind turbine blade images. IEEE Access,10: 123378-123386 [DOI: 10.1109/ACCESS.2022.3224446], articleTitle=Efficient and accurate damage detector for wind turbine blade images, refAbstract=null), Reference(id=1322960977246954287, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=17, pageStart=28409, pageEnd=28418, url=null, language=null, rfNumber=null, rfOrder=64, authorNames=Ma L M, Jiang X Y, Tang Z Z, Zhi S D, Wang T Y, journalName=IEEE Sensors Journal, refType=null, unstructuredReference=Ma L M,Jiang X Y,Tang Z Z,Zhi S D and Wang T Y. 2024. Wind turbine blade defect detection algorithm based on lightweight MES-YOLOv8n. IEEE Sensors Journal,24(17): 28409-28418 [DOI: 10.1109/JSEN.2024.3430351], articleTitle=Wind turbine blade defect detection algorithm based on lightweight MES-YOLOv8n, refAbstract=null), Reference(id=1322960977351811888, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2021, volume=25, issue=2, pageStart=463, pageEnd=482, url=null, language=null, rfNumber=null, rfOrder=65, authorNames=Mao Y L, Wang S X, Yu D L, Zhao J C, journalName=Intelligent Data Analysis, refType=null, unstructuredReference=Mao Y L,Wang S X,Yu D L and Zhao J C. 2021. Automatic image detection of multi-type surface defects on wind turbine blades based on cascade deep learning network. Intelligent Data Analysis,25(2): 463-482 [DOI: 10.3233/IDA-205143], articleTitle=Automatic image detection of multi-type surface defects on wind turbine blades based on cascade deep learning network, refAbstract=null), Reference(id=1322960977427309361, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=17, issue=3, pageStart=null, pageEnd=673, url=null, language=null, rfNumber=null, rfOrder=66, authorNames=Memari M, Shekaramiz M, Masoum M A S, Seibi A C, journalName=Energies, refType=null, unstructuredReference=Memari M,Shekaramiz M,Masoum M A S and Seibi A C. 2024. Data fusion and ensemble learning for advanced anomaly detection using multi-spectral RGB and thermal imaging of small wind turbine blades. Energies,17(3): #673 [DOI: 10.3390/en17030673], articleTitle=Data fusion and ensemble learning for advanced anomaly detection using multi-spectral RGB and thermal imaging of small wind turbine blades, refAbstract=null), Reference(id=1322960977511195442, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=1, pageEnd=4, url=null, language=null, rfNumber=null, rfOrder=67, authorNames=Moll J, journalName=null, refType=null, unstructuredReference=Moll J. 2019. Numerical and experimental analysis of defect detection in jointed electromagnetic waveguides//Proceedings of the 13th European Conference on Antennas and Propagation. Krakow,Poland: IEEE: 1-4, articleTitle=Numerical and experimental analysis of defect detection in jointed electromagnetic waveguides, refAbstract=null), Reference(id=1322960977574110003, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2018, volume=162, issue=null, pageStart=51, pageEnd=60, url=null, language=null, rfNumber=null, rfOrder=68, authorNames=Moll J, Simon J, Malzer M, Krozer V, Pozdniakov D, Salman R, journalName=Progress in Electromagnetics Research, refType=null, unstructuredReference=Moll J,Simon J,Malzer M,Krozer V,Pozdniakov D,Salman R,et al. 2018. Radar imaging system for in-service wind turbine blades inspections: initial results from a field installation at a 2 MW wind turbine. Progress in Electromagnetics Research,162: 51-60 [DOI: 10.2528/PIER18021905], articleTitle=Radar imaging system for in-service wind turbine blades inspections: initial results from a field installation at a 2 MW wind turbine, refAbstract=null), Reference(id=1322960977628635956, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=1, pageEnd=5, url=null, language=null, rfNumber=null, rfOrder=69, authorNames=Moreno S, Peña M, Toledo A, Treviño R, Ponce H, journalName=null, refType=null, unstructuredReference=Moreno S,Peña M,Toledo A,Treviño R and Ponce H. 2018. A new vision-based method using deep learning for damage inspection in wind turbine blades//Proceedings of the 15th International Conference on Electrical Engineering,Computing Science and Automatic Control. Mexico City,Mexico: IEEE: 1-5 [DOI: 10.1109/ICEEE.2018.8533924], articleTitle=A new vision-based method using deep learning for damage inspection in wind turbine blades, refAbstract=null), Reference(id=1322960977695744821, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2012, volume=20, issue=2, pageStart=986, pageEnd=1000, url=null, language=null, rfNumber=null, rfOrder=70, authorNames=Nakauchi S, Nishino K, Yamashita T, journalName=Optics Express, refType=null, unstructuredReference=Nakauchi S,Nishino K and Yamashita T. 2012. Selection of optimal combinations of band-pass filters for ice detection by hyperspectral imaging. Optics Express,20(2): 986-1000 [DOI: 10.1364/OE.20.000986], articleTitle=Selection of optimal combinations of band-pass filters for ice detection by hyperspectral imaging, refAbstract=null), Reference(id=1322960977771242294, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2025, volume=null, issue=null, pageStart=null, pageEnd=null, url=https://www.nea.gov.cn/20250121/097bfd7c1cd3498897639857d86d5dac/c.html (, language=null, rfNumber=null, rfOrder=71, authorNames=National Energy Administration, journalName=null, refType=null, unstructuredReference=National Energy Administration. 2025. National electricity industry statistics data of 2024 [EB/OL]. [2025-10-12]., articleTitle=National electricity industry statistics data of 2024, refAbstract=null), Reference(id=1322960979042116408, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2025, volume=null, issue=null, pageStart=null, pageEnd=null, url=https://www.nea.gov.cn/20250121/097bfd7c1cd3498897639857d86d5dac/c.html, language=null, rfNumber=null, rfOrder=72, authorNames=国家能源局, journalName=null, refType=null, unstructuredReference=国家能源局. 2025. 国家能源局发布2024年全国电力工业统计数据 [EB/OL]. [2025-10-12]., articleTitle=国家能源局发布2024年全国电力工业统计数据, refAbstract=null), Reference(id=1322960979167945529, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=null, rfOrder=73, authorNames=Nijssen R, Manrique E, journalName=TNO, refType=null, unstructuredReference=Nijssen R and Manrique E. 2020. Literature Review of Structural and Non-Structural Wind Turbine Blade Damage. TNO, articleTitle=Literature Review of Structural and Non-Structural Wind Turbine Blade Damage, refAbstract=null), Reference(id=1322960979293774650, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=28, issue=9, pageStart=2587, pageEnd=2615, url=null, language=null, rfNumber=null, rfOrder=74, authorNames=Pan X Y, Jia N X, Mu Y Z, Gao X R, journalName=Journal of Image and Graphics, refType=null, unstructuredReference=Pan X Y,Jia N X,Mu Y Z and Gao X R. 2023. Survey of small object detection. Journal of Image and Graphics,28(9): 2587-2615, articleTitle=Survey of small object detection, refAbstract=null), Reference(id=1322960979381855035, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=28, issue=9, pageStart=2587, pageEnd=2615, url=null, language=null, rfNumber=null, rfOrder=75, authorNames=潘晓英, 贾凝心, 穆元震, 高炫蓉, journalName=中国图象图形学报, refType=null, unstructuredReference=潘晓英,贾凝心,穆元震,高炫蓉. 2023. 小目标检测研究综述. 中国图象图形学报,28(9): 2587-2615 [DOI: 10.11834/jig.220455], articleTitle=小目标检测研究综述, refAbstract=null), Reference(id=1322960979465741116, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2011, volume=65, issue=1, pageStart=88, pageEnd=96, url=null, language=null, rfNumber=null, rfOrder=76, authorNames=Parent O, Ilinca A, journalName=Cold Regions Science and Technology, refType=null, unstructuredReference=Parent O and Ilinca A. 2011. Anti-icing and de-icing techniques for wind turbines: critical review. Cold Regions Science and Technology,65(1): 88-96 [DOI: 10.1016/j.coldregions.2010.01.005], articleTitle=Anti-icing and de-icing techniques for wind turbines: critical review, refAbstract=null), Reference(id=1322960979528655677, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2022a, volume=14, issue=1, pageStart=null, pageEnd=87, url=null, language=null, rfNumber=null, rfOrder=77, authorNames=Peng Y P, Tang Z, Zhao G P, Cao G Z, Wu C, journalName=Remote Sensing, refType=null, unstructuredReference=Peng Y P,Tang Z,Zhao G P,Cao G Z and Wu C. 2022a. Motion blur removal for UAV-based wind turbine blade images using synthetic datasets. Remote Sensing,14(1): #87 [DOI: 10.3390/rs14010087], articleTitle=Motion blur removal for UAV-based wind turbine blade images using synthetic datasets, refAbstract=null), Reference(id=1322960979595764542, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2022b, volume=170, issue=null, pageStart=null, pageEnd=108797, url=null, language=null, rfNumber=null, rfOrder=78, authorNames=Peng Y P, Wang W J, Tang Z, Cao G Z, Zhou S X, journalName=Mechanical Systems and Signal Processing, refType=null, unstructuredReference=Peng Y P,Wang W J,Tang Z,Cao G Z and Zhou S X. 2022b. Non-uniform illumination image enhancement for surface damage detection of wind turbine blades. Mechanical Systems and Signal Processing,170: #108797 [DOI: 10.1016/j.ymssp.2021.108797], articleTitle=Non-uniform illumination image enhancement for surface damage detection of wind turbine blades, refAbstract=null), Reference(id=1322960979679650623, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2025, volume=13, issue=4, pageStart=null, pageEnd=283, url=null, language=null, rfNumber=null, rfOrder=79, authorNames=Pratt R, Allen C, Masoum M A S, Seibi A, journalName=Machines, refType=null, unstructuredReference=Pratt R,Allen C,Masoum M A S and Seibi A. 2025. Defect detection and classification on wind turbine blades using deep learning with fuzzy voting. Machines,13(4): #283 [DOI: 10.3390/machines13040283], articleTitle=Defect detection and classification on wind turbine blades using deep learning with fuzzy voting, refAbstract=null), Reference(id=1322960979750953792, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2019, volume=28, issue=4, pageStart=null, pageEnd=043023, url=null, language=null, rfNumber=null, rfOrder=80, authorNames=Qiu Z, Wang S, Zeng Z, Yu D, journalName=Journal of Electronic Imaging, refType=null, unstructuredReference=Qiu Z,Wang S,Zeng Z and Yu D. 2019. Automatic visual defects inspection of wind turbine blades via YOLO-based small object detection approach. Journal of Electronic Imaging,28(4): #043023 [DOI: 10.1117/1.JEI.28.4.043023], articleTitle=Automatic visual defects inspection of wind turbine blades via YOLO-based small object detection approach, refAbstract=null), Reference(id=1322960979826451265, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2022, volume=10, issue=null, pageStart=122171, pageEnd=122181, url=null, language=null, rfNumber=null, rfOrder=81, authorNames=Ran X K, Zhang S, Wang H T, Zhang Z Y, journalName=IEEE Access, refType=null, unstructuredReference=Ran X K,Zhang S,Wang H T and Zhang Z Y. 2022. An improved algorithm for wind turbine blade defect detection. IEEE Access,10: 122171-122181 [DOI: 10.1109/ACCESS.2022.3216613], articleTitle=An improved algorithm for wind turbine blade defect detection, refAbstract=null), Reference(id=1322960979943891778, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2019, volume=147, issue=null, pageStart=null, pageEnd=106823, url=null, language=null, rfNumber=null, rfOrder=82, authorNames=Reddy A, Indragandhi V, Ravi L, Subramaniyaswamy V, journalName=Measurement, refType=null, unstructuredReference=Reddy A,Indragandhi V,Ravi L and Subramaniyaswamy V. 2019. Detection of cracks and damage in wind turbine blades using artificial intelligence-based image analytics. Measurement,147: #106823 [DOI: 10.1016/j.measurement.2019.07.051], articleTitle=Detection of cracks and damage in wind turbine blades using artificial intelligence-based image analytics, refAbstract=null), Reference(id=1322960980031972163, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2020, volume=18, issue=null, pageStart=null, pageEnd=100291, url=null, language=null, rfNumber=null, rfOrder=83, authorNames=Rizk P, Al Saleh N, Younes R, Ilinca A, Khoder J, journalName=Remote Sensing Applications: Society and Environment, refType=null, unstructuredReference=Rizk P,Al Saleh N,Younes R,Ilinca A and Khoder J. 2020. Hyperspectral imaging applied for the detection of wind turbine blade damage and icing. Remote Sensing Applications: Society and Environment,18: #100291 [DOI: 10.1016/j.rsase.2020.100291], articleTitle=Hyperspectral imaging applied for the detection of wind turbine blade damage and icing, refAbstract=null), Reference(id=1322960980090692420, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024a, volume=18, issue=8, pageStart=6299, pageEnd=6307, url=null, language=null, rfNumber=null, rfOrder=84, authorNames=Rizvi S Z, Jamil M, Huang W M, journalName=Signal,Image and Video Processing, refType=null, unstructuredReference=Rizvi S Z,Jamil M and Huang W M. 2024a. Pixel U-Net: an improved version of U-Net for binary segmentation of wind turbine blades. Signal,Image and Video Processing,18(8): 6299-6307 [DOI: 10.1007/s11760-024-03317-8], articleTitle=Pixel U-Net: an improved version of U-Net for binary segmentation of wind turbine blades, refAbstract=null), Reference(id=1322960980161995589, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024b, volume=120, issue=null, pageStart=null, pageEnd=109615, url=null, language=null, rfNumber=null, rfOrder=85, authorNames=Rizvi S Z, Jamil M, Huang W M, journalName=Computers and Electrical Engineering, refType=null, unstructuredReference=Rizvi S Z,Jamil M and Huang W M. 2024b. Enhanced defect detection on wind turbine blades using binary segmentation masks and YOLO. Computers and Electrical Engineering,120: #109615 [DOI: 10.1016/j.compeleceng.2024.109615], articleTitle=Enhanced defect detection on wind turbine blades using binary segmentation masks and YOLO, refAbstract=null), Reference(id=1322960980220715846, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2018, volume=120, issue=null, pageStart=423, pageEnd=433, url=null, language=null, rfNumber=null, rfOrder=86, authorNames=Rubert T, McMillan D, Niewczas P, journalName=Renewable Energy, refType=null, unstructuredReference=Rubert T,McMillan D and Niewczas P. 2018. A decision support tool to assist with lifetime extension of wind turbines. Renewable Energy,120: 423-433 [DOI: 10.1016/j.renene.2017.12.064], articleTitle=A decision support tool to assist with lifetime extension of wind turbines, refAbstract=null), Reference(id=1322960980292019015, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2021, volume=12, issue=8, pageStart=8535, pageEnd=8548, url=null, language=null, rfNumber=null, rfOrder=87, authorNames=Sarkar D, Gunturi S K, journalName=Journal of Ambient Intelligence and Humanized Computing, refType=null, unstructuredReference=Sarkar D and Gunturi S K. 2021. Wind turbine blade structural state evaluation by hybrid object detector relying on deep learning models. Journal of Ambient Intelligence and Humanized Computing,12(8): 8535-8548 [DOI: 10.1007/s12652-020-02587-7], articleTitle=Wind turbine blade structural state evaluation by hybrid object detector relying on deep learning models, refAbstract=null), Reference(id=1322960980363322184, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=https://data.mendeley.com/datasets/hd96prn3nc/2, language=null, rfNumber=null, rfOrder=88, authorNames=Shihavuddin A, Chen X, DTU-drone inspection images of wind turbine, journalName=null, refType=null, unstructuredReference=Shihavuddin A and Chen X. 2018. DTU-drone inspection images of wind turbine. Mendeley Data,V2 [EB/OL]. [2025-10-12]., articleTitle=Mendeley Data,V2, refAbstract=null), Reference(id=1322960980451402569, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2019, volume=12, issue=4, pageStart=null, pageEnd=676, url=null, language=null, rfNumber=null, rfOrder=89, authorNames=Shihavuddin A, Chen X, Fedorov V, Nymark Christensen A, Andre Brogaard Riis N, Branner K, journalName=Energies, refType=null, unstructuredReference=Shihavuddin A,Chen X,Fedorov V,Nymark Christensen A,Andre Brogaard Riis N,Branner K,et al. 2019. Wind turbine surface damage detection by deep learning aided drone inspection analysis. Energies,12(4): #676 [DOI: 10.3390/en12040676], articleTitle=Wind turbine surface damage detection by deep learning aided drone inspection analysis, refAbstract=null), Reference(id=1322960980531094346, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=46, pageEnd=51, url=null, language=null, rfNumber=null, rfOrder=90, authorNames=Sun H H, Zhang W T, Ma D L, Zhang Y M, Li D X, Gao X D, journalName=YOLO-WTDL: a lightweight wind turbine blades defect detection model based on YOLOv8//Proceedings of the 7th Asia Conference on Energy and Electrical Engineering, refType=null, unstructuredReference=Sun H H,Zhang W T,Ma D L,Zhang Y M,Li D X and Gao X D. 2024. YOLO-WTDL: a lightweight wind turbine blades defect detection model based on YOLOv8//Proceedings of the 7th Asia Conference on Energy and Electrical Engineering. Chengdu,China: IEEE: 46-51 [DOI: 10.1109/ACEEE62329.2024.10651628], articleTitle=null, refAbstract=null), Reference(id=1322960980610786123, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2011, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=null, rfOrder=91, authorNames=Szeliski R, journalName=null, refType=null, unstructuredReference=Szeliski R. 2011. Computer vision: algorithms and applications. Springer. [DOI: 10.1007/978-1-84882-935-0], articleTitle=Computer vision: algorithms and applications, refAbstract=null), Reference(id=1322960980690477900, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2017, volume=5, issue=null, pageStart=1184, pageEnd=1191, url=null, language=null, rfNumber=null, rfOrder=92, authorNames=Tiwari K A, Raisutis R, Samaitis S, journalName=Procedia Structural Integrity, refType=null, unstructuredReference=Tiwari K A,Raisutis R and Samaitis S. 2017. Signal processing methods to improve the signal-to-noise ratio (SNR) in ultrasonic non-destructive testing of wind turbine blade. Procedia Structural Integrity,5: 1184-1191 [DOI: 10.1016/j.prostr.2017.07.036], articleTitle=Signal processing methods to improve the signal-to-noise ratio (SNR) in ultrasonic non-destructive testing of wind turbine blade, refAbstract=null), Reference(id=1322960980782752589, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2008, volume=null, issue=null, pageStart=1096, pageEnd=1103, url=null, language=null, rfNumber=null, rfOrder=93, authorNames=Vincent P, Larochelle H, Bengio Y, Manzagol P A, journalName=null, refType=null, unstructuredReference=Vincent P,Larochelle H,Bengio Y and Manzagol P A. 2008. Extracting and composing robust features with denoising autoencoders//Proceedings of the 25th International Conference on Machine Learning. Helsinki,Finland: ACM: 1096-1103 [DOI: 10.1145/1390156.1390294], articleTitle=Extracting and composing robust features with denoising autoencoders, refAbstract=null), Reference(id=1322960980875027278, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=80, issue=1, pageStart=625, pageEnd=642, url=null, language=null, rfNumber=null, rfOrder=94, authorNames=Wang H J, Shen Q Z, Dai Q, Gao Y C, Gao J, Zhang T, journalName=Computers,Materials and Continua, refType=null, unstructuredReference=Wang H J,Shen Q Z,Dai Q,Gao Y C,Gao J and Zhang T. 2024. Evolutionary variational YOLOv8 network for fault detection in wind turbines. Computers,Materials and Continua,80(1): 625-642 [DOI: 10.32604/cmc.2024.051757], articleTitle=Evolutionary variational YOLOv8 network for fault detection in wind turbines, refAbstract=null), Reference(id=1322960980996662095, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=56, issue=10, pageStart=80, pageEnd=95, url=null, language=null, rfNumber=null, rfOrder=95, authorNames=Wang L, Liu Y B, Teng W, Huang X W, Liu J T, journalName=Electric Power, refType=null, unstructuredReference=Wang L,Liu Y B,Teng W,Huang X W and Liu J T. 2023. Research and development of nondestructive detection technology for wind turbine blades. Electric Power,56(10): 80-95, articleTitle=Research and development of nondestructive detection technology for wind turbine blades, refAbstract=null), Reference(id=1322960981122491216, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023b, volume=56, issue=10, pageStart=80, pageEnd=95, url=null, language=null, rfNumber=null, rfOrder=96, authorNames=王磊, 柳亦兵, 滕伟, 黄心伟, 刘剑韬, journalName=中国电力, refType=null, unstructuredReference=王磊,柳亦兵,滕伟,黄心伟,刘剑韬. 2023b. 风电机组叶片无损检测技术研究与进展. 中国电力,56(10): 80-95 [DOI: 10.11930/j.issn.1004-9649.202303073], articleTitle=风电机组叶片无损检测技术研究与进展, refAbstract=null), Reference(id=1322960981218960209, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2017, volume=64, issue=9, pageStart=7293, pageEnd=7303, url=null, language=null, rfNumber=null, rfOrder=97, authorNames=Wang L, Zhang Z J, journalName=IEEE Transactions on Industrial Electronics, refType=null, unstructuredReference=Wang L and Zhang Z J. 2017. Automatic detection of wind turbine blade surface cracks based on UAV-taken images. IEEE Transactions on Industrial Electronics,64(9): 7293-7303 [DOI: 10.1109/TIE.2017.2682037], articleTitle=Automatic detection of wind turbine blade surface cracks based on UAV-taken images, refAbstract=null), Reference(id=1322960981332206418, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2019a, volume=24, issue=3, pageStart=1271, pageEnd=1281, url=null, language=null, rfNumber=null, rfOrder=98, authorNames=Wang L, Zhang Z J, Lou X, journalName=IEEE/ASME Transactions on Mechatronics, refType=null, unstructuredReference=Wang L,Zhang Z J and Lou X. 2019a. A two-stage data-driven approach for image-based wind turbine blade crack inspections. IEEE/ASME Transactions on Mechatronics,24(3): 1271-1281 [DOI: 10.1109/TMECH.2019.2908233], articleTitle=A two-stage data-driven approach for image-based wind turbine blade crack inspections, refAbstract=null), Reference(id=1322960981416092499, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2021, volume=50, issue=3, pageStart=257, pageEnd=274, url=null, language=null, rfNumber=null, rfOrder=99, authorNames=Wang Q Z, Jin G S, Li Q, Wang K, Yang Z Y, Wang H, journalName=Information and Control, refType=null, unstructuredReference=Wang Q Z,Jin G S,Li Q,Wang K,Yang Z Y and Wang H. 2021. Industrial edge computing: vision and challenges. Information and Control,50(3): 257-274, articleTitle=Industrial edge computing: vision and challenges, refAbstract=null), Reference(id=1322960981504172884, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2021, volume=50, issue=3, pageStart=257, pageEnd=274, url=null, language=null, rfNumber=null, rfOrder=100, authorNames=王其朝, 金光淑, 李庆, 王锴, 杨祖业, 王宏, journalName=信息与控制, refType=null, unstructuredReference=王其朝,金光淑,李庆,王锴,杨祖业,王宏. 2021. 工业边缘计算研究现状与展望. 信息与控制,50(3): 257-274 [DOI: 10.13976/j.cnki.xk.2021.1030], articleTitle=工业边缘计算研究现状与展望, refAbstract=null), Reference(id=1322960981592253269, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2022, volume=15, issue=15, pageStart=null, pageEnd=5672, url=null, language=null, rfNumber=null, rfOrder=101, authorNames=Wang W J, Xue Y, He C K, Zhao Y N, journalName=Energies, refType=null, unstructuredReference=Wang W J,Xue Y,He C K and Zhao Y N. 2022. Review of the typical damage and damage-detection methods of large wind turbine blades. Energies,15(15): #5672 [DOI: 10.3390/en15155672], articleTitle=Review of the typical damage and damage-detection methods of large wind turbine blades, refAbstract=null), Reference(id=1322960981671945046, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=8, issue=2, pageStart=1069, pageEnd=1076, url=null, language=null, rfNumber=null, rfOrder=102, authorNames=Wang Y F, Lu Q, Ren B B, journalName=IEEE Robotics and Automation Letters, refType=null, unstructuredReference=Wang Y F,Lu Q and Ren B B. 2023. Wind turbine crack inspection using a quadrotor with image motion blur avoided. IEEE Robotics and Automation Letters,8(2): 1069-1076 [DOI: 10.1109/LRA.2023.3236576], articleTitle=Wind turbine crack inspection using a quadrotor with image motion blur avoided, refAbstract=null), Reference(id=1322960981747442519, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2019b, volume=11, issue=1, pageStart=#3, pageEnd=null, url=null, language=null, rfNumber=null, rfOrder=103, authorNames=Wang Y N, Yoshihashi R, Kawakami R, You S D, Harano T, Ito M, journalName=IPSJ Transactions on Computer Vision and Applications, refType=null, unstructuredReference=Wang Y N,Yoshihashi R,Kawakami R,You S D,Harano T,Ito M,et al. 2019b. Unsupervised anomaly detection with compact deep features for wind turbine blade images taken by a drone. IPSJ Transactions on Computer Vision and Applications,11(1): #3. [DOI: 10.1186/s41074-019-0056-0], articleTitle=Unsupervised anomaly detection with compact deep features for wind turbine blade images taken by a drone, refAbstract=null), Reference(id=1322960981810357080, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2019, volume=11, issue=5, pageStart=null, pageEnd=053305, url=null, language=null, rfNumber=null, rfOrder=104, authorNames=Xu D H, Wen C B, Liu J H, journalName=Journal of Renewable and Sustainable Energy, refType=null, unstructuredReference=Xu D H,Wen C B and Liu J H. 2019. Wind turbine blade surface inspection based on deep learning and UAV-taken images. Journal of Renewable and Sustainable Energy,11(5): #053305 [DOI: 10.1063/1.5113532], articleTitle=Wind turbine blade surface inspection based on deep learning and UAV-taken images, refAbstract=null), Reference(id=1322960981873271641, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2018, volume=20, issue=3, pageStart=44, pageEnd=50, url=null, language=null, rfNumber=null, rfOrder=105, authorNames=Xu G D, Ye H Y, Xie H B, journalName=Strategic Study of CAE, refType=null, unstructuredReference=Xu G D,Ye H Y and Xie H B. 2018. The current state and future development of wind turbine technology. Strategic Study of CAE,20(3): 44-50, articleTitle=The current state and future development of wind turbine technology, refAbstract=null), Reference(id=1322960981940380506, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2018, volume=20, issue=3, pageStart=44, pageEnd=50, url=null, language=null, rfNumber=null, rfOrder=106, authorNames=许国东, 叶杭冶, 解鸿斌, journalName=中国工程科学, refType=null, unstructuredReference=许国东,叶杭冶,解鸿斌. 2018. 风电机组技术现状及发展方向. 中国工程科学,20(3): 44-50 [DOI: 10.15302/J-SSCAE-2018.03.007], articleTitle=风电机组技术现状及发展方向, refAbstract=null), Reference(id=1322960981999100763, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=203, issue=null, pageStart=267, pageEnd=279, url=null, language=null, rfNumber=null, rfOrder=107, authorNames=Yang C, Liu X, Zhou H, Ke Y, See J, journalName=Renewable Energy, refType=null, unstructuredReference=Yang C,Liu X,Zhou H,Ke Y and See J. 2023. Towards accurate image stitching for drone-based wind turbine blade inspection. Renewable Energy,203: 267-279 [DOI: 10.1016/j.renene.2022.12.063], articleTitle=Towards accurate image stitching for drone-based wind turbine blade inspection, refAbstract=null), Reference(id=1322960982078792540, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2025a, volume=15, issue=1, pageStart=null, pageEnd=3930, url=null, language=null, rfNumber=null, rfOrder=108, authorNames=Yang C C, Ding S H, Zhou G S, journalName=Scientific Reports, refType=null, unstructuredReference=Yang C C,Ding S H and Zhou G S. 2025a. Wind turbine blade damage detection based on acoustic signals. Scientific Reports,15(1): #3930 [DOI: 10.1038/s41598-025-88276-x], articleTitle=Wind turbine blade damage detection based on acoustic signals, refAbstract=null), Reference(id=1322960982162678621, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=8, pageEnd=213, url=null, language=null, rfNumber=null, rfOrder=109, authorNames=Yang G Y, Xiong W, Lei J Q, Yang K, journalName=Blade-YOLOv, refType=null, unstructuredReference=Yang G Y,Xiong W,Lei J Q and Yang K. 2024. Blade-YOLOv8: improved YOLOv8 for wind turbine blade defect detection//Proceedings of the 4th Power System and Green Energy Conference. Shanghai,China: IEEE: 209-213 [DOI: 10.1109/PSGEC62376.2024.10721051], articleTitle=improved YOLOv8 for wind turbine blade defect detection, refAbstract=null), Reference(id=1322960982221398878, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2025b, volume=28, issue=12, pageStart=2529, pageEnd=2539, url=null, language=null, rfNumber=null, rfOrder=110, authorNames=Yang J W, Chen X C, Cao S X, Zhao B, Tang Z H, Wang G, journalName=Journal of Applied Science and Engineering, refType=null, unstructuredReference=Yang J W,Chen X C,Cao S X,Zhao B,Tang Z H,Wang G,et al. 2025b. Research on surface loss detection algorithm of wind turbine blade based on FRE-DETR network. Journal of Applied Science and Engineering,28(12): 2529-2539 [DOI: 10.6180/jase.202512_28(12).0002], articleTitle=Research on surface loss detection algorithm of wind turbine blade based on FRE-DETR network, refAbstract=null), Reference(id=1322960982275924831, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2021, volume=163, issue=null, pageStart=386, pageEnd=397, url=null, language=null, rfNumber=null, rfOrder=111, authorNames=Yang X Y, Zhang Y F, Lyu W, Wang D, journalName=Renewable Energy, refType=null, unstructuredReference=Yang X Y,Zhang Y F,Lyu W and Wang D. 2021. Image recognition of wind turbine blade damage based on a deep learning model with transfer learning and an ensemble learning classifier. Renewable Energy,163: 386-397 [DOI: 10.1016/j.renene.2020.08.125], articleTitle=Image recognition of wind turbine blade damage based on a deep learning model with transfer learning and an ensemble learning classifier, refAbstract=null), Reference(id=1322960982330450784, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=16, issue=9, pageStart=null, pageEnd=3776, url=null, language=null, rfNumber=null, rfOrder=112, authorNames=Yao Y, Wang G Z, Fan J H, journalName=Energies, refType=null, unstructuredReference=Yao Y,Wang G Z and Fan J H. 2023. WT-YOLOX: an efficient detection algorithm for wind turbine blade damage based on YOLOX. Energies,16(9): #3776 [DOI: 10.3390/en16093776], articleTitle=WT-YOLOX: an efficient detection algorithm for wind turbine blade damage based on YOLOX, refAbstract=null), Reference(id=1322960982410142561, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=136, issue=null, pageStart=null, pageEnd=108908, url=null, language=null, rfNumber=null, rfOrder=113, authorNames=Ye X Y, Wang L, Huang C, Lou X, journalName=Engineering Applications of Artificial Intelligence, refType=null, unstructuredReference=Ye X Y,Wang L,Huang C and Lou X. 2024. Wind turbine blade defect detection with a semi-supervised deep learning framework. Engineering Applications of Artificial Intelligence,136: #108908 [DOI: 10.1016/j.engappai.2024.108908], articleTitle=Wind turbine blade defect detection with a semi-supervised deep learning framework, refAbstract=null), Reference(id=1322960982477251426, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=12, issue=1, pageStart=null, pageEnd=205, url=null, language=null, rfNumber=null, rfOrder=114, authorNames=Yu H, Wang J G, Han Y X, Fan B, Zhang C, journalName=Processes, refType=null, unstructuredReference=Yu H,Wang J G,Han Y X,Fan B and Zhang C. 2024. Research on an intelligent identification method for wind turbine blade damage based on CBAM-BiFPN-YOLOV8. Processes,12(1): #205 [DOI: 10.3390/pr12010205], articleTitle=Research on an intelligent identification method for wind turbine blade damage based on CBAM-BiFPN-YOLOV8, refAbstract=null), Reference(id=1322960982544360291, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023a, volume=23, issue=2, pageStart=1318, pageEnd=1327, url=null, language=null, rfNumber=null, rfOrder=115, authorNames=Yu J F, He Y Z, Liu H, Zhang F, Li J, Sun G S, journalName=IEEE Sensors Journal, refType=null, unstructuredReference=Yu J F,He Y Z,Liu H,Zhang F,Li J,Sun G S,et al. 2023a. An improved U-net model for infrared image segmentation of wind turbine blade. IEEE Sensors Journal,23(2): 1318-1327 [DOI: 10.1109/JSEN.2022.3224837], articleTitle=An improved U-net model for infrared image segmentation of wind turbine blade, refAbstract=null), Reference(id=1322960982640829284, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023b, volume=23, issue=8, pageStart=8727, pageEnd=8736, url=null, language=null, rfNumber=null, rfOrder=116, authorNames=Yu J F, He Y Z, Zhang F, Sun G S, Hou Y J, Liu H, journalName=IEEE Sensors Journal, refType=null, unstructuredReference=Yu J F,He Y Z,Zhang F,Sun G S,Hou Y J,Liu H,et al. 2023b. An infrared image stitching method for wind turbine blade using UAV flight data and U-Net. IEEE Sensors Journal,23(8): 8727-8736 [DOI: 10.1109/JSEN.2023.3257824], articleTitle=An infrared image stitching method for wind turbine blade using UAV flight data and U-Net, refAbstract=null), Reference(id=1322960982733103973, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2020, volume=376, issue=null, pageStart=1, pageEnd=9, url=null, language=null, rfNumber=null, rfOrder=117, authorNames=Yu Y J, Cao H, Yan X Y, Wang T, Ge S S, journalName=Neurocomputing, refType=null, unstructuredReference=Yu Y J,Cao H,Yan X Y,Wang T and Ge S S. 2020. Defect identification of wind turbine blades based on defect semantic features with transfer feature extractor. Neurocomputing,376: 1-9 [DOI: 10.1016/j.neucom.2019.09.071], articleTitle=Defect identification of wind turbine blades based on defect semantic features with transfer feature extractor, refAbstract=null), Reference(id=1322960982812795750, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2020, volume=12, issue=5, pageStart=null, pageEnd=053302, url=null, language=null, rfNumber=null, rfOrder=118, authorNames=Zhang C, Wen C B, Liu J H, journalName=Journal of Renewable and Sustainable Energy, refType=null, unstructuredReference=Zhang C,Wen C B and Liu J H. 2020. Mask-MRNet: a deep neural network for wind turbine blade fault detection. Journal of Renewable and Sustainable Energy,12(5): #053302 [DOI: 10.1063/5.0014223], articleTitle=Mask-MRNet: a deep neural network for wind turbine blade fault detection, refAbstract=null), Reference(id=1322960982888293223, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=16, pageStart=null, pageEnd=6009, url=null, language=null, rfNumber=null, rfOrder=119, authorNames=Zhang C, Yang T, Yang J, journalName=Sensors, refType=null, unstructuredReference=Zhang C,Yang T and Yang J. 2022. Image recognition of wind turbine blade defects using attention-based MobileNetv1-YOLOv4 and transfer learning. Sensors,22(16): #6009 [DOI: 10.3390/s22166009], articleTitle=Image recognition of wind turbine blade defects using attention-based MobileNetv1-YOLOv4 and transfer learning, refAbstract=null), Reference(id=1322960982980567912, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2021, volume=76, issue=null, pageStart=323, pageEnd=336, url=null, language=null, rfNumber=null, rfOrder=120, authorNames=Zhang H, Xu H, Tian X, Jiang J J, Ma J Y, journalName=Information Fusion, refType=null, unstructuredReference=Zhang H,Xu H,Tian X,Jiang J J and Ma J Y. 2021. Image fusion meets deep learning: a survey and perspective. Information Fusion,76: 323-336 [DOI: 10.1016/j.inffus.2021.06.008], articleTitle=Image fusion meets deep learning: a survey and perspective, refAbstract=null), Reference(id=1322960983064453993, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2022, volume=5, issue=7, pageStart=null, pageEnd=2100631, url=null, language=null, rfNumber=null, rfOrder=121, authorNames=Zhang R, Wen C B, journalName=Advanced Theory and Simulations, refType=null, unstructuredReference=Zhang R and Wen C B. 2022. SOD-YOLO: a small target defect detection algorithm for wind turbine blades based on improved YOLOV5. Advanced Theory and Simulations,5(7): #2100631 [DOI: 10.1002/adts.202100631], articleTitle=SOD-YOLO: a small target defect detection algorithm for wind turbine blades based on improved YOLOV5, refAbstract=null), Reference(id=1322960983135757162, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=35, issue=10, pageStart=null, pageEnd=105408, url=null, language=null, rfNumber=null, rfOrder=122, authorNames=Zhang Y, Fang Y, Gao W W, Liu X T, Yang H, Tong Y M, journalName=Measurement Science and Technology, refType=null, unstructuredReference=Zhang Y,Fang Y,Gao W W,Liu X T,Yang H,Tong Y M,et al. 2024. Attention mechanism based on deep learning for defect detection of wind turbine blade via multi-scale features. Measurement Science and Technology,35(10): #105408 [DOI: 10.1088/1361-6501/ad6024], articleTitle=Attention mechanism based on deep learning for defect detection of wind turbine blade via multi-scale features, refAbstract=null), Reference(id=1322960983232226155, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=220, issue=null, pageStart=null, pageEnd=113222, url=null, language=null, rfNumber=null, rfOrder=123, authorNames=Zhang Y S, Yang Y L, Sun J Q, Ji R, Zhang P Y, Shan H L, journalName=Measurement, refType=null, unstructuredReference=Zhang Y S,Yang Y L,Sun J Q,Ji R,Zhang P Y and Shan H L. 2023. Surface defect detection of wind turbine based on lightweight YOLOv5s model. Measurement,220: #113222 [DOI: 10.1016/j.measurement.2023.113222], articleTitle=Surface defect detection of wind turbine based on lightweight YOLOv5s model, refAbstract=null), Reference(id=1322960983303529324, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=17, issue=15, pageStart=null, pageEnd=3731, url=null, language=null, rfNumber=null, rfOrder=124, authorNames=Zhang Z Y, Shu Z R, journalName=Energies, refType=null, unstructuredReference=Zhang Z Y and Shu Z R. 2024. Unmanned aerial vehicle (UAV)-assisted damage detection of wind turbine blades: a review. Energies,17(15): #3731 [DOI: 10.3390/en17153731], articleTitle=Unmanned aerial vehicle (UAV)-assisted damage detection of wind turbine blades: a review, refAbstract=null), Reference(id=1322960983588741997, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=12, issue=null, pageStart=99349, pageEnd=99361, url=null, language=null, rfNumber=null, rfOrder=125, authorNames=Zhao Y X, Qiu Z B, Zhang Y, Quan H, Wei Z H, Zhu X, journalName=IEEE Access, refType=null, unstructuredReference=Zhao Y X,Qiu Z B,Zhang Y,Quan H,Wei Z H and Zhu X. 2024. Detection of defects in wind turbin blade based on cascaded adaptive hybrid attention network. IEEE Access,12: 99349-99361 [DOI: 10.1109/ACCESS.2024.3429632], articleTitle=Detection of defects in wind turbin blade based on cascaded adaptive hybrid attention network, refAbstract=null), Reference(id=1322960983664239470, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2025, volume=62, issue=10, pageStart=#1000005, pageEnd=null, url=null, language=null, rfNumber=null, rfOrder=126, authorNames=Zhong S, Wang L P, journalName=Laser and Optoelectronics Progress, refType=null, unstructuredReference=Zhong S and Wang L P. 2025. Review of research on object detection in UAV aerial images. Laser and Optoelectronics Progress,62(10): #1000005, articleTitle=Review of research on object detection in UAV aerial images, refAbstract=null), Reference(id=1322960983735542639, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2025, volume=62, issue=10, pageStart=null, pageEnd=1000005, url=null, language=null, rfNumber=null, rfOrder=127, authorNames=钟帅, 王丽萍, journalName=激光与光电子学进展, refType=null, unstructuredReference=钟帅,王丽萍. 2025. 无人机航拍图像目标检测技术研究综述. 激光与光电子学进展,62(10): #1000005 [DOI: 10.3788/LOP242120], articleTitle=无人机航拍图像目标检测技术研究综述, refAbstract=null), Reference(id=1322960983823623024, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=13, issue=6, pageStart=1235, pageEnd=1243, url=null, language=null, rfNumber=null, rfOrder=128, authorNames=Zhou L M, Li S X, Zhu Z R, Chen F K, Liu C, Meng F R, journalName=Computer Science and Application, refType=null, unstructuredReference=Zhou L M,Li S X,Zhu Z R,Chen F K,Liu C and Meng F R. 2023. A review of transmission line defect detection research based on deep learning. Computer Science and Application,13(6): 1235-1243, articleTitle=A review of transmission line defect detection research based on deep learning, refAbstract=null), Reference(id=1322960983966229361, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=13, issue=6, pageStart=1235, pageEnd=1243, url=null, language=null, rfNumber=null, rfOrder=129, authorNames=周立明, 李士心, 朱致仁, 陈范凯, 刘宸, 孟范润, journalName=计算机科学与应用, refType=null, unstructuredReference=周立明,李士心,朱致仁,陈范凯,刘宸,孟范润. 2023. 基于深度学习的输电线路缺陷检测研究综述. 计算机科学与应用,13(6): 1235-1243 [DOI: 10.12677/CSA.2023.136121], articleTitle=基于深度学习的输电线路缺陷检测研究综述, refAbstract=null), Reference(id=1322960984045921138, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2023, volume=72, issue=null, pageStart=null, pageEnd=3509208, url=null, language=null, rfNumber=null, rfOrder=130, authorNames=Zhou W J, Wang Z J, Zhang M S, Wang L W, journalName=IEEE Transactions on Instrumentation and Measurement, refType=null, unstructuredReference=Zhou W J,Wang Z J,Zhang M S and Wang L W. 2023. Wind turbine actual defects detection based on visible and infrared image fusion. IEEE Transactions on Instrumentation and Measurement,72: #3509208 [DOI: 10.1109/TIM.2023.3251413], articleTitle=Wind turbine actual defects detection based on visible and infrared image fusion, refAbstract=null), Reference(id=1322960984125612915, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2022, volume=328, issue=null, pageStart=null, pageEnd=120241, url=null, language=null, rfNumber=null, rfOrder=131, authorNames=Zhu X X, Hang X Y, Gao X X, Yang X, Xu Z X, Wang Y, journalName=Applied Energy, refType=null, unstructuredReference=Zhu X X,Hang X Y,Gao X X,Yang X,Xu Z X,Wang Y,et al. 2022. Research on crack detection method of wind turbine blade based on a deep learning method. Applied Energy,328: #120241 [DOI: 10.1016/j.apenergy.2022.120241], articleTitle=Research on crack detection method of wind turbine blade based on a deep learning method, refAbstract=null), Reference(id=1322960984196916084, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2024, volume=14, issue=19, pageStart=null, pageEnd=8763, url=null, language=null, rfNumber=null, rfOrder=132, authorNames=Zou L, Chen A Q, Li C Z, Yang X H, Sun Y B, journalName=Applied Sciences, refType=null, unstructuredReference=Zou L,Chen A Q,Li C Z,Yang X H and Sun Y B. 2024. DCW-YOLO: an improved method for surface damage detection of wind turbine blades. Applied Sciences,14(19): #8763 [DOI: 10.3390/app14198763], articleTitle=DCW-YOLO: an improved method for surface damage detection of wind turbine blades, refAbstract=null), Reference(id=1322960984284996469, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2025, volume=15, issue=1, pageStart=null, pageEnd=5833, url=null, language=null, rfNumber=null, rfOrder=133, authorNames=Zou L, Chen A Q, Yang X H, Sun Y B, journalName=Scientific Reports, refType=null, unstructuredReference=Zou L,Chen A Q,Yang X H and Sun Y B. 2025. An improved method of AUD-YOLO for surface damage detection of wind turbine blades. Scientific Reports,15(1): #5833 [DOI: 10.1038/s41598-025-89864-7], articleTitle=An improved method of AUD-YOLO for surface damage detection of wind turbine blades, refAbstract=null), Reference(id=1322960984356299638, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, doi=null, pmid=null, pmcid=null, year=2022, volume=12, issue=18, pageStart=#9338, pageEnd=null, url=null, language=null, rfNumber=null, rfOrder=134, authorNames=Zou L, Cheng H, journalName=Applied Sciences, refType=null, unstructuredReference=Zou L and Cheng H,2022. Research on wind turbine blade surface damage identification based on improved convolution neural network. Applied Sciences,12(18): #9338 [DOI: 10.3390/app12189338], articleTitle=Research on wind turbine blade surface damage identification based on improved convolution neural network, refAbstract=null)], funds=[Fund(id=1322960968669602539, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, awardId=null, language=EN, fundingSource=Fundamental Research Funds for the Central Universities (2024MS112) ;The S&T Program of Hebei (SZX2020034), fundOrder=null, country=null), Fund(id=1322960968745100012, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, awardId=null, language=CN, fundingSource=中央高校基本科研业务费专项资金资助(2024MS112);河北省省级科技计划资助(SZX2020034), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1322960962290066066, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=1, ext=[AuthorCompanyExt(id=1322960962298454675, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962290066066, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Yanzhao Electric Power Laboratory of North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962306843284, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962290066066, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 华北电力大学燕赵电力实验室,保定 071003)]), AuthorCompany(id=1322960962361369237, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=2, ext=[AuthorCompanyExt(id=1322960962369757846, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962378146455, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 华北电力大学控制与计算机工程学院,保定 071003)]), AuthorCompany(id=1322960962445255320, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=3, ext=[AuthorCompanyExt(id=1322960962453643929, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962445255320, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 Hebei Key Laboratory of Knowledge Computing for Energy and Power,Baoding 071003,China), AuthorCompanyExt(id=1322960962462032538, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962445255320, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 河北省能源电力知识计算重点实验室,保定 071003)]), AuthorCompany(id=1322960962520752795, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=4, ext=[AuthorCompanyExt(id=1322960962529141404, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962520752795, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4 Engineering Research Center of Intelligent Computing for Complex Energy Systems,Ministry of Education,Baoding 071003,China), AuthorCompanyExt(id=1322960962537530013, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962520752795, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4 复杂能源系统智能计算教育部工程研究中心,保定 071003)]), AuthorCompany(id=1322960962596250270, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=5, ext=[AuthorCompanyExt(id=1322960962604638879, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962596250270, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=5 Hebei Key Laboratory of Power Internet of Things Technology,North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962608833184, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962596250270, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=5 河北省电力物联网技术重点实验室,保定 071003)])], figs=[ArticleFig(id=1322960966400484047, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, label=Fig.1, caption=Common NDT techniques for WTB defect detection, figureFileSmall=d93LU6IJ9B3BgCh01nAnJQ==, figureFileBig=76Xkm2q+QfLthlax0eB85g==, tableContent=null), ArticleFig(id=1322960966463398608, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, label=图1, caption=常见WTB缺陷的NDT技术, figureFileSmall=d93LU6IJ9B3BgCh01nAnJQ==, figureFileBig=76Xkm2q+QfLthlax0eB85g==, tableContent=null), ArticleFig(id=1322960966618587857, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, label=Fig.2, caption=Schematic of the paper structure, figureFileSmall=lwWKRQLvWRYBYy163L6QpA==, figureFileBig=YoV74Et0AjxzEq17DJYAzA==, tableContent=null), ArticleFig(id=1322960966681502418, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, label=图2, caption=整体研究结构示意, figureFileSmall=lwWKRQLvWRYBYy163L6QpA==, figureFileBig=YoV74Et0AjxzEq17DJYAzA==, tableContent=null), ArticleFig(id=1322960966752805587, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, label=Fig.3, caption=Typical defects in wind turbine blades ((a) crack;, figureFileSmall=+16NHK19MJQSBqzd0ctVlg==, figureFileBig=v9pFcs8i6zUz0NMaxqTlwA==, tableContent=null), ArticleFig(id=1322960966824108756, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, label=图3, caption=典型风机叶片缺陷

(b) erosion;(c) pinholes;(d) coat peeling;(e) icing;

(f) lightning damage;(g) gelcoat peeling;(h) blistering)

, figureFileSmall=+16NHK19MJQSBqzd0ctVlg==, figureFileBig=v9pFcs8i6zUz0NMaxqTlwA==, tableContent=null), ArticleFig(id=1322960966928966357, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, label=Fig.4, caption=Employ SRCNN for image quality improvement (Sarkar and Gunturi,2021), figureFileSmall=I5vX0tmWc9aiXY05E3dPuQ==, figureFileBig=528hU97/hCOy3k+lTLym7Q==, tableContent=null), ArticleFig(id=1322960967021241046, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, label=图4, caption=利用SRCNN提高图像质量(Sarkar和Gunturi,2021), figureFileSmall=I5vX0tmWc9aiXY05E3dPuQ==, figureFileBig=528hU97/hCOy3k+lTLym7Q==, tableContent=null), ArticleFig(id=1322960967105127127, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, label=Fig.5, caption=Milestones in the development of object detection networks, figureFileSmall=bU8GwtxTrxNucR1xo5rLdQ==, figureFileBig=eQWQPhlEuaS14CXJy6orOw==, tableContent=null), ArticleFig(id=1322960967176430296, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, label=图5, caption=目标检测网络发展脉络, figureFileSmall=bU8GwtxTrxNucR1xo5rLdQ==, figureFileBig=eQWQPhlEuaS14CXJy6orOw==, tableContent=null), ArticleFig(id=1322960967264510681, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, label=Fig.6, caption=Representative learning strategies for wind turbine blade defect detection ((a) transfer learning;, figureFileSmall=JXVO+fWZs34HJjwQIqVc6w==, figureFileBig=mn/eDVYvtmZQeBpco/50Tg==, tableContent=null), ArticleFig(id=1322960967335813850, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, label=图6, caption=WTB缺陷检测中常见学习策略

(b) ensemble learning;(c) cascade strategy;(d) unsupervised learning;(e) semi-supervised learning)

, figureFileSmall=JXVO+fWZs34HJjwQIqVc6w==, figureFileBig=mn/eDVYvtmZQeBpco/50Tg==, tableContent=null), ArticleFig(id=1322960967407117019, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, label=Table 1, caption=Comparison of recent reviews on wind turbine blade defect detection, figureFileSmall=null, figureFileBig=null, tableContent=
方法涉及到的检测技术优点不足
Du等人(2020)传感器技术、信号处理与数据融合强调数据融合和信号处理技术,关注多源数据结合与智能化处理深度学习和无人机检测方法覆盖有限
Nijssen和Manrique(2020)综合多种传感和检测技术分类明确,区分结构与非结构损伤以理论研究为主,缺少具体应用案例指导
Katsaprakakis等人(2021)声学传感器、振动、应变传感器等系统分析各种叶片损伤类型及其机理,涵盖多种检测方法主要聚焦传统检测技术,缺少对基于深度学习等智能检测技术的详细探讨;对最新传感器融合技术关注有限
Wang等人(2022)传感器融合、图像处理关注大型风机叶片,结合最新传感技术与图像分析缺少对算法性能指标的量化比较,且对智能检测方法的深入研究不足
Zhang和Shu(2024)UAV视觉检测、图像处理结合最新视觉算法,覆盖应用场景广泛主要集中于表面视觉检测,对深层结构损伤和多模态传感融合关注较少
), ArticleFig(id=1322960967474225884, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, label=表1, caption=

WTB缺陷检测文献综述对比

, figureFileSmall=null, figureFileBig=null, tableContent=
方法涉及到的检测技术优点不足
Du等人(2020)传感器技术、信号处理与数据融合强调数据融合和信号处理技术,关注多源数据结合与智能化处理深度学习和无人机检测方法覆盖有限
Nijssen和Manrique(2020)综合多种传感和检测技术分类明确,区分结构与非结构损伤以理论研究为主,缺少具体应用案例指导
Katsaprakakis等人(2021)声学传感器、振动、应变传感器等系统分析各种叶片损伤类型及其机理,涵盖多种检测方法主要聚焦传统检测技术,缺少对基于深度学习等智能检测技术的详细探讨;对最新传感器融合技术关注有限
Wang等人(2022)传感器融合、图像处理关注大型风机叶片,结合最新传感技术与图像分析缺少对算法性能指标的量化比较,且对智能检测方法的深入研究不足
Zhang和Shu(2024)UAV视觉检测、图像处理结合最新视觉算法,覆盖应用场景广泛主要集中于表面视觉检测,对深层结构损伤和多模态传感融合关注较少
), ArticleFig(id=1322960967549723357, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, label=Table 2, caption=Typical defects in wind turbine blades, figureFileSmall=null, figureFileBig=null, tableContent=
损伤名称位置形态学特征产生原因对应内部损伤对风机的影响
裂纹前缘、后缘或表面线性或网状,尖端延伸方向与所受应力有关机械应力、环境因素制造缺陷、疲劳损伤、应力损伤、载荷损伤结构受损,影响捕风能力
侵蚀表面材料磨损或剥落,常见于前缘不规则鳞片状,从微小点状剥落逐渐扩大,边缘部分可能翘起长期暴露于风沙、盐雾等侵蚀环境-捕风能力下降
砂眼表面出现小砂眼,可能分布于整个叶片表面圆形或近似圆形的小孔,继续发展易导致胶层大范围脱落制造过程中的缺陷或长期风沙侵蚀制造缺陷捕风能力下降
白斑表面出现白色斑点,可能分布于整个叶片表面圆形、椭圆形、雪花状,易导致涂层的局部脱落材料老化、气流冲击、紫外线照射或表面微小损伤腐蚀损伤表层强度下降、微小裂纹形成
积冰表面积冰,常见于低温环境下的叶片前缘透明或半透明,片状、层状湿冷气候,水汽冷凝冻结-增重失衡,风能转化效率下降
雷击损伤-表面黑灰化、裂纹、断裂等雷击-结构受损,影响捕风能力
脱胶损伤粘接区域脱胶,见于叶片上下壳体、壳体与梁的粘接区域表面凸起或脱落,颜色存在明显差异,圆形、椭圆、不规则形状制造缺陷或长期运行中的机械应力导致制造缺陷、应力损伤影响叶片整体结构稳定,粘接强度下降,结构松动
污垢、油渍叶片表面表面有不规则油迹、污垢附着空气中灰尘、油污等颗粒附着-表面粗糙度增加,气动性能下降
涂层脱落叶片表面,常见于前缘和外层胶衣处表面局部剥落、出现裸露基体紫外老化、雨滴冲刷、冰雹或颗粒物冲击-叶片易腐蚀、气动性能下降
尖端开裂叶片尖端沿纵向或横向出现裂纹气动载荷集中、雷击等冲击制造缺陷、内部分层叶片强度下降,裂纹易扩展导致断裂
锈蚀叶片金属部件,如连接螺栓、防雷系统金属部位金属表面出现褐色锈蚀、点蚀、剥落潮湿、高盐、雨水或结露环境-影响结构连接的可靠性,降低防雷性能
导航漆脱落特定位置表面局部脱落、褪色或不完整长期紫外线照射、雨滴或沙尘冲蚀,涂层老化-降低叶片的可识别度和完全性,裸露区域可能加速表层老化
), ArticleFig(id=1322960967646192350, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, label=表2, caption=

典型风机叶片缺陷

, figureFileSmall=null, figureFileBig=null, tableContent=
损伤名称位置形态学特征产生原因对应内部损伤对风机的影响
裂纹前缘、后缘或表面线性或网状,尖端延伸方向与所受应力有关机械应力、环境因素制造缺陷、疲劳损伤、应力损伤、载荷损伤结构受损,影响捕风能力
侵蚀表面材料磨损或剥落,常见于前缘不规则鳞片状,从微小点状剥落逐渐扩大,边缘部分可能翘起长期暴露于风沙、盐雾等侵蚀环境-捕风能力下降
砂眼表面出现小砂眼,可能分布于整个叶片表面圆形或近似圆形的小孔,继续发展易导致胶层大范围脱落制造过程中的缺陷或长期风沙侵蚀制造缺陷捕风能力下降
白斑表面出现白色斑点,可能分布于整个叶片表面圆形、椭圆形、雪花状,易导致涂层的局部脱落材料老化、气流冲击、紫外线照射或表面微小损伤腐蚀损伤表层强度下降、微小裂纹形成
积冰表面积冰,常见于低温环境下的叶片前缘透明或半透明,片状、层状湿冷气候,水汽冷凝冻结-增重失衡,风能转化效率下降
雷击损伤-表面黑灰化、裂纹、断裂等雷击-结构受损,影响捕风能力
脱胶损伤粘接区域脱胶,见于叶片上下壳体、壳体与梁的粘接区域表面凸起或脱落,颜色存在明显差异,圆形、椭圆、不规则形状制造缺陷或长期运行中的机械应力导致制造缺陷、应力损伤影响叶片整体结构稳定,粘接强度下降,结构松动
污垢、油渍叶片表面表面有不规则油迹、污垢附着空气中灰尘、油污等颗粒附着-表面粗糙度增加,气动性能下降
涂层脱落叶片表面,常见于前缘和外层胶衣处表面局部剥落、出现裸露基体紫外老化、雨滴冲刷、冰雹或颗粒物冲击-叶片易腐蚀、气动性能下降
尖端开裂叶片尖端沿纵向或横向出现裂纹气动载荷集中、雷击等冲击制造缺陷、内部分层叶片强度下降,裂纹易扩展导致断裂
锈蚀叶片金属部件,如连接螺栓、防雷系统金属部位金属表面出现褐色锈蚀、点蚀、剥落潮湿、高盐、雨水或结露环境-影响结构连接的可靠性,降低防雷性能
导航漆脱落特定位置表面局部脱落、褪色或不完整长期紫外线照射、雨滴或沙尘冲蚀,涂层老化-降低叶片的可识别度和完全性,裸露区域可能加速表层老化
), ArticleFig(id=1322960967725884127, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, label=Table 3, caption=Comparison of object detection tasks in power system, figureFileSmall=null, figureFileBig=null, tableContent=
检测任务检测目标主要检测难点公开数据集数量
输电线异物检测(周立明 等,2023)鸟巢等异物,腐蚀、破损等损伤异物形状多样性、尺度差异大、环境复杂、相似度高较多
变压器油泄露检测(Lu等,2020)油渍目标边缘不明显较少
绝缘子缺陷检测(刘传洋 等,2025)绝缘子破损、脱落等背景复杂、目标多尺度较多
风机叶片表面检测裂纹、侵蚀、积冰等缺陷形状多样性、背景复杂多变、多尺度目标、积冰等特征不明显较少
), ArticleFig(id=1322960967818158816, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, label=表3, caption=

电力系统中目标检测任务对比

, figureFileSmall=null, figureFileBig=null, tableContent=
检测任务检测目标主要检测难点公开数据集数量
输电线异物检测(周立明 等,2023)鸟巢等异物,腐蚀、破损等损伤异物形状多样性、尺度差异大、环境复杂、相似度高较多
变压器油泄露检测(Lu等,2020)油渍目标边缘不明显较少
绝缘子缺陷检测(刘传洋 等,2025)绝缘子破损、脱落等背景复杂、目标多尺度较多
风机叶片表面检测裂纹、侵蚀、积冰等缺陷形状多样性、背景复杂多变、多尺度目标、积冰等特征不明显较少
), ArticleFig(id=1322960967918822113, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, label=Table 4, caption=Representative image preprocessing techniques in WTB damage detection, figureFileSmall=null, figureFileBig=null, tableContent=
分类方法任务
图像质量改善去噪、去模糊、超分辨率改善图像清晰度、细节和噪声水平,提升图像的视觉效果和后续处理的基础质量
图像几何校正图像拼接、图像配准解决图像之间的空间关系,保证多视角、多时刻图像间的对齐与融合
图像结构与提取图像分割、图像分解与融合分离目标与背景,融合信息,提升语义理解和后续分析能力
), ArticleFig(id=1322960968006902498, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, label=表4, caption=

WTB缺陷检测中常见预处理方法

, figureFileSmall=null, figureFileBig=null, tableContent=
分类方法任务
图像质量改善去噪、去模糊、超分辨率改善图像清晰度、细节和噪声水平,提升图像的视觉效果和后续处理的基础质量
图像几何校正图像拼接、图像配准解决图像之间的空间关系,保证多视角、多时刻图像间的对齐与融合
图像结构与提取图像分割、图像分解与融合分离目标与背景,融合信息,提升语义理解和后续分析能力
), ArticleFig(id=1322960968078205667, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, label=Table 5, caption=Summary of improvement methods for YOLOv5 model, figureFileSmall=null, figureFileBig=null, tableContent=
文献主干网络修改Neck损失函数mAP参数量/M

运算量/

检测速度

Zhang等人(2022)CBAM注意力机制4层微尺度检测层-@0.5: 0.951(↑0.115)37.841.3帧/s
Ran等人(2022)CA注意力机制双向FPNEIoU@0.5: 83.7(↑3.2)11.663.4帧/s
Zhu等人(2022)MobileNetv3 + GhostNet-Alpha-IoU@0.5: 93.2(↑2.0)--
Zhang等人(2023)MobileNetv3双向FPNEIoU@0.5: 90.15(↑5.51)8.471.15帧/s
Dai(2024)--CIoU---
Liu等人(2024b)Ghost卷积 + GhostBottleNeckv2 CBAMBiFPN-@0.5: 94.56(↑5.05)6.87.3 GFLOPs
Lei等人(2024)--动态标签分配策略@0.5: 0.893(↑0.11)0.138帧/s *
Ye等人(2024)第1个卷积层替换为DConv; 其余替换为GSConv

@0.5: 0.792(↑0.029),

@0.5-0.95: 0.457(↑0.010)

7.1115.0 GFLOPs
), ArticleFig(id=1322960968141120228, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, label=表5, caption=

YOLOv5模型的改进方法汇总

, figureFileSmall=null, figureFileBig=null, tableContent=
文献主干网络修改Neck损失函数mAP参数量/M

运算量/

检测速度

Zhang等人(2022)CBAM注意力机制4层微尺度检测层-@0.5: 0.951(↑0.115)37.841.3帧/s
Ran等人(2022)CA注意力机制双向FPNEIoU@0.5: 83.7(↑3.2)11.663.4帧/s
Zhu等人(2022)MobileNetv3 + GhostNet-Alpha-IoU@0.5: 93.2(↑2.0)--
Zhang等人(2023)MobileNetv3双向FPNEIoU@0.5: 90.15(↑5.51)8.471.15帧/s
Dai(2024)--CIoU---
Liu等人(2024b)Ghost卷积 + GhostBottleNeckv2 CBAMBiFPN-@0.5: 94.56(↑5.05)6.87.3 GFLOPs
Lei等人(2024)--动态标签分配策略@0.5: 0.893(↑0.11)0.138帧/s *
Ye等人(2024)第1个卷积层替换为DConv; 其余替换为GSConv

@0.5: 0.792(↑0.029),

@0.5-0.95: 0.457(↑0.010)

7.1115.0 GFLOPs
), ArticleFig(id=1322960968208229093, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, label=Table 6, caption=Summary of improvement methods for YOLOv8 model, figureFileSmall=null, figureFileBig=null, tableContent=
文献主干网络修改Neck损失函数mAP参数量/M

运算量/

检测速度

李冰等人(2024)LSK Attention

BiFPN

FasterBlock

Inner IoU

@0.5: 76.6(↑3.3)

@0.5-0.95: 46.5(↑2.4)

2.26.9 GFLOPs
Yu等人(2024)CBAM AttentionBiFPNEIoU

@0.5: 83.3(↑1.4)

@0.5-0.95: 54.7(↑1.1)

3.048.3 GFLOPs
Wang等人(2024)PConv-EfficiCIoU@0.5: 0.995(↑0.026)2.077.0帧/s
Liu等人(2024a)Focal Block

ResNet-EMA

Slim-Neck

-@0.5: 79.9(↑3.0)3.017.8 GFLOPs (0.9)
Yang等人(2024)

FasterNetBlock

C2f-EMA

GMF-SPPF-

@0.5: 86.6(↑4.3)

@0.5-0.95: 64(↑7.5)

8.7116.3帧/s
Zou等人(2024)-

DSConv

CARAFE

WIoUv3

@0.5: 0.937(↑0.23)

@0.5-0.95: 0.712(↑0.023)

3.358.5 GFLOPS
Ma等人(2024)

MKGG

ECA

SBiFPN-@0.5: 0.855(↑0.017)2.1-
Zhang等人(2024)CA AttentionRep-GFPNInner-WIoU@0.5: 92.0(↑5.5)3.2120.5帧/s
Cao等人(2024)

MLCA Atteniton

焦点特征

聚合模块

AFPNInner-IoU

@0.5: 78.6(↑4)

@0.5-0.95: 51.7(↑5.3)

-8.5 GFLOPs
Sun等人(2024)RFCBAMConv

CAA-HSFPN

共享卷积;任务对齐

结构的Head

-

@0.5: 78.6

@0.5-0.95: 51.7

1.828.5 GFLOPs
Zou等人(2025)

UniRepLKNet

ADdown

DySample-

@0.5: 0.92(↑0.03)

@0.5-0.95: 0.687(↑0.062)

2.537.2 GFLOPs
), ArticleFig(id=1322960968287920870, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, label=表6, caption=

YOLOv8模型的改进方法汇总

, figureFileSmall=null, figureFileBig=null, tableContent=
文献主干网络修改Neck损失函数mAP参数量/M

运算量/

检测速度

李冰等人(2024)LSK Attention

BiFPN

FasterBlock

Inner IoU

@0.5: 76.6(↑3.3)

@0.5-0.95: 46.5(↑2.4)

2.26.9 GFLOPs
Yu等人(2024)CBAM AttentionBiFPNEIoU

@0.5: 83.3(↑1.4)

@0.5-0.95: 54.7(↑1.1)

3.048.3 GFLOPs
Wang等人(2024)PConv-EfficiCIoU@0.5: 0.995(↑0.026)2.077.0帧/s
Liu等人(2024a)Focal Block

ResNet-EMA

Slim-Neck

-@0.5: 79.9(↑3.0)3.017.8 GFLOPs (0.9)
Yang等人(2024)

FasterNetBlock

C2f-EMA

GMF-SPPF-

@0.5: 86.6(↑4.3)

@0.5-0.95: 64(↑7.5)

8.7116.3帧/s
Zou等人(2024)-

DSConv

CARAFE

WIoUv3

@0.5: 0.937(↑0.23)

@0.5-0.95: 0.712(↑0.023)

3.358.5 GFLOPS
Ma等人(2024)

MKGG

ECA

SBiFPN-@0.5: 0.855(↑0.017)2.1-
Zhang等人(2024)CA AttentionRep-GFPNInner-WIoU@0.5: 92.0(↑5.5)3.2120.5帧/s
Cao等人(2024)

MLCA Atteniton

焦点特征

聚合模块

AFPNInner-IoU

@0.5: 78.6(↑4)

@0.5-0.95: 51.7(↑5.3)

-8.5 GFLOPs
Sun等人(2024)RFCBAMConv

CAA-HSFPN

共享卷积;任务对齐

结构的Head

-

@0.5: 78.6

@0.5-0.95: 51.7

1.828.5 GFLOPs
Zou等人(2025)

UniRepLKNet

ADdown

DySample-

@0.5: 0.92(↑0.03)

@0.5-0.95: 0.687(↑0.062)

2.537.2 GFLOPs
), ArticleFig(id=1322960968363418343, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, label=Table 7, caption=Summary of non-public datasets in the references, figureFileSmall=null, figureFileBig=null, tableContent=
文献缺陷类别样本数量分辨率/像素
Wang等人(2019b)损伤1605 280 × 2 970
Shihavuddin和Chen (2019)前缘侵蚀;涡流发生器面板;涡流发生器缺齿;雷击传感器458-
Xu等人(2019)导航漆脱落;胶衣脱落;表面侵蚀;油污25 7735 472 × 3 684
Qiu等人(2019)裂纹;油污;砂眼4 3691 920 × 1 080
Yu等人(2020)裂纹;胶衣脱落;刮痕;砂眼;后缘边缘损伤171-
Reddy等人(2019)Data1: 损伤1 675-
Data2: 损伤(天空;雷击损伤;材料损伤;尖端开裂;侵蚀;边缘侵蚀;裂纹)1 647-
Mao等人(2021)裂纹;破损;油污;复合损伤1 164300 × 500; 608 × 608; 224 × 224
Zhang等人(2021)裂纹;侵蚀;孔隙;其他1912 592 × 1 936
Yang等人(2021)损伤1 350224 × 224
Guo等人(2021)材料损伤;裂纹;涂层损伤;边缘侵蚀725高分辨率
Hacıefendioğlu等人(2022)积冰278-
Zhu等人(2022)损伤513640 × 512
Zhang和Wen(2022)砂眼;裂纹;侵蚀;涂层脱落4005 184 × 3 456
Zou和Cheng(2022)裂纹;表面脱落4 3965 472 × 3 684
Ran等人(2022)-559640 × 640
Zhang等人(2022)剥落;点蚀;裂纹;污染1 952-
Lyu等人(2022)胶衣脱落;裂纹;侵蚀10 0004 000 × 3 000
Diaz和Tittus(2023)累计损伤;污渍;前缘侵蚀;缺齿;破损;裂纹;其他701-
Dong等人(2023)污渍;划痕;孔隙198-
Yao等人(2023)污渍;裂纹;破损7252 560 × 1 920
Gao等人(2023)涂层;裂纹2 577-
Zhang等人(2023)裂纹;污渍;孔洞;侵蚀;锈蚀;雷击7 995319 × 312
Yu等人(2023a)-312640 × 512
Zhou等人(2023)损伤1 040RGB 6 400 × 6 400;IR 640 × 512
Wang等人(2024)表面污垢;前缘侵蚀;涂层脱落;油渍2 916-
Liu等人(2024)剥落腐蚀;砂眼;油污;裂纹;渗透开裂8512 048 × 1 536
Yu等人(2024)破损;裂纹;划痕4 500-
Yang等人(2024)裂纹;雷击;砂眼;侵蚀;表面损伤6 072-
Liao等人(2024)裂纹;砂眼;破损2 995586 × 371
Fu等人(2024)油污;裂纹13 642649 × 640
Cao等人(2024)胶衣脱落;裂纹;表面侵蚀7841 200 × 900
Zhang等人(2024)裂纹;孔隙;边缘侵蚀;涂层脱落890-
Ma等人(2024)损伤2 990-
), ArticleFig(id=1322960968434721512, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, label=表7, caption=

文献中涉及的私有数据集

, figureFileSmall=null, figureFileBig=null, tableContent=
文献缺陷类别样本数量分辨率/像素
Wang等人(2019b)损伤1605 280 × 2 970
Shihavuddin和Chen (2019)前缘侵蚀;涡流发生器面板;涡流发生器缺齿;雷击传感器458-
Xu等人(2019)导航漆脱落;胶衣脱落;表面侵蚀;油污25 7735 472 × 3 684
Qiu等人(2019)裂纹;油污;砂眼4 3691 920 × 1 080
Yu等人(2020)裂纹;胶衣脱落;刮痕;砂眼;后缘边缘损伤171-
Reddy等人(2019)Data1: 损伤1 675-
Data2: 损伤(天空;雷击损伤;材料损伤;尖端开裂;侵蚀;边缘侵蚀;裂纹)1 647-
Mao等人(2021)裂纹;破损;油污;复合损伤1 164300 × 500; 608 × 608; 224 × 224
Zhang等人(2021)裂纹;侵蚀;孔隙;其他1912 592 × 1 936
Yang等人(2021)损伤1 350224 × 224
Guo等人(2021)材料损伤;裂纹;涂层损伤;边缘侵蚀725高分辨率
Hacıefendioğlu等人(2022)积冰278-
Zhu等人(2022)损伤513640 × 512
Zhang和Wen(2022)砂眼;裂纹;侵蚀;涂层脱落4005 184 × 3 456
Zou和Cheng(2022)裂纹;表面脱落4 3965 472 × 3 684
Ran等人(2022)-559640 × 640
Zhang等人(2022)剥落;点蚀;裂纹;污染1 952-
Lyu等人(2022)胶衣脱落;裂纹;侵蚀10 0004 000 × 3 000
Diaz和Tittus(2023)累计损伤;污渍;前缘侵蚀;缺齿;破损;裂纹;其他701-
Dong等人(2023)污渍;划痕;孔隙198-
Yao等人(2023)污渍;裂纹;破损7252 560 × 1 920
Gao等人(2023)涂层;裂纹2 577-
Zhang等人(2023)裂纹;污渍;孔洞;侵蚀;锈蚀;雷击7 995319 × 312
Yu等人(2023a)-312640 × 512
Zhou等人(2023)损伤1 040RGB 6 400 × 6 400;IR 640 × 512
Wang等人(2024)表面污垢;前缘侵蚀;涂层脱落;油渍2 916-
Liu等人(2024)剥落腐蚀;砂眼;油污;裂纹;渗透开裂8512 048 × 1 536
Yu等人(2024)破损;裂纹;划痕4 500-
Yang等人(2024)裂纹;雷击;砂眼;侵蚀;表面损伤6 072-
Liao等人(2024)裂纹;砂眼;破损2 995586 × 371
Fu等人(2024)油污;裂纹13 642649 × 640
Cao等人(2024)胶衣脱落;裂纹;表面侵蚀7841 200 × 900
Zhang等人(2024)裂纹;孔隙;边缘侵蚀;涂层脱落890-
Ma等人(2024)损伤2 990-
), ArticleFig(id=1322960968506024681, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=EN, label=Table 8, caption=Summary of performance evaluation metrics, figureFileSmall=null, figureFileBig=null, tableContent=
评估指标定义公式
图像质量峰值信噪比通过计算原始图像与失真图像之间的均方误差(mean square error,MSE)衡量失真程度,再通过信号最大功率与噪声功率的比值进行评估
PSNR=10logMAXI2MSE
结构相似度指数从亮度(luminance)、对比度(contrast)和结构(structure)3个方面衡量恢复前后图像的相似度
SSIMx, y=lx, yαcx, yβsx, yγ
分类精度准确率(accuracy)预测值中,预测正确占比
Acc=TP+TNTP+FP+FN+TN
精确率(precision)在所有预测为正例的结果中,预测为正例且预测正确的占比
Precision=TPTP+FP
召回率(recall)在所有正例中,预测为正例且预测正确的占比
Recall=TPTP+FN
F1-score用于平衡precision和recall,使得模型更加稳定
F1=2Precision×RecallPrecision+Recall
平均准确率(AP)用于衡量单类别的模型平均准确度
AP=∑i=1n-1ri+1-riPinterri+1
平均精度均值(mAP)多类别分类时,各类AP的均值
mAP=∑i=1kAPik
定位精度像素准确率图像中正确分类的像素百分比
PA=∑i=0npii∑i=0n∑j=0npij
交并比预测边界框与真实标签区域之间的重叠区域除以预测分割与标签之间的联合区域
IoU=A⋂BA⋃B
Dice系数两倍的交集除以像素和
Dice=2A⋂BA+B
模型参数量模型中可训练参数的总数-
检测速度FPS每秒内可以处理的图像数量或者处理一幅图像所需时间
FPS =帧数/时间
GFLOPs模型在进行一次前行传播时的浮点运算次数-
), ArticleFig(id=1322960968568939242, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, language=CN, label=表8, caption=

性能评估指标总结

, figureFileSmall=null, figureFileBig=null, tableContent=
评估指标定义公式
图像质量峰值信噪比通过计算原始图像与失真图像之间的均方误差(mean square error,MSE)衡量失真程度,再通过信号最大功率与噪声功率的比值进行评估
PSNR=10logMAXI2MSE
结构相似度指数从亮度(luminance)、对比度(contrast)和结构(structure)3个方面衡量恢复前后图像的相似度
SSIMx, y=lx, yαcx, yβsx, yγ
分类精度准确率(accuracy)预测值中,预测正确占比
Acc=TP+TNTP+FP+FN+TN
精确率(precision)在所有预测为正例的结果中,预测为正例且预测正确的占比
Precision=TPTP+FP
召回率(recall)在所有正例中,预测为正例且预测正确的占比
Recall=TPTP+FN
F1-score用于平衡precision和recall,使得模型更加稳定
F1=2Precision×RecallPrecision+Recall
平均准确率(AP)用于衡量单类别的模型平均准确度
AP=∑i=1n-1ri+1-riPinterri+1
平均精度均值(mAP)多类别分类时,各类AP的均值
mAP=∑i=1kAPik
定位精度像素准确率图像中正确分类的像素百分比
PA=∑i=0npii∑i=0n∑j=0npij
交并比预测边界框与真实标签区域之间的重叠区域除以预测分割与标签之间的联合区域
IoU=A⋂BA⋃B
Dice系数两倍的交集除以像素和
Dice=2A⋂BA+B
模型参数量模型中可训练参数的总数-
检测速度FPS每秒内可以处理的图像数量或者处理一幅图像所需时间
FPS =帧数/时间
GFLOPs模型在进行一次前行传播时的浮点运算次数-
)], attaches=null, journal=Journal(id=1249023527618129992, delFlag=0, nameCn=中国图象图形学报, nameEn=Journal of Image and Graphics, nameHistory1=null, nameHistory2=null, issn=1006-8961, eissn=null, cn=11-3758, coden=CODEN ZTTXFZ, 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=uirXtX858YS3zEpFXZttJA==, journalPrice=null, startedYear=null, abbrevIsoEn=Journal of Image and Graphics, journalRemark=null, publicationField=null, createdTime=1775720014721, updatedTime=1788951488513, createdBy=18614031015, updatedBy=13041195026, firstLetterCn=Z, firstLetterEn=Z, subjectCode=Electronic Information, subjectName=null, subjectCodeEn=Electronic Information, subjectNameEn=null, picCn=uirXtX858YS3zEpFXZttJA==, picEn=bud7qaxfvWHeFsbyBTAiKQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1304520351199814396, language=CN, name=中国图象图形学报, 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=1788951488542, updatedTime=1788951488542, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=, submissionAuthorUrl=https://journal.ids.fzyun.cn/auth/realms/journal/protocol/openid-connect/auth?client_id=journal-cjig-author&redirect_uri=https%3A%2F%2Fcjig.portal.founderss.cn%2Foauth%2Fcallback&response_type=code&scope=phone+openid+email+profile&state=e6369def-2842-41d8, submissionEditorUrl=https://journal.ids.fzyun.cn/auth/realms/journal/protocol/openid-connect/auth?client_id=journal-cjig-editor&redirect_uri=https%3A%2F%2Fcjigeditor.portal.founderss.cn%2Foauth%2Fcallback&response_type=code&scope=phone+openid+email+profile&state=9ccec05b-6bd, submissionReviewUrl=https://journal.ids.fzyun.cn/auth/realms/journal/protocol/openid-connect/auth?client_id=journal-cjig-author&redirect_uri=https%3A%2F%2Fcjig.portal.founderss.cn%2Foauth%2Fcallback&response_type=code&scope=phone+openid+email+profile&state=1e8a31c8-5434-4f78, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1304520351250146045, language=EN, name=Journal of Image and Graphics, 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=1788951488554, updatedTime=1788951488554, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=, submissionAuthorUrl=https://journal.ids.fzyun.cn/auth/realms/journal/protocol/openid-connect/auth?client_id=journal-cjig-author&redirect_uri=https%3A%2F%2Fcjig.portal.founderss.cn%2Foauth%2Fcallback&response_type=code&scope=phone+openid+email+profile&state=e6369def-2842-41d8, submissionEditorUrl=https://journal.ids.fzyun.cn/auth/realms/journal/protocol/openid-connect/auth?client_id=journal-cjig-editor&redirect_uri=https%3A%2F%2Fcjigeditor.portal.founderss.cn%2Foauth%2Fcallback&response_type=code&scope=phone+openid+email+profile&state=9ccec05b-6bd, submissionReviewUrl=https://journal.ids.fzyun.cn/auth/realms/journal/protocol/openid-connect/auth?client_id=journal-cjig-author&redirect_uri=https%3A%2F%2Fcjig.portal.founderss.cn%2Foauth%2Fcallback&response_type=code&scope=phone+openid+email+profile&state=1e8a31c8-5434-4f78, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1249024232475115590, websiteList=[Website(id=1249025782459334881, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1249024232475115590, 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/zgtxtxxb/CN, language=CN, createTime=1775720552315, createBy=18614031015, updateTime=1775720586268, updateBy=18614031015, name=中国图象图形学报-中文, tplId=1146099689490845704, title=中国图象图形学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1249026166254928133, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782459334881, code=articleTextType, value=kx, createTime=1775720643819, updateTime=1775720643819, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026166221373698, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782459334881, code=banner, value=null, createTime=1775720643811, updateTime=1775720643811, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026166271705352, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782459334881, code=grayFlag, value=0, createTime=1775720643823, updateTime=1775720643823, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026166212985089, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782459334881, code=logo, value=https://castjournals.cast.org.cn/joweb/zgtxtxxb/CN/file/pic?fileId=TDRjKTHfgAnvFKZaDA70wA==, createTime=1775720643809, updateTime=1775720643809, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026166288482570, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782459334881, code=minRunFlag, value=0, createTime=1775720643827, updateTime=1775720643827, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026166246539524, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782459334881, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/zgtxtxxb/CN/file/pic, createTime=1775720643817, updateTime=1775720643817, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026166280093961, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782459334881, code=silenceFlag, value=0, createTime=1775720643825, updateTime=1775720643825, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026166233956611, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782459334881, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1775720643814, updateTime=1775720643814, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026166259122438, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782459334881, code=themeColor, value=null, createTime=1775720643820, updateTime=1775720643820, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026166267511047, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782459334881, code=themeStyle, value=null, createTime=1775720643822, updateTime=1775720643822, creator=18614031015, updator=18614031015)]), Website(id=1249025782681633001, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1249024232475115590, 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/zgtxtxxb/EN, language=EN, createTime=1775720552368, createBy=18614031015, updateTime=1775720607118, updateBy=18614031015, name=中国图象图形学报-英文, tplId=1146101810881728533, title=Journal of Image and Graphics, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1249026195371786511, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782681633001, code=articleTextType, value=kx, createTime=1775720650761, updateTime=1775720650761, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026195355009292, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782681633001, code=banner, value=null, createTime=1775720650757, updateTime=1775720650757, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026195392758034, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782681633001, code=grayFlag, value=0, createTime=1775720650766, updateTime=1775720650766, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026195342426379, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782681633001, code=logo, value=https://castjournals.cast.org.cn/joweb/zgtxtxxb/EN/file/pic?fileId=TDRjKTHfgAnvFKZaDA70wA==, createTime=1775720650754, updateTime=1775720650754, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026195409535252, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782681633001, code=minRunFlag, value=0, createTime=1775720650770, updateTime=1775720650770, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026195367592206, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782681633001, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/zgtxtxxb/EN/file/pic, createTime=1775720650760, updateTime=1775720650760, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026195401146643, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782681633001, code=silenceFlag, value=0, createTime=1775720650768, updateTime=1775720650768, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026195359203597, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782681633001, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1775720650758, updateTime=1775720650758, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026195380175120, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782681633001, code=themeColor, value=null, createTime=1775720650763, updateTime=1775720650763, creator=18614031015, updator=18614031015), WebsiteProps(id=1249026195388563729, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1249025782681633001, code=themeStyle, value=null, createTime=1775720650765, updateTime=1775720650765, creator=18614031015, updator=18614031015)])], journalTitle=中国图象图形学报, weixinUrl=null, journalUrl=https://www.cjig.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Journal of Image and Graphics, journalPhotoCn=uirXtX858YS3zEpFXZttJA==, journalPhotoEn=bud7qaxfvWHeFsbyBTAiKQ==, journalFirstLetter=Z, 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/zgtxtxxb/CN/10.11834/jig.250458, detailUrlEn=https://castjournals.cast.org.cn/joweb/zgtxtxxb/EN/10.11834/jig.250458, pdfUrlCn=https://castjournals.cast.org.cn/joweb/zgtxtxxb/CN/PDF/10.11834/jig.250458, pdfUrlEn=https://castjournals.cast.org.cn/joweb/zgtxtxxb/EN/PDF/10.11834/jig.250458, aliStartDate=0, aliEndDate=0, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1786809600000, fullTextJson=null, articleText=null, reference=null)
收藏切换
风力发电机叶片缺陷目标检测研究综述
收藏切换
PDF下载
张弪弘 1, 2 , 苏攀 2, 3, * , 朱清源 2, 4 , 张超刚 2, 4 , 李冰 1, 2 , 曹旺斌 1, 5
中国图象图形学报 | 综述 2026,31(8): 2844-2867
收起
收藏切换
中国图象图形学报 |综述 2026 , 31 (8) : 2844 -2867
风力发电机叶片缺陷目标检测研究综述
全屏
[Author(id=1322960962684330658, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=17395714117@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1322960962772411045, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960962684330658, language=EN, stringName=Jinghong Zhang, firstName=Jinghong, middleName=null, lastName=Zhang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 Yanzhao Electric Power Laboratory of North China Electric Power University,Baoding 071003,China
2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1322960962852102822, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960962684330658, language=CN, stringName=张弪弘, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 华北电力大学燕赵电力实验室,保定 071003
2 华北电力大学控制与计算机工程学院,保定 071003, bio={"content":"

张弪弘,女,硕士研究生,主要研究方向为新能源与目标检测。E-mail:

"}, bioImg=null, bioContent=

张弪弘,女,硕士研究生,主要研究方向为新能源与目标检测。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1322960962290066066, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=1, ext=[AuthorCompanyExt(id=1322960962298454675, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962290066066, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Yanzhao Electric Power Laboratory of North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962306843284, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962290066066, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 华北电力大学燕赵电力实验室,保定 071003)]), AuthorCompany(id=1322960962361369237, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=2, ext=[AuthorCompanyExt(id=1322960962369757846, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962378146455, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 华北电力大学控制与计算机工程学院,保定 071003)])]), Author(id=1322960962910823080, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=supan@ncepu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1322960962986320555, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960962910823080, language=EN, stringName=Pan Su, firstName=Pan, middleName=null, lastName=Su, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, 3, *, address=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China
3 Hebei Key Laboratory of Knowledge Computing for Energy and Power,Baoding 071003,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1322960963066012332, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960962910823080, language=CN, stringName=苏攀, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, 3, *, address=2 华北电力大学控制与计算机工程学院,保定 071003
3 河北省能源电力知识计算重点实验室,保定 071003, bio={"content":"

苏攀,通信作者,男,副教授,主要研究方向为机器学习与医学图像处理。E-mail:

"}, bioImg=null, bioContent=

苏攀,通信作者,男,副教授,主要研究方向为机器学习与医学图像处理。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1322960962361369237, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=2, ext=[AuthorCompanyExt(id=1322960962369757846, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962378146455, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 华北电力大学控制与计算机工程学院,保定 071003)]), AuthorCompany(id=1322960962445255320, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=3, ext=[AuthorCompanyExt(id=1322960962453643929, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962445255320, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 Hebei Key Laboratory of Knowledge Computing for Energy and Power,Baoding 071003,China), AuthorCompanyExt(id=1322960962462032538, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962445255320, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 河北省能源电力知识计算重点实验室,保定 071003)])]), Author(id=1322960963128926894, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=18957139414@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1322960963225395889, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960963128926894, language=EN, stringName=Qingyuan Zhu, firstName=Qingyuan, middleName=null, lastName=Zhu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, 4, address=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China
4 Engineering Research Center of Intelligent Computing for Complex Energy Systems,Ministry of Education,Baoding 071003,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1322960963296699058, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960963128926894, language=CN, stringName=朱清源, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, 4, address=2 华北电力大学控制与计算机工程学院,保定 071003
4 复杂能源系统智能计算教育部工程研究中心,保定 071003, bio={"content":"

朱清源,男,硕士研究生,主要研究方向为人工智能与新能源。E-mail:

"}, bioImg=null, bioContent=

朱清源,男,硕士研究生,主要研究方向为人工智能与新能源。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1322960962361369237, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=2, ext=[AuthorCompanyExt(id=1322960962369757846, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962378146455, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 华北电力大学控制与计算机工程学院,保定 071003)]), AuthorCompany(id=1322960962520752795, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=4, ext=[AuthorCompanyExt(id=1322960962529141404, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962520752795, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4 Engineering Research Center of Intelligent Computing for Complex Energy Systems,Ministry of Education,Baoding 071003,China), AuthorCompanyExt(id=1322960962537530013, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962520752795, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4 复杂能源系统智能计算教育部工程研究中心,保定 071003)])]), Author(id=1322960963388973748, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=18231204373@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1322960963456082615, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960963388973748, language=EN, stringName=Chaogang Zhang, firstName=Chaogang, middleName=null, lastName=Zhang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, 4, address=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China
4 Engineering Research Center of Intelligent Computing for Complex Energy Systems,Ministry of Education,Baoding 071003,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1322960963506414264, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960963388973748, language=CN, stringName=张超刚, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, 4, address=2 华北电力大学控制与计算机工程学院,保定 071003
4 复杂能源系统智能计算教育部工程研究中心,保定 071003, bio={"content":"

张超刚,男,硕士研究生,主要研究方向为基于轻量化模型的风机叶片缺陷检测。E-mail:

"}, bioImg=null, bioContent=

张超刚,男,硕士研究生,主要研究方向为基于轻量化模型的风机叶片缺陷检测。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1322960962361369237, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=2, ext=[AuthorCompanyExt(id=1322960962369757846, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962378146455, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 华北电力大学控制与计算机工程学院,保定 071003)]), AuthorCompany(id=1322960962520752795, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=4, ext=[AuthorCompanyExt(id=1322960962529141404, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962520752795, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4 Engineering Research Center of Intelligent Computing for Complex Energy Systems,Ministry of Education,Baoding 071003,China), AuthorCompanyExt(id=1322960962537530013, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962520752795, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4 复杂能源系统智能计算教育部工程研究中心,保定 071003)])]), Author(id=1322960963560940218, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=li_bing_hb@126.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1322960963653214909, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960963560940218, language=EN, stringName=Bing Li, firstName=Bing, middleName=null, lastName=Li, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 Yanzhao Electric Power Laboratory of North China Electric Power University,Baoding 071003,China
2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1322960963716129470, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960963560940218, language=CN, stringName=李冰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 华北电力大学燕赵电力实验室,保定 071003
2 华北电力大学控制与计算机工程学院,保定 071003, bio={"content":"

李冰,男,副教授,主要研究方向为机器学习与计算机视觉。E-mail:

"}, bioImg=null, bioContent=

李冰,男,副教授,主要研究方向为机器学习与计算机视觉。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1322960962290066066, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=1, ext=[AuthorCompanyExt(id=1322960962298454675, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962290066066, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Yanzhao Electric Power Laboratory of North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962306843284, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962290066066, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 华北电力大学燕赵电力实验室,保定 071003)]), AuthorCompany(id=1322960962361369237, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=2, ext=[AuthorCompanyExt(id=1322960962369757846, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962378146455, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962361369237, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 华北电力大学控制与计算机工程学院,保定 071003)])]), Author(id=1322960963812598464, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, orderNo=5, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=wbin.cao@ncepu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1322960963892290243, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960963812598464, language=EN, stringName=Wangbin Cao, firstName=Wangbin, middleName=null, lastName=Cao, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 5, address=1 Yanzhao Electric Power Laboratory of North China Electric Power University,Baoding 071003,China
5 Hebei Key Laboratory of Power Internet of Things Technology,North China Electric Power University,Baoding 071003,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1322960963946816196, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, authorId=1322960963812598464, language=CN, stringName=曹旺斌, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 5, address=1 华北电力大学燕赵电力实验室,保定 071003
5 河北省电力物联网技术重点实验室,保定 071003, bio={"content":"

曹旺斌,男,副教授,主要研究方向为人工智能与新能源。E-mail:

"}, bioImg=null, bioContent=

曹旺斌,男,副教授,主要研究方向为人工智能与新能源。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1322960962290066066, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=1, ext=[AuthorCompanyExt(id=1322960962298454675, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962290066066, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Yanzhao Electric Power Laboratory of North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962306843284, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962290066066, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 华北电力大学燕赵电力实验室,保定 071003)]), AuthorCompany(id=1322960962596250270, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, xref=5, ext=[AuthorCompanyExt(id=1322960962604638879, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962596250270, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=5 Hebei Key Laboratory of Power Internet of Things Technology,North China Electric Power University,Baoding 071003,China), AuthorCompanyExt(id=1322960962608833184, tenantId=1146029695717560320, journalId=1249024232475115590, articleId=1322960951045137026, companyId=1322960962596250270, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=5 河北省电力物联网技术重点实验室,保定 071003)])])]
张弪弘1, 2 , 苏攀2, 3, * , 朱清源2, 4 , 张超刚2, 4 , 李冰1, 2 , 曹旺斌1, 5
作者信息
  • 1 华北电力大学燕赵电力实验室,保定 071003
  • 2 华北电力大学控制与计算机工程学院,保定 071003
  • 3 河北省能源电力知识计算重点实验室,保定 071003
  • 4 复杂能源系统智能计算教育部工程研究中心,保定 071003
  • 5 河北省电力物联网技术重点实验室,保定 071003
通讯作者:
苏攀,通信作者,男,副教授,主要研究方向为机器学习与医学图像处理。E-mail:
作者简介:

张弪弘,女,硕士研究生,主要研究方向为新能源与目标检测。E-mail:

苏攀,通信作者,男,副教授,主要研究方向为机器学习与医学图像处理。E-mail:

朱清源,男,硕士研究生,主要研究方向为人工智能与新能源。E-mail:

张超刚,男,硕士研究生,主要研究方向为基于轻量化模型的风机叶片缺陷检测。E-mail:

李冰,男,副教授,主要研究方向为机器学习与计算机视觉。E-mail:

曹旺斌,男,副教授,主要研究方向为人工智能与新能源。E-mail:

Review of damage detection methods for wind turbine blades
Jinghong Zhang1, 2 , Pan Su2, 3, * , Qingyuan Zhu2, 4 , Chaogang Zhang2, 4 , Bing Li1, 2 , Wangbin Cao1, 5
Affiliations
  • 1 Yanzhao Electric Power Laboratory of North China Electric Power University,Baoding 071003,China
  • 2 School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China
  • 3 Hebei Key Laboratory of Knowledge Computing for Energy and Power,Baoding 071003,China
  • 4 Engineering Research Center of Intelligent Computing for Complex Energy Systems,Ministry of Education,Baoding 071003,China
  • 5 Hebei Key Laboratory of Power Internet of Things Technology,North China Electric Power University,Baoding 071003,China
出版时间: 2026-08-16 doi: 10.11834/jig.250458
文章导航
收藏切换

全球能源转型推动了风力发电机的大规模部署,基于无人机图像采集的风机叶片(wind turbine black,WTB)缺陷目标检测技术已成为研究热点。风力发电机叶片的无人机图像存在质量低、缺陷与背景纹理相似、缺陷特征多尺度等问题,且标注缺乏统一标准并高度依赖人工,这些因素显著增加了缺陷检测的难度。当前关于风力发电机叶片缺陷检测的综述主要聚焦于无损检测方法,针对图像检测技术的系统性综述仍然不足,且多集中于模型架构的优化,而忽略了图像预处理和学习策略对检测性能的重要影响。因此,本文将图像预处理和学习策略纳入视角,系统综述了风力发电机叶片缺陷目标检测的研究进展:1)详细列举叶片缺陷类型及特征。2)与电力系统其他检测任务进行对比研究并总结主要问题。3)从图像预处理、网络改进和学习策略3个维度总结现有风力发电机叶片缺陷目标检测方法。在图像预处理方面,基于任务导向将现有方法归纳为3类:图像质量增强、几何校正以及结构特征提取;在网络优化方面,沿目标检测模型发展脉络,梳理了各类架构的改进策略;最后,总结了该领域常用的学习策略。4)对近年来数据集与常用评估指标进行汇总。5)对风机叶片缺陷目标检测方法面临的挑战及未来研究方向进行总结和展望。

风机叶片缺陷检测  /  无人机图像  /  图像预处理  /  目标检测  /  学习策略

With the accelerating transition of the global energy landscape toward renewables,wind power has been widely promoted as a vital component of clean energy. Alongside the rapid expansion of wind turbine installed capacity,the structural health of wind turbine blades (WTBs),which are among the critical power-generating components,directly influences the safety and efficiency of wind power systems. In recent years,unmanned aerial vehicles (UAVs) have been equipped with high-resolution imaging devices for blade image acquisition. By combining advanced image processing and object detection techniques,they have achieved automated,noncontact defect detection,which not only enhances detection efficiency and coverage but also reduces the risks and costs of traditional manual inspections. Therefore,this approach has become a mainstream methodology in the industry and a focal point of research. However,images captured by UAVs often suffer from poor quality (primarily manifested as noise,motion blur,and low resolution),posing critical challenges to defect detection. Moreover,defect features,such as erosion marks,stains,or manufacturing traces,frequently exhibit high visual similarity to blade surface textures and present multiscale and diverse morphological characteristics. This visual ambiguity further complicates accurate localization and classification. In addition,the lack of standardized annotation protocols and the heavy reliance on manual labeling hinder the construction of high-quality datasets. These issues collectively constrain the detection accuracy and generalization performance of deep learning-based object detection models. Existing survey literature on WTB defect detection has primarily emphasized nondestructive testing methods,with a noticeable scarcity of comprehensive reviews dedicated exclusively to image-based detection approaches. Current surveys in this field focus `on architectural advancements within deep learning models,including the integration of lightweight modules,attention mechanisms,and multiscale feature extraction techniques. Although these architectural improvements have enhanced detection accuracy,they often overlook essential aspects,including image preprocessing methods and learning strategies,both of which play a crucial role in determining overall model performance. In this review,we systematically incorporate perspectives on image preprocessing and learning strategies to analyze defect detection methods for WTBs. The main contributions of this review are as follows. 1) From the perspective of defect manifestation in WTBs,defects can be broadly categorized into internal and surface types. We systematically present and analyze the causes,formation mechanisms,and visual characteristics of these defect categories with respect to their structural properties. We also summarize the most frequently observed defect types in datasets. These datasets provide valuable data support for the design and optimization of subsequent defect detection models. 2) This review summarizes the characteristics of defects in detection tasks across various power system components,such as foreign object recognition on transmission lines and insulator defect detection. Through a comparative analysis,this review demonstrates that WTB defect detection uniquely encompasses all the aforementioned challenges and variabilities. Consequently,WTB defect detection necessitates simultaneously addressing a spectrum of challenges,including poor image quality,multiscale defect manifestations,class imbalance,lack of standardized annotation protocols,and the need for lightweight deployable models. Such a problem-driven perspective provides a foundation for inspiring future methodological advancements. 3) We conduct an in-depth investigation into the key components of deep learning-based image-driven WTB defect detection. Three critical aspects,namely,image preprocessing,model architecture,and dominant learning strategies,are presented. With respect to image preprocessing,task-driven methodologies are categorized into image quality enhancement,geometric correction,and structural feature extraction. Each of them is examined for its critical role in improving the saliency of defects and bolstering the robustness of detection models. At the architectural level,the enhancement strategies across mainstream detection frameworks,including two-stage,one-stage,and Transformer-based models,are presented. In terms of learning strategies,structured analyses of the practical effectiveness and potential of approaches,including transfer learning,ensemble learning,cascade strategies,and unsupervised/semisupervised learning,are presented. These strategies are evaluated in terms of their capability to address limited data availability,class imbalance,and label uncertainty in defect detections. 4) This review also provides an overview of publicly available datasets released in recent years,examining their data types,annotation protocols,and defect categories,thereby offering guidance for researchers in selecting appropriate research sources. A summary of commonly used evaluation metrics is provided,underscoring the necessity of image quality assessment in diverse application scenarios. 5) This review outlines critical challenges in image-based WTB defect detection,including imbalanced defect distributions,heavy reliance on manual annotations,limited model generalization under environmental variations,insufficient data modality diversity,and lack of innovation in model architectures. In light of these challenges,several promising research directions are proposed; these include the effective integration of multimodal data,creative design of model structures,and in-depth adoption of weakly or self-supervised learning strategies,with the aim of inspiring and guiding future advancements in WTB defect detection.

wind turbine blade (WTB) defect detection  /  UAV imagery  /  image preprocessing  /  object detection  /  learning strategies
张弪弘, 苏攀, 朱清源, 张超刚, 李冰, 曹旺斌. 风力发电机叶片缺陷目标检测研究综述. 中国图象图形学报, 2026 , 31 (8) : 2844 -2867 . DOI: 10.11834/jig.250458
Jinghong Zhang, Pan Su, Qingyuan Zhu, Chaogang Zhang, Bing Li, Wangbin Cao. Review of damage detection methods for wind turbine blades[J]. Journal of Image and Graphics, 2026 , 31 (8) : 2844 -2867 . DOI: 10.11834/jig.250458
全球对可持续能源需求的激增正推动风力发电机(wind turbine,WT)快速部署。根据《2025年全球风能报告》,2024年全球风电累计装机容量较2023年增加117 GW,达1 136 GW(Global Wind Energy Council,2025)。其中,我国装机容量占全球装机容量的68%,年增加18%,约79 GW,达到521 GW(国家能源局,2025)。风电作为全球能源转型的关键支柱和清洁能源的核心组成部分,其设备结构健康监测的重要性日益凸显(许国东 等,2018)。
WT主要结构包含塔筒、机舱、转子和底座,其中转子由轮毂和叶片组成,叶片作为捕风部件,其性能直接决定发电效率。风机叶片(wind turbine blade,WTB)需兼具结构强度与功能完整性:在承受风压、自重及循环载荷的同时,须维持最优气动外形;还需抵御雷击、鸟撞等突发冲击及环境腐蚀(王磊 等,2023)。因此,建立精准的状态监测体系并实施预防性维护,成为保障机组长期可靠运行的核心策略。
风机叶片的检测方法主要分为人工检测和自动检测。传统的人工巡检方法操作简便但检测效率低,通常仅适用于识别裂纹、侵蚀等明显的表面缺陷,难以实现早期检测(Cawley,2018)。相比之下,自动检测技术通过不同信号分析叶片的物理和力学特性,实现更为高效、准确的缺陷检测(王磊 等,2023)。其中,鉴于风机具有较高的经济价值,有损检测方法并不适用于其缺陷检测;而无损检测技术(non-destructive testing,NDT)可在不影响经济效益的前提下,有效实现风机叶片的早期健康检测,从而减少突发维修成本(unscheduled maintenance cost)和平准化能源成本(levelized cost of energy),延长设备运行时间(Rubert等,2018)。常见NDT手段可分为接触式与非接触式两类,具有检测灵敏度高、可自动化集成等优势。然而,随着WTB尺寸的不断扩大,现场检测面临信号弱、环境干扰强、检测设备难以部署等挑战,导致其在复杂工况下的精度与鲁棒性受限(García M􀅡rquez和Peco Chacón,2020)。
如图1所示,常见的无损检测方法主要基于声学(如超声波)、电学(如阻抗)及光学(如红外)等物理信号。该类方法通过传感器或无人机采集信号或图像,并据此分析其中的异常变化,从而定位缺陷并评估其严重程度。然而,基于传感器的信号获取方法在实际应用中面临诸多挑战,例如:高空环境导致传感器安装困难;检测精度对传感器布局与固定方式敏感;传感器本身易受外部噪声干扰;物理信号和噪声混叠增加了数据处理难度。这些因素共同制约了基于传感器的无损检测方法在实际检测中的性能和实际效果。
鉴于上述检测方法存在的固有局限,基于多维图像信号的NDT技术凭借其优异的可解释性成为当前研究的热点方向(Khan和Byun,2023)。该方法主要通过无人机搭载高分辨率拍摄设备获取叶片图像,结合先进目标检测算法,实现早期缺陷的精准检测。该方法因操作简便性及优异鲁棒性,成为WTB缺陷检测的主流技术。然而复杂环境条件、多因素干扰导致的图像质量下降以及缺陷特征的多样性,仍对其检测精度构成挑战。同时,WTB损伤类型的多样性及其标注标准的不一致性,导致高质量公开数据集的获取面临显著挑战。这些难题促使研究者探索小目标检测、图像增强及自监督学习等新型方法,以突破现有检测性能的瓶颈。
表1展示了近几年WTB检测文献综述对比。区别于表1中的现有综述,本文聚焦于WTB损伤目标检测,深入探讨目标检测模型的改进与应用。如图2所示,本文以目标检测模型的演进为脉络,系统分析了WTB缺陷检测技术的架构变迁,阐述了两阶段检测器、单阶段检测器及基于Transformer的检测网络在该领域的迭代发展,总结了不同模型网络特征,重点从特征提取、网络架构及训练策略3个层面展开比较研究。最后基于现有研究进展,进一步探讨了WTB缺陷检测领域的技术瓶颈与发展趋势,并提出了潜在的突破方向与研究空白。
长期动态载荷作用下,WTB易发生材料性能退化和疲劳损伤,进而导致其气动性能衰减,最终威胁风电机组运行安全(Chen和Eder,2020; Katsaprakakis等,2021)。基于WTB的结构特征、缺陷成因及缺陷表现形式,常见缺陷可分为内部缺陷与表面缺陷两类。
内部结构缺陷往往是在长期载荷循环、制造缺陷及环境侵蚀等多种因素共同作用下逐步积累形成的。根据其诱发机制的不同,内部缺陷可进一步细分为制造缺陷、疲劳损伤、应力损伤、外部载荷诱导损伤、腐蚀损伤以及胶层脱粘等类型。如表2所示,内部缺陷通常在早期隐蔽性强,难以被常规检测手段察觉,一旦发展至临界状态,将外显为表面缺陷。
表面缺陷(如图3所示)会显著降低叶片的气动性能,导致捕风效率下降。并且局部应力集中会进一步加速材料老化过程。此外,强风、雷击等极端天气因素,甚至可能造成叶片断裂,引发重大安全事故;在高海拔或寒冷潮湿地区,叶片结冰现象频发,显著增加旋转阻力,削弱气动性能,降低发电效率,严重时甚至诱发失速或结构损伤。
因此,针对叶片损伤开展早期识别与状态监测,对于减缓故障扩展、延长叶片服役周期以及保障风电机组的安全稳定运行具有重要意义。
传统WTB无损检测技术主要依赖传感器方法,通过分析时序信号中的异常信息实现损伤识别。常见声信号、光信号和电信号3类信号传感器(Gholizadeh,2016)。
声信号传感器在复杂环境噪声下的鲁棒性和抗干扰能力有限(Kumar等,2024),尤其对叶片整体的早期微损伤的捕捉灵敏度不足(Tiwari等,2017)。光信号传感器的稳定性和适应性要求更高,在复杂部位缺陷识别和损伤类型区分上存在不足,限制其应用场景。电信号传感器的信号采集过程易受电磁干扰,导致信号不稳定,影响故障识别准确性。此外,WTB对附加载荷敏感,限制了传感器布设范围;不同类型传感器间、与背景噪声间存在信号耦合,进一步降低了数据质量。基于电磁波的无损检测方法受限于叶片的轴向方向(Moll,2019),且多应用于检测裂纹,在其他缺陷上的检测尚不成熟(Moll等,2018)。综上,基于传感器的WTB损伤检测方法在实际应用中仍面临诸多挑战。具体包括:1)传感器系统易受外部环境干扰,且对高温、高湿和高风等恶劣工况适应性较差;2)传感器布设与维护过程依赖大量人力与物力,成本较高(Gómez Muñoz等,2019);3)信号处理流程复杂,且对早期微小损伤的敏感性不足,存在漏检风险(Kumar等,2024);4)当前高精度定位算法主要面向薄壁等二维材料的缺陷定位,在复杂三维叶片结构中的应用存在局限,难以实现准确的缺陷定位。
无人机航拍为无损检测技术提供了新的数据采集方式,并结合计算机视觉,为WTB缺陷检测提供了高效解决方案。该方法借助图像处理与深度学习算法,对无人机拍摄图像进行分析,并提取缺陷区域的几何形貌与纹理特征,从而实现缺陷的智能识别与分类。相较于传统传感器方法,该技术显著提升了检测的空间覆盖性与环境适应性,尤其适用于早期损伤识别,为风电叶片结构健康监测开辟了新途径。
风力发电机作为新能源系统的起点之一,其运行状态对整个发电过程具有决定性影响。基于计算机视觉对WTB进行缺陷检测,不仅能实现对风机健康的早期检测,其在算法框架与技术路径上也与输电线路、变电设备等电力关键部件的智能巡检任务具有高度互通性。
为系统性揭示电力视觉巡检任务间的共性和WTB缺陷检测的特殊性,本文在表3中将其与输电线异物检测、变压器油泄露检测等典型任务进行了对比分析。对比结果表明,WTB缺陷检测主要面临两大挑战:在检测层面,其缺陷不仅形状各异(与输电线异物检测类似)、颜色较浅(与变压器油泄露检测类似),更易与复杂背景混叠,对模型的细微特征分辨能力提出了极高要求;在数据层面,公开数据集匮乏且标注标准不统一,进一步加剧了模型训练的难度。
本节对WTB缺陷目标检测中存在的问题进行分类与细化。
无人机拍摄图像常存在噪声、运动模糊等问题,对焦不准确会进一步加剧图像模糊;在飞行过程中抖动导致图像几何畸变,会使图像中叶片出现几何变形(钟帅和王丽萍,2025)。自然光照条件的变化以及拍摄角度的不理想常造成图像亮度不均,从而掩盖或夸大部分区域的缺陷特征(卢荣胜 等,2018)。高空拍摄中常出现目标像素不足、背景复杂等情况,增加误检漏检概率(钟帅和王丽萍,2025)。图像的存储与传输过程可能引入非自然结构或异常视觉效果。具体表现为图像边缘块状结构或特征不连续,影响模型对缺陷的判别能力。
针对上述问题,多种图像预处理技术可以减弱低质量图像对检测性能的不利影响,以提升图像清晰度与特征信息的可信度。
由于无人机采集视角和距离的差异,且缺陷类型多、特征多样,目标通常呈现出显著的多尺度变化(钟帅和王丽萍,2025)。当目标在图像中的尺寸范围较大时,模型易面临感受野未对齐或锚框尺寸与目标不匹配问题,导致训练不稳定,甚至对某一特定尺度出现过拟合(潘晓英 等,2023)。小尺度目标存在特征信息有限及上下文语义匮乏的问题,亟需精细的特征提取机制,但神经网络深度的加深以及下采样操作的累积显著缩减小目标在特征图中的像素占比,导致模型表征能力下降,形成特征表达瓶颈。此外,在多尺度目标分布下,可能感兴趣区域(region of interest,RoI)急剧增多,空间上的不确定性随之增加,使得锚框回归精度降低,目标定位任务面临更大挑战(Wang等,2023b)。
因此,目标尺度差异大成为影响风机缺陷检测性能的关键因素之一,对检测模型的结构设计和训练策略提出更高的要求。
风电场分布与环境条件多样,导致不同缺陷的发生概率差异显著,致使数据集中类别分布明显不均衡。在多数公开数据集中,积冰和雷击损伤的样本数量远少于裂纹、侵蚀及脱漆等主导缺陷。这种不均衡分布会导致以下问题。首先,在模型训练过程中,数量多的类别样本占据优势,使得梯度更新更偏向学习多数类特征,导致少数类特征难以有效建模与泛化(李新叶 等,2020)。其次,常用的评价指标往往偏向整体性能,难以准确反映模型在少数类缺陷检测中的实际表现(李艳霞 等,2019)。
针对上述问题,现有检测模型普遍采用数据增强技术以提升少数类样本的比例,或通过调整损失函数权重,增强对少数类缺陷特征的关注,从而改善模型的检测效果。
低质量图像不仅易导致模型检测性能降低,同时也显著增加了数据标注的难度。以裂纹等缺陷为例,其在图像中常呈现模糊、渐变和形态不规则的特征,使得边界难以明确界定。此外,不同类型的缺陷在外观上可能高度相似,导致在标注结果上不同人员存在显著差异,增加了分类判断难度。例如,公开数据集DTU-Drone(Shihavuddin等,2019)缺乏统一的标注标准(Gohar等,2023),这阻碍了不同研究间的算法性能比较。若不统一标注标准,将会导致模型过拟合、性能不稳定及泛化能力下降。
为减轻高成本、低一致性的人工作业负担,现有研究尝试引入半监督学习、主动学习或自监督机制,以减少对高质量标注的依赖并提升模型稳健性。
在WTB缺陷检测中,如何在保持高检测精度的同时实现模型的轻量化、低功耗与实时响应,是当前研究与应用的重点。一方面,边缘计算设备通常受限于计算能力、内存容量和能耗上限,难以承载大规模深度神经网络的高算力需求(王其朝 等,2021);另一方面,边采集边处理的检测系统需要实现高频率巡检和快速响应,边采集集中处理的检测系统需具备低延迟与高吞吐量,以确保检测效率与系统稳定性。此外,在偏远或高危环境下,轻量模型还可显著提升系统的整体部署灵活性与安全性。
为在性能与效率之间取得平衡,当前研究广泛采用轻量网络架构、模型剪枝、知识蒸馏及量化压缩等策略,实现对模型结构与计算需求的优化。
图像预处理可显著降低低质量图像对检测结果的影响。本节参考Gonzalez和Woods(2018)以及Szeliski(2011)中的分类方法,以任务为导向,将图像预处理方法分为图像质量改善、图像几何校正以及图像结构与提取。表4系统总结了WTB缺陷检测中常用的图像预处理方法。
在WTB图像采集过程中,环境因素(如光照变化、云雾干扰)和风机运行引起的结构噪声会导致图像出现噪声干扰和局部遮挡等问题。Guo等人(2021)结合手工特征提取与浅层学习模型去除图像噪声,提升WTB损伤识别精度,但在复杂背景下泛化能力有限。
去噪自编码器(denoising autoencoder,DAE)通过在输入图像上叠加噪声,生成损坏图像,引导模型学习从扰动中恢复原始无噪声图像,进而提取关键的边缘与结构信息(Vincent等,2008)。Jia和Chen(2025)利用柏林噪声构建DAE,并结合记忆辅助机制存储正常图像的特征,通过在推理阶段检索相似表征,辅助识别DAE难以还原的异常区域。
此外,基于深度学习的去噪方法可显著提升检测性能。Gao等人(2025)引入去噪扩散概率模型,通过在清晰图像上逐步施加高斯噪声,模拟图像退化过程;并利用时间条件U-Net和基于马尔可夫链迭代预测噪声分布,逐步还原无噪声图像。然而,该论文中展示的裂纹图像清晰度较高,其在实际场景中的适应性仍有待验证。
近年来,基于卷积神经网络(convolutional neural network,CNN)和生成对抗网络(generative adversarial network,GAN)的去模糊方法在计算机视觉领域取得了显著进展(Amrollahi,2023)。Du等人(2023)通过多层卷积与反卷积架构,结合渐进上采样与残差学习,实现了从低级到高级特征的提取与细节重建。但该方法通过模糊核估计实现对图像退化过程的逆向建模,其性能在复杂场景下易受核估计精度的制约。
DeblurGAN-v2作为一种基于GAN的高效图像去模糊模型,通过残差结构和多尺度判别器,实现对复杂模糊图像的自适应建模与还原(Kupyn等,2019)。Lu等人(2024)利用预训练的DeblurGAN-v2模型对数据集进行去模糊处理,但在重度运动模糊及特征噪声相似时仍存误检漏检。Peng等人(2022a)引入Inception-ResNet-v2作为DeblurGAN-v2的主干网络,显著增强边缘纹理细节,提升检测模型在中重度模糊条件下的缺陷检测性能。
除了通过对图像进行去模糊以降低运动模糊的影响,还可通过优化无人机飞行稳定来主动规避运动模糊。Wang等人(2023)基于闭环稳定性控制,通过调节无人机速度抑制运动模糊,降低叶片对飞行稳定性的影响,从而提高图像质量。
在风电叶片巡检中,通过超分辨率重建方法,可在低成本硬件条件下实现高分辨率图像获取,有效提升对微小缺陷的早期缺陷的识别能力。
超分辨率方法主要通过CNN学习低分辨率与高分辨率图像之间的映射关系,揭示图像细节恢复的隐空间规律。如图4所示,Sarkar和Gunturi(2021)采用SRCNN(super-resolution convolutional neural network)算法提升模糊图像分辨率,实现高分辨率图像的重建。该方法可显著增强缺陷检测系统在多视角、动态光照及复杂背景条件下的鲁棒性。
GAN的引入标志着超分辨率图像重建的重要突破,通过生成器与判别器间的博弈,提升生成图像的视觉质量与模型的细节还原能力。专注递归生成对抗网络在此基础上引入注意力机制与递归结构,提升模型在复杂光照条件下的表现(李健和郭鹏,2021)。
DAT(dual-attention Transformer)基于Transformer架构,通过融合浅层空间特征与深层语义特征,协同建模空间细节与通道上下文;并借助上采样重建以增强纹理与结构保真度。Zhao等人(2024)进一步引入基于指数变换的对比度增强算法,提升图像在低照度环境下的纹理细节与对比度,其出色的重建能力和结构还原能力更适用于对图像精度与真实性要求较高的应用场景。
基于图像拼接的高分辨率全景图像为WTB缺陷检测提供了更广阔的视野和更完整的特征信息,从而提升了缺陷定位与检测的准确性。该方法还能有效消除局部遮挡和阴影干扰,获取更全面的表面特征信息。
Li等人(2020)提出了一种基于变分的图像拼接方法,在保留叶片关键特征的同时有效抑制背景干扰,实现叶片边缘的精确提取,并基于Harris角点和随机抽样一致(random sample consensus,RANSAC)算法实现边缘匹配和图像融合,最终结合插值和加权平滑算法实现图像拼接。Yang等人(2023)根据细粒度对纹理特征和形状进行拼接,利用Deep Lab分割算法提取叶片信息,增加拼接图像的准确性和可靠性。Yu等人(2023a)基于无人机飞行数据和像素位移,精确对齐叶片边缘,实现高质量红外图像的无缝拼接,生成完整全景图像。
在WTB缺陷检测中,常将红外图像(infrared image,IR)与可见光图像进行融合。Zhou等人(2023)将透明度为50%的红外图像叠加于可见光图像之上,同时引入Rep-ResBlock的结构设计和可学习的权重因子,将IR图像特征融入图像表征,增强细节特征表达,提升检测精度与鲁棒性。
此外,高光谱成像技术可用于积冰检测(Nakauchi等,2012)。基于图像的积冰缺陷检测极具挑战性(Parent和Ilinca,2011)。积冰因视觉特性与叶片颜色相近、边缘模糊,导致传统检测方法效果有限;同时,环境条件(如雨雾、光照及红外温差减弱)也显著降低了目标与背景的可区分性(Nakauchi等,2012)。Rizk等人(2020)通过配准3种覆盖不同光谱区域的高光谱传感器所得到的图像,以精准检测结冰特征。
为提高风机缺陷检测精度,常采用图像分割方法消除背景干扰,突出目标区域,为后续的缺陷识别提供更具清晰的结构化输入。
Yu等人(2023a)在U-Net网络结构引入分层深度可分卷积块,增强对关键特征的捕捉能力;并将特征图分组处理,实现多层级信息融合。改进后的U-Net模型性能显著提高但推理速度降低。Yu等人(2023b)利用改进U-Net和Sobel算子剔除复杂背景,降低后续图像拼接难度。Rizvi等人(2024a)在U-Net上引入Pixel Shuffle和Pixel Unshuffle操作,在保持空间分辨率的同时增强特征还原能力,提升模型分割能力。在此基础上,进一步通过Canny边缘检测算子提取叶片轮廓信息,构建了一个仅包含叶片区域的精简数据集(Rizvi等,2024b),分割后的数据集显著提升了检测精度和模型收敛速度。
图像分解与融合在图像处理中协同作用,其中分解提供结构化特征表示,融合则整合多源特征以提升信息利用效率与表达能力。
Peng等人(2022b)提出了一种针对非均匀光照下的风机图像增强方法。该方法对分解后的卡通和纹理特征进行图像增强,通过在高斯尺度空间中建立光照模型以去除光照不均匀现象,并使用正弦函数和Gabor变换,恢复边缘细节,从而提升损伤检测性能。Ji等人(2025)提出的CMFCANet网络框架,通过通道级联将RGB和其对应的HSI(hyperspectral image)颜色空间进行结合,并采用多通道卷积方法,提升检测模型在复杂背景下对微小损伤的检测性能。
基于本节分析,图像去噪、去模糊和超分辨率技术能有效改善WTB缺陷的图像质量,显著提升了缺陷特征的清晰度和可辨识性;同时,图像拼接和配准为构建更具判别性的数据集提供了技术支持;此外,图像分割去除复杂背景对检测网络的干扰,图像分解与融合则提供了更为精细的边缘纹理细节。这些方法共同为后续缺陷检测与定位研究奠定了基础。
近年来,基于卷积神经网络和Transformer等先进模型的视觉检测技术,已成为WTB缺陷检测领域的重要研究方向。Moreno等人(2018)通过在离线环境下训练CNN模型,证实了CNN在WTB缺陷检测领域的可行性,但对低分辨率图像及小尺寸目标的检测效果仍有待提升;Reddy等人(2019)结合滑动窗口技术,实现缺陷区域的二分类识别及多类缺陷精细分类,但在处理小尺寸缺陷时存在计算效率瓶颈,且模型的高计算需求限制了其在边缘设备中部署。Altice等人(2024)在ResNet-50骨干网络中融合Transformer结构,并采用可变形卷积模块和基于内容感知特征重组的特征金字塔模块,显著提升定位精度与召回率。
尽管CNN在目标检测领域取得显著进展,但其对小目标的检测精度和空间定位准确性存在不足。为此,研究者提出了两类主流检测框架,根据是否使用区域建议网络可分为:以R-CNN系列为代表的两阶段检测模型和以YOLO(you only look once)为核心的单阶段检测模型。近年来,除传统CNN架构外,Transformer凭借自注意力机制,展现出强大的全局建模能力、鲁棒性与语义表达性能,为目标检测任务开辟了新的研究方向,如图5所示。
为应对WTB缺陷检测中所面临的多重挑战,近年来研究者针对检测网络的各组成部分进行了系统优化。采用注意力机制、轻量化网络结构和改进损失函数等策略,显著提高了模型在复杂WTB检测中的适应性、精度和效率。下文按照不同检测模型,简要梳理在网络结构优化方面的最新研究进展。
为克服传统卷积神经网络在目标定位方面的局限性,Girshick等人(2014)提出了区域卷积神经网络R-CNN,首次将候选区域与深度卷积特征结合,建立了“候选区域生成—特征提取—分类器”组成的两阶段目标检测框架。Mao等人(2021)提出了改进的级联深度学习网络,结合可变形卷积、可变形RoI与上下文信息提取,增强小缺陷识别能力,验证了两阶段级联网络在该类任务中的适用性与扩展潜力。
尽管R-CNN在检测精度方面取得了突破,但其检测流程存在候选区域数量庞大、特征提取过程重复冗余等问题。Fast R-CNN通过共享特征图与RoI Pooling减少重复计算,并引入区域建议网络实现精度与效率间的平衡。Shihavuddin等人(2019)针对风机表面胶衣脱落问题,使用ResNeXt-101作为Faster R-CNN特征提取网络,并结合DCNv2提升对脱落部分形状特征多变的表征能力。
在Faster R-CNN实现目标检测与候选框回归的基础上,He等人(2016)进一步提出了Mask R-CNN,以实现更精细的目标检测任务。Zhang等人(2020)在Mask R-CNN-512中加入MRNet(MinReact net),实现对低噪声图像中叶片提取和缺陷分类。随后,Zhang等人(2021)比较了YOLOv3、YOLOv4和Mask R-CNN的性能,并引入图像增强和数据增强构建IE Mask R-CNN模型,提升在不同光照下的特征识别能力和鲁棒性。Diaz和Tittus(2023)通过级联策略,对引入深度可分离卷积的Mask R-CNN进行训练,具有较高的检测准确性和较低的计算成本。
上述两阶段检测方法因候选区域与特征提取分离导致计算复杂、实时性差,难以满足目标检测任务对速度和精度的双重需求。为此,研究者提出了端到端的单阶段检测方法。
SSD(single shot multibox detector)采用多尺度特征图实现同步检测,在保持精度的同时提升了速度,其网络结构更简洁、推理速度更快。季利鹏等人(2023)采用ResNet50(residual network)作为SSD的特征提取网络,但是未分析参数量与速度,也未与YOLOv4及以后的系列进行对比。EADD(efficient and accurate damage detector)(Lyu等,2022)在主干中引入FDSB(factorized depth-wise separable bottleneck)和FAM(feature aggregation module)构成的密集连接块,并结合双向特征金字塔网络(bidirectional feature pyramid network,BiFPN),实现特征高效表达与融合,但整体速度仍低于YOLOv4。
SSD虽然具有检测速度快、结构简单等优点,但在小目标检测任务中对高质量定位不敏感,锚框匹配依赖强,不满足风机检测的高精度需求。YOLO将目标检测转化为回归问题,直接从图像像素到边界框和类别进行端到端预测,实现单阶段检测。
Sarkar和Gunturi(2021)利用超分辨率重建图像构建高质量缺陷数据集,并采用YOLOv3网络实现了6类表面缺陷的自动识别,但未与YOLOv4等模型对比,亦未讨论参数与速度指标。Liu等人(2023)通过骨干网更换、过滤器修剪和知识蒸馏等,降低YOLOv3复杂度和计算量,但同样缺乏横向对比实验。
YOLOv4在YOLOv3的基础上进一步提高检测的准确率和模型的泛化能力。Zhang等人(2022)采用轻量级MobileNetv1和深度可分离卷积降低计算成本,并比较通道注意力机制SE(squeeze-and-excitation)、卷积块注意力模块(convolutional block attention module,CBAM)和高效通道注意力(efficient channel attention,ECA)在不同位置的作用效果,选取最佳模型进行缺陷检测。Huang等人(2024)充分利用YOLOv4主干网络的特征提取能力,并在合适位置中插入空间和通道压缩与激励注意力机制(spatial and channel squeeze and excitation,scSE),提升空间和通道信息提取能力,但未探讨轻量化问题。
YOLOX在YOLOv3和YOLOv4基础上优化,具有多尺度检测能力,广泛适用于工业检测和小目标识别。面对工业场景中目标多尺度问题和轻量化问题,WT-YOLOX(Yao等,2023)通过SimRepVGG作为主干网络并增加CFFM(cascade feature fusion module)模块,实现小目标特征提取。
PP-YOLO在保留YOLOv4核心架构的基础上进行了性能优化。其后续版本PP-YOLOv2重点提升了训练和推理效率。最新推出的PP-YOLOE则进行了全面架构重构,实现了在检测精度、推理速度及部署适应性等方面全面提升。WTD-YOLO(Cao等,2025)针对缺陷目标过小问题,在PP-YOLOE基础上引入相邻层聚合注意力机制优化低层次特征提取能力,并通过相似性矩阵实现多层次间高效融合,同时结合Transformer编码器与CBAM提升复杂场景下特征表达。
YOLOv5在YOLO系列基础上引入Focus模块,并改进PANet,实现轻量化。同时增加自动锚框聚类机制,自动学习调整锚框尺寸,提升了模型对多尺度特征的提取能力。AFB-YOLOv5(Ran等,2022)利用残差连接和路径聚合网络(path aggregation network,PANet)双向路径,融合深浅层特征,并引入不同学习权重以更好地平衡WTB缺陷特征。然而,该研究中数据集规模小且仅涵盖损伤和污渍两种类型。Liu等人(2024b)同样采用结构化剪枝和知识蒸馏,训练轻量化GCBL-YOLOv5s模型,GCBL-YOLOv5s在骨干网络中用GhostBottleneckv2模块替代原始的C3模块,并集成CBAM注意力机制,保持对小特征关注的同时实现轻量化。针对颜色相近的浅色特征,Zhu等人(2022)在backbone中融合MobileNetv3Bone和GhostNet以提升浅色裂纹特征;同时用C3TR模块替换C3模块以提高对低分辨率图像的检测能力。该模型对小数据集友好且抗污垢干扰,但未提及参数量和检测速度。针对YOLOv5的其他改进措施如表5所示。
YOLOv6在YOLOv5基础上,引入了EfficientRep、RepVGG等模块以提升推理效率,具备优异的工业级部署特性。然而,其高度集成的结构在换取高效率的同时,也牺牲了灵活性与可扩展性,限制了针对WTB缺陷检测任务的深度优化,导致相关改进研究匮乏。
YOLOv7在YOLOv5基础上引入E-ELAN(efficient expandable layer aggregation network)、模型参数化与统一任务头,并结合CSM注意力、轻量化PWK卷积与BiFPN结构,同时改进Wise IoU以优化回归性能。YOLOv7尽管整体精度提升明显,但在小目标和低分辨率目标识别及多尺度适应性方面仍存不足。因此,Fu等人(2024)提出轻量化改进模型LE-YOLO。该模型结合GSConv和无参数的SimAM注意力机制,在轻量化的同时增强不同信道间的信息融合。为达成类似目标,Liao等人(2024)采用DGST模块和下采样模块替换原有的ELAN模块和MP-1模块,以减轻特征提取的困难;同时引入ECA模块和MIoU,提升定位精度和收敛效率。该模型鲁棒性较好,但其召回率表现稍逊。
YOLOv8是YOLO系列首个实现全架构重构的版本,广泛适用于从研究到工业部署的各类场景。Yu等人(2024)提出增强型YOLOv8模型,通过引入CBAM、BiFPN及EIoU提升对风电叶片多尺度缺陷的检测能力,但不适用于复杂天气且对缺陷尺寸敏感。Zou等人(2024)引入DSConv与CARAFE,增强裂纹细长结构特征提取与上下文感知,结合WIoUv3自适应梯度分配策略显著提升检测精度,计算开销稍有增加的情况下,该模型在特定数据集上准确率达100%,但泛化能力仍待验证。Cao等人(2024)结合MLCA注意力、焦点特征聚合与AFPN的改进算法,有效提升复杂背景下多尺度小目标的检测性能,在保持检测速度的同时显著提升平均精度均值(mean average precision,mAP),虽然优于现有模型,但未探讨模型规模对性能的影响。针对YOLOv8的其他改进措施如表6所示。
Transformer在自然语言处理领域取得突破,被迅速拓展至计算机视觉领域。ViT(vision Transformer)通过将图像重构为序列化表示,利用自注意力机制建立全局依赖关系,突破了传统卷积神经网络的局部建模局限。Dwivedi等人(2024)首次将ViT应用于电力系统缺陷检测,通过自注意力机制捕捉图像全局上下文依赖,最终由多层感知机实现缺陷检测。
ViT虽然在检测精度和推理速度方面均具备显著优势,但缺乏局部特征的建模能力和空间层级结构,在高分辨率图像处理和密集预测任务中存在明显局限。针对这一问题,Swin-Transformer采用局部窗口自注意力机制和窗口平移策略,在保持计算效率的同时增强局部与全局特征提取,显著提升高分辨率图像处理能力。Han等人(2025)通过采集停机状态下不同光照条件的可见光/红外图像及运行状态下的振动与声学信号,10次重复实验的结果表明Swin-Transformer模型的平均识别准确率显著优于其他对比模型,但该研究未验证风机运行状态下的模型性能。
此前,DETR(detection Transformer)首次将Transformer架构应用于目标检测,但面临收敛速度慢和小目标检测效果欠佳等局限。为此,DINO(DETR with improved denoising and anchor boxes)通过引入噪声学习与动态锚框机制,显著提升了小目标检测精度和整体性能。WTB-DINO(Gao等,2023)针对风机数据集的特点和检测要求,在主干网络中引入Focus down-sampling模块和ECA注意力机制,增强对细粒度小特征的提取能力,并在编码器末端增加多头注意力机制的并行编码器层,促进不同类型损伤特征间的交互学习。
RT-DETR(real-time detection Transformer)通过优化特征提取与查询机制并采用更高效的匹配策略,在精度与速度间实现更优平衡,适用于工业检测等高吞吐率场景。Yang等人(2025b)提出改进的Fre-DETR模型,采用轻量化FasterNet和RepConv实现高效特征提取与推理加速,并结合基于注意力的尺度内特征交互(attention-based intra-scale feature interaction,AIFI)模块和感知查询机制,实现全局特征联动,同时设计融合通道与空间注意力机制的SMLCA模块,有效提升模型对多尺度缺陷的检测能力。
基于Transformer的分割模型在该领域也有所应用。Li等人(2024)提出一种基于像素匹配的Segformer模型,结合在线难例挖掘(online hard example mining,OHEM)算法聚焦于难学习样本,通过计算实际损伤区域与预测掩码图的相似度,精确定位缺陷区域,实现WTB表面裂纹、孔洞、剥落和侵蚀等缺陷的像素级识别,并增强了模型在光照变化下的分割质量与检测鲁棒性。
学习策略作为连接模型结构与训练过程的关键桥梁,直接影响模型的收敛特性与泛化性能。在WTB缺陷检测任务中,缺陷样本稀缺、边界模糊及背景干扰等挑战凸显了传统架构优化的局限性。通过设计高效的学习策略,不仅能够实现高效的数据深层特征提取,更能缓解实际应用中常见的类不平衡、标注噪声等问题,从而显著提升检测系统的鲁棒性。
迁移学习在大规模(标注或未标注)数据上进行预训练以获得可迁移知识,随后将模型迁移至目标任务中;若下游任务无标注数据,则引入来自同任务、异分布源域的标注样本提升性能(如图6(a)所示)(Jiang等,2022)。该方法不仅使模型能够快速适应新任务并获得性能提升,同时增强了模型在不同环境条件下的检测鲁棒性(Dhiman和Nizami,2024)。
积冰检测面临严重数据缺失,Alvela Nieto等人(2023)提出一种基于深度迁移学习的检测方法,采用在ImageNet上预训练的VGG19(Visual Geometry Group 19)、MobileNetV2和Xception作为基础网络,从而有效捕获目标领域的特征表示。Yu等人(2020)同样通过ImageNet上预训练的DCNN(deep CNN)提取缺陷语义特征,并结合支持向量机(support vector machine,SVM)分类器实现高精度的缺陷识别,该方案虽存在少量误分类案例,但准确率获得显著提升。
集成学习通过整合多个弱学习器,构建更强泛化能力的模型(如图6(b)所示),有效利用各模型间的互补特性,显著增强了模型对复杂环境的鲁棒性,提升了其在应对噪声干扰、光照变化以及目标特征多样性等方面的适应能力。
Memari等人(2024)融合DenseNet的局部特征提取能力与ViT的全局依赖建模优势,采用ViT权重为0.6的加权平均集成策略,其分类准确率明显优于简单平均和堆叠方法。Pratt等人(2025)基于多变量模糊投票系统,整合VGG、Xception和ResNet 3种骨干网络的Mask R-CNN预测结果,最终模型在各类损伤检测中实现性能平和,且准确率普遍优于单一模型准确率。此外,通过调整模糊函数参数,可针对特定缺陷类型优化识别性能。
级联机制通过分阶段处理策略实现模型性能优化,其核心是将复杂任务分解为递进子任务序列(Hwang等,1996)。每一级模型基于前序输出进行渐进式特征提取与决策优化(如图6(c)所示),从而在保证计算效率的同时提升判别精度。这种层级化处理方式通过逐步细化的特征学习实现针对性决策。
在裂纹检测任务中,Wang和Zhang(2017)提出了一种结合扩展Haar-like特征与级联分类器的检测框架,以提高检测的准确性和效率。该方法通过裂纹位置检测与裂纹精度优化两阶段实现高效检测。在裂纹位置检测阶段,首先通过扩展Haar-like特征提取裂纹的边缘与纹理信息,随后结合滑动窗口和级联分类器(LogitBoost/决策树/SVM)对各候选区域进行分类;在裂纹精度优化阶段,利用基于LogitBoost的扩展级联分类器进行精细化分析。在上述基础上,Wang等人(2019b)通过并行处理和Jaya优化算法对裂纹检测流程进行了进一步优化,提升裂纹边界提取的准确性。He等人(2024)提出了一种基于“先粗后精”策略的级联检测方法,通过逐步检测高、中、低层特征,提升分类与定位的精度。
无监督学习通过从未标注数据中挖掘潜在结构与分布规律(如图6(d)所示),有效克服数据标注成本高、人工标注主观性强等局限性。这一方法显著提升了数据利用效率与模型泛化能力,为WTB缺陷检测研究开辟了新途径。例如,Wang等人(2019b)采用一类支持向量机(one-class support vector machine,OCSVM)的无监督方法,区分叶片图像中正常与异常区域。
半监督学习通过协同利用少量标注样本与大量未标注样本(如图6(e)所示),在监督学习与无监督学习之间建立了有效平衡。针对WTB缺陷检测中高质量标注依赖大量人力和专业知识的问题,半监督学习通过引入伪标签生成、对比学习及一致性正则化等策略,有效从未标注数据中挖掘潜在信息,实现数据集的高效利用与模型精度的提升。
Barker等人(2021)融合了无监督与半监督两种学习机制,显著提升了风力涡轮叶片表面缺陷的检测效率。无监督阶段采用基于U-Net架构的卷积神经网络进行图像分割,准确识别叶片结构特征以区分目标区域;随后采用融合VAE、GANomaly和PANDA的半监督生成框架,基于超像素重构与判别机制实现异常检测。Huang等人(2024)基于GAN构建半监督学习框架,其中编码器部分采用YOLOv4的主干网络,提取无标签图像的结构特征;解码器通过上采样模块将编码特征重构为图像,该设计在不依赖额外标注的前提下,有效提高了模型的泛化能力。Ye等人(2024)基于委员会机制的不确定性采样策略,利用两个模型对未标注数据进行并行检测,通过比较两模型输出的交并比(intersection-over-union,IoU)值计算成员间的不确定性度量。随着数据集扩充,模型召回率略有降低,但模型预测精度显著提高。
1) DTU-Drone。Shihavuddin团队2018年建立该数据集,包含559幅图像,分辨率为5 280 × 2 970像素(Shihavuddin等,2018)。记录裂纹、胶衣脱落、边缘侵蚀及雷击传感器等多种典型损伤,已成为该领域广泛采用的基准数据集。
2)YOLO Annotated Wind Turbine Surface Damage。2019年在Kaggle平台公开发布,共13 000幅图像,分辨率为586 × 371像素,采用YOLO格式标注裂纹、涂层破损、边缘侵蚀、雷击传感器及污渍等多种损伤类型。
3)Wind Turbine Damage new。2023年通过Roboflow平台发布,共2 248幅图像,分辨率为640 × 640像素,包含裂纹、涂层破损和雷击损伤,同时提供YOLO和VOC两种标注格式。该数据集是目前公开数据中少数包含雷击损伤标注的珍贵资源。
4)Blade30。2023年在GitHub平台发布,共1 302幅图像,包含裂纹、砂眼和边缘侵蚀等缺陷类型,所有样本均采用矩形框标注。
5)Small-WTB-Thermal1。2024年在GitHub平台开源,共1 000幅图像,分辨率为320 × 320像素,包含裂纹、砂眼和边缘侵蚀,所有样本均采用矩形框标注。
公开数据集链接如下:
另外,现有文献中涉及的私有数据集汇总如表7所示。
WTB缺陷检测中常用评估指标汇总如表8。
1)图像恢复任务中的评估指标。图像恢复旨在将低质量图像恢复为高质量图像,通过对比恢复前后图像质量,评估算法有效性。本文中所涉及到的评估指标包含峰值信噪比(peak signal-to-noise ratio,PSNR)和结构相似度指数(structural similarity index measure,SSIM)。PSNR通过比较去噪后图像的峰值与噪声的均方根进行计算。PSNR高于40 dB表示图像质量良好;30~40 dB表示图像质量中等;20~30 dB表示图像质量差;低于20 dB表示图像不可接受。SSIM考虑图像结构信息,其值越接近1,去噪后图像与原始图像越相似。
2)缺陷检测任务中的评估指标。目标检测任务分为定位和识别两大模块。分类任务中,通过衡量4种分类情况:实际正样本预测为正样本(true positive,TP)、实际正样本预测为负样本(false negative,FN)、实际负样本预测为正样本(false positive,FP)和实际负样本预测为负样本(true negative,TN),对最终分类结果进行评估。同时,为涵盖定位的准确性,通过IoU等指标进行评估。
3)计算开销与检测效率评估指标。模型的参数量和检测速度是评估模型轻量化、边缘化的重要指标。参数量的大小直接影响模型的存储需求和内存调用,检测速度直接影响模型的实时性。
通过前文对近几年WTB缺陷检测方法的分析与探讨,发现通过融合图像预处理技术、优化目标检测网络结构及合理设计学习策略,能够在兼顾工业轻量化需求的同时显著提升检测精度。但是,通过与其他领域和前沿检测模型进行对比,WTB缺陷检测过程中仍存在许多挑战。下文对目前存在的挑战及可能存在的解决方法进行总结与展望。
目前,多数公开数据集集中于裂纹和边缘侵蚀等明显缺陷,其他缺陷类型(如积冰)样本稀缺。这种类不平衡严重制约了模型对少数类缺陷的识别能力,尤其影响了我国高海拔及海上风电场景中积冰问题的研究进展。
未来研究可尝试利用生成模型实现对少数类别的图像增强或生成,按照需求灵活扩充数据集,缓解类不平衡问题,增强模型对少数类缺陷的识别能力。例如,借鉴扩散模型在隐空间根据专家知识对生成结果进行约束,从而提升生成样本的多样性与可靠性,增强模型对实际场景中少类别缺陷的判别能力。
手工标注质量会影响检测精度,导致误检漏检。尽管半监督学习可减轻对人工标注的依赖,相关研究仍多依赖人工标注数据集开展。不同研究者对缺陷的定义与标注标准存在显著差异,损伤、破损和缺陷等中文术语在语义和形态特征上高度相似,却缺乏明确的区分标准,导致标注过程依赖主观判断。此外,为提高检测精度,多数研究倾向于将形态相似的缺陷归为同一类别进行标注,这种做法虽然提升了模型性能,却可能掩盖了不同缺陷类型的本质差异。
在数据集标注方面,未来研究可探索非全监督策略自动学习缺陷特征,实现对数据集的自动标注,以降低检测模型对大量人工标注数据的依赖,进而提升模型在实际应用中的检测精度和对新增加缺陷的识别能力。例如,自监督—对比学习的基础框架(如MoCo)或根据自编码方式实现自监督学习。
风机分布地域广阔,不同区域缺陷发生概率存在显著差异,这对模型的泛化能力提出了严峻挑战。尤其在光照不足、雨雪天气或叶片积冰等复杂条件下,缺陷特征与叶片背景及环境特征高度相似,导致检测难度大幅提升。值得注意的是,当前研究多局限于单一地区的特定风电场,其模型的泛化性仍存在较大争议。
结合当前研究热点,未来研究可探索视觉大模型与持续学习、主动学习等策略的结合,构建能够适应多天气、多地域的通用检测模型。同时,利用大语言模型在特征描述与知识推理方面的优势,辅助模型理解复杂环境下的缺陷语义。
基于二维图像的视觉检测虽然能精准识别多数缺陷,但对表面特征相似的白斑、污渍等缺陷,在低光照条件下仍存在区分困难;而且对于材料内部裂纹或气孔等无表面特征的缺陷,二维图像因无法穿透物体表面而难以获取有效信息。因此,单一图像模态在缺陷检测中存在固有局限性。
在多模态数据融合检测方法的研究中,研究者应突破可见光与红外图像融合的局限,整合图像的宏观纹理、形态信息与时序信号的内部物理状态数据。该跨维度融合策略旨在利用时序信号的物理化学特征,有效提升模型在复杂工况下辨识缺陷、特别是区分表观相似缺陷的能力。
目前WTB缺陷检测算法主要集中于YOLO架构的优化,通过增强对多目标特征的提取能力和轻量化主干网络结构,实现对小RoI、低像素缺陷的检测。然而,这些改进方法存在思路趋同的问题,且基于Transformer的检测网络在该领域的应用潜力尚未得到充分挖掘。
在目标检测模型的架构拓展方面,未来的研究视野应突破当前YOLO系列模型的主导格局。随着基于Transformer的检测网络日益受到学界关注,有必要系统考察包括该架构在内的多种先进检测框架,同时探索原创网络结构在缺陷检测任务中的适用性。
基于无人机图像的叶片缺陷检测正逐渐成为新能源电力系统中的重要研究前沿与技术挑战。首先,本文对WTB常见缺陷类型进行归纳,并分析了该技术面临的主要挑战;其次,介绍了WTB缺陷目视检测中的常用预处理方法;之后,从算法层面展开讨论,详细阐述了基于两阶段模型、单阶段模型以及Transformer架构的检测网络,并从特征提取、模型优化和学习策略3个维度剖析了当前研究的核心方向;此外,阐述了现有公开数据集及私有数据集,并总结常见评估指标;最后,基于现有研究的局限性,进一步提出了创新性研究思路和发展方向。WTB缺陷检测技术的突破不仅有助于保障风机结构安全,其方法论对电力系统其他检测任务也具有重要的借鉴价值。
  • 中央高校基本科研业务费专项资金资助(2024MS112);河北省省级科技计划资助(SZX2020034)
参考文献 引证文献
排序方式:
Altice B,Nazario E,Davis M,Shekaramiz M,Moon T K and Masoum M A S. 2024b. Anomaly detection on small wind turbine blades using deep learning algorithms. Energies,17(5): #982 [DOI: 10.3390/en17050982]
Alvela Nieto M T,Gelbhardt H,Ohlendorf J H and Thoben K D. 2023. Detecting ice on wind turbine rotor blades: towards deep transfer learning for image data//Proceedings of the 6th International Conference on System-Integrated Intelligence. Genova,Italy: Springer: 459-468 [DOI: 10.1007/978-3-031-16281-7_54]
Amrollahi Biyouki S and Hwangbo H. 2023. A comprehensive survey on deep neural image deblurring [EB/OL]. [2025-10-12]. https://arxiv.org/pdf/2310.04719.pdf
Barker J W,Bhowmik N and Breckon T P. 2021. Semi-supervised surface anomaly detection of composite wind turbine blades from drone imagery [EB/OL]. [2025-10-12]. https://arxiv.org/pdf /2112.00556. pdf
Cao J F,Peng B,Gao M Z,Hao H C,Guo J H,Liu X L,et al. 2025. Detecting small damage on wind turbine surfaces using an improved YOLO in drone-captured scenes. Journal of Failure Analysis and Prevention,25(2): 725-740 [DOI: 10.1007/s11668-025-02130-w]
Cao Z,He R D,Zhang S F,Qi Z T,Li S,Liu T,et al. 2024. Integrating local channel attention and focused feature modulation for wind turbine blade defect detection. International Journal of Advanced Computer Science and Applications,15(12): 262-270 [DOI: 10.14569/IJACSA.2024.0151228]
Cawley P. 2018. Structural health monitoring: closing the gap between research and industrial deployment. Structural Health Monitoring,17(5): 1225-1244 [DOI: 10.1177/1475921717750047]
Chen X and Eder M A. 2020. A critical review of damage and failure of composite wind turbine blade structures. IOP Conference Series: Materials Science and Engineering,942: #012001 [DOI: 10.1088/1757-899X/942/1/012001]
Dai Z J. 2024. Image acquisition technology for unmanned aerial vehicles based on YOLO — Illustrated by the case of wind turbine blade inspection. Systems and Soft Computing,6: #200126 [DOI: 10.1016/j.sasc.2024.200126]
Dhiman H S and Nizami T K. 2024. Wind turbine blade erosion detection using visual inspection and transfer learning//Proceedings of 2024 International Conference on Control,Automation and Diagnosis. Paris,France: IEEE: 1-6 [DOI: 10.1109/ICCAD60883.2024.10553885]
Diaz P M and Tittus P. 2023. Fast detection of wind turbine blade damage using Cascade Mask R-DSCNN-aided drone inspection analysis. Signal,Image and Video Processing,17(5): 2333-2341 [DOI: 10.1007/s11760-022-02450-6]
Dong C,Zhang C,Li J and Liu J. 2023. Surface defect recognition of wind turbine blades based on improved YOLOX-X model. Journal of Computers,34(2): 19-27 [DOI: 10.53106/199115992023043402002]
Du Y,Wu H K and Garcia Cava D. 2023. A motion-blurred restoration method for surface damage detection of wind turbine blades. Measurement,217: #113031 [DOI: 10.1016/j.measurement.2023.113031]
Du Y,Zhou S X,Jing X J,Peng Y P,Wu H K and Kwok N. 2020. Damage detection techniques for wind turbine blades: a review. Mechanical Systems and Signal Processing,141: #106445 [DOI: 10.1016/j.ymssp.2019.106445]
Dwivedi D,Babu K V S M,Yemula P K,Chakraborty P and Pal M. 2024. Identification of surface defects on solar PV panels and wind turbine blades using attention based deep learning model. Engineering Applications of Artificial Intelligence,131: #107836 [DOI: 10.1016/j.engappai.2023.107836]
Fu Z J,Zhang F,Ren X Y,Hao B,Zhang X Y,Yin C L,et al. 2024. LE-YOLO: lightweight and efficient detection model for wind turbine blade defects based on improved YOLO. IEEE Access,12: 135985-135998 [DOI: 10.1109/ACCESS.2024.3463391]
Gao R X,Jin H Q,Wang T F,Li X Y and Liu Q P. 2025. Wind turbine equipment motion blur image restoration algorithm optimized by diffusion model denoising diffusion probabilistic model strategy. Journal of Electronic Imaging,34(2): #023002 [DOI: 10.1117/1.JEI.34.2.023002]
Gao R X,Ma Y F and Wang T F. 2023. Early stage damage detection of wind turbine blades based on UAV images and deep learning. Journal of Renewable and Sustainable Energy,15(4): #043305 [DOI: 10.1063/5.0157624]
García M􀅡rquez F P and Peco Chacón A M. 2020. A review of non-destructive testing on wind turbines blades. Renewable Energy,161: 998-1010 [DOI: 10.1016/j.renene.2020.07.145]
Gholizadeh S. 2016. A review of non-destructive testing methods of composite materials. Procedia Structural Integrity,1: 50-57 [DOI: 10.1016/j.prostr.2016.02.008]
Girshick R,Donahue J,Darrell T and Malik J,2014. Rich feature hierarchies for accurate object detection and semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition: 580-587 [DOI: 10.1109/CVPR.2014.81]
Global Wind Energy Council. 2025. GWEC Global Wind Report 2025. Brussels: Global Wind Energy Council
Gohar I,See J,Halimi A and Yew W K. 2023. Automatic defect detection in wind turbine blade images: model benchmarks and re-annotations//Proceedings of 2023 IEEE International Conference on Multimedia and Expo Workshops. Brisbane,Australia: IEEE: 290-295 [DOI: 10.1109/ICMEW59549.2023.00056]
Gómez Muñoz C Q,García Marquez F P,Hernandez Crespo B and Makaya K. 2019. Structural health monitoring for delamination detection and location in wind turbine blades employing guided waves. Wind Energy,22(5): 698-711 [DOI: 10.1002/we.2316]
Gonzalez R C and Woods R E. 2018 . Digital image processing (4th ed.). Pearson.
Guo J H,Liu C,Cao J F and Jiang D X. 2021. Damage identification of wind turbine blades with deep convolutional neural networks. Renewable Energy,174: 122-133 [DOI: 10.1016/j.renene.2021.04.040]
Hacıefendioğlu K,Başağa H B,Yavuz Z and Karimi M T. 2022. Intelligent ice detection on wind turbine blades using semantic segmentation and class activation map approaches based on deep learning method. Renewable Energy,182: 1-16 [DOI: 10.1016/j.renene.2021.10.025]
Han Y,Liu T M and Li K. 2025. Fault diagnosis of wind turbine blades under wide-weather multi-operating conditions based on multi-modal information fusion and deep learning. Structural Health Monitoring [DOI: 10.1177/14759217251333761]
He K M,Zhang X Y,Ren S Q and Sun J. 2016. Deep residual learning for image recognition//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas,USA: IEEE: 770-778 [DOI: 10.1109/CVPR.2016.90]
He Y,Niu X B,Hao C P,Li Y F,Kang L and Wang Y. 2024. An adaptive detection approach for multi-scale defects on wind turbine blade surface. Mechanical Systems and Signal Processing,219: #111592 [DOI: 10.1016/j.ymssp.2024.111592]
Huang C,Chen M H and Wang L. 2024. Semi-supervised surface defect detection of wind turbine blades with YOLOv4. Global Energy Interconnection,7(3): 284-292 [DOI: 10.1016/j.gloei.2024.06.010]
Hwang J N,You S S,Lay S R and Jou I C. 1996. The cascade-correlation learning: a projection pursuit learning perspective. IEEE Transactions on Neural Networks,7(2): 278-289 [DOI: 10.1109/72.485631]
Ji L P,Liu F and Zhu D Q. 2025. CMFCANet: a wind turbine blade damage recognition network based on color-multiscale dual-feature fusion and channel attention mechanism. IEEE Transactions on Geoscience and Remote Sensing,63: #4700612 [DOI: 10.1109/TGRS.2024.3525192]
Ji L P,Wu S L,Nie T,Yang W W and Yang J Y. 2023. Defect detection of wind turbine blade based on improved SSD. Flight Control and Detection,6(3): 63-71
季利鹏,吴世龙,聂涛,杨文威,杨迦迤. 2023. 基于改进SSD的风机叶片表面缺陷检测方法. 飞控与探测,6(3): 63-71 [DOI: 10.20249/j.cnki.2096-5974.2023.03.008]
Jia X D and Chen X. 2025. Unsupervised wind turbine blade damage detection with memory-aided denoising reconstruction. IEEE Transactions on Industrial Informatics,21(1): 762-770 [DOI: 10.1109/TII.2024.3459612]
Jiang J G,Shu Y,Wang J M and Long M S. 2022. Transferability in deep learning: a survey [EB/OL]. [2025-10-12]. https://arxiv.org/pdf /2201.05867.pdf
Katsaprakakis D A,Papadakis N and Ntintakis I. 2021. A comprehensive analysis of wind turbine blade damage. Energies,14(18): #5974 [DOI: 10.3390/en14185974]
Khan P W and Byun Y C. 2023. A review of machine learning techniques for wind turbine’s fault detection,diagnosis,and prognosis. International Journal of Green Energy,21(4): 771-786 [DOI: 10.1080/15435075.2023.2217901]
Kumar D,Pandey R C and Mishra A K. 2024. A review of image features extraction techniques and their applications in image forensic. Multimedia Tools and Applications,83(40): 87801-87902 [DOI: 10.1007/s11042-023-17950-x]
Kupyn O,Martyniuk T,Wu J R and Wang Z Y. 2019. DeblurGAN-V2: deblurring (orders-of-magnitude) faster and better//Proceedings of 2019 IEEE/CVF International Conference on Computer Vision. Seoul,Korea (South): IEEE: 8878-8887 [DOI: 10.1109/ICCV.2019.00897]
Lei M W,Wang X F,Wang M H and Cheng Y T. 2024. Improved YOLOv5 based on multi-strategy integration for multi-category wind turbine surface defect detection. Energies,17(8): #1796 [DOI: 10.3390/en17081796]
Li B,Zhang Y M,Wei L T,Wang Y and Zhai Y J. 2024. Fan blade defect detection algorithm based on improved YOLOv8. Electronic Measurement Technology,47(13): 89-99
李冰,张易牧,魏乐涛,王月,翟永杰. 2024. 基于改进YOLOv8的风机桨叶缺陷检测算法. 电子测量技术,47(13): 89-99 [DOI: 10.19651/j.cnki.emt.2415882]
Li J and Guo P. 2021. ARGAN-based surface shadow preprocessing and transfer learning fault identification of wind turbine blade. Journal of North China Electric Power University,48(2): 73-79
李健,郭鹏. 2021. 基于ARGAN表面阴影预处理与迁移学习风电机组叶片故障识别. 华北电力大学学报,48(2): 73-79 [DOI: 10.3969/j.issn.1007-2691.2021.02.09]
Li W B,Hu B,Song C,Zhao F,Ma H L and Wang Y J. 2020. An image stitching method for blades of wind turbine based on background removal preprocessing//Proceedings of the 5th International Conference on Communication,Image and Signal Processing. Chengdu,China: IEEE: 174-178 [DOI: 10.1109/CCISP51026.2020.9273502]
Li W R,Pan Z H,Zhu Q X and Du Y F. 2024. Wind turbine blade defect detection and measurement technology based on improved SegFormer and pixel matching. Optics and Laser Technology,179: #111381 [DOI: 10.1016/j.optlastec.2024.111381]
Li X Y,Long S P and Zhu J. 2020. Survey of few-shot learning based on deep neural network. Application Research of Computers,37(8): 2241-2247
李新叶,龙慎鹏,朱婧. 2020. 基于深度神经网络的少样本学习综述. 计算机应用研究,37(8): 2241-2247 [DOI: 10.19734/j.issn.1001-3695.2019.03.0036]
Li Y X,Chai Y,Hu Y Q and Yin H P. 2019. Review of imbalanced data classification methods. Control and Decision,34(4): 673-688
李艳霞,柴毅,胡友强,尹宏鹏. 2019. 不平衡数据分类方法综述. 控制与决策,34(4): 673-688 [DOI: 10.13195/j.kzyjc.2018.0865]
Liao Y K,Lyu M Y,Huang M Y,Qu M W,Zou K H,Chen L,et al. 2024. An improved YOLOv7 model for surface damage detection on wind turbine blades based on low-quality UAV images. Drones,8(9): #436 [DOI: 10.3390/drones8090436]
Liu C Y,Wu Y Q and Liu J J. 2025. Research progress of deep learning methods for insulator defect detection in UAV based aerial images. Transactions of China Electrotechnical Society,40(9): 2897-2916
刘传洋,吴一全,刘景景. 2025. 无人机航拍图像中绝缘子缺陷检测的深度学习方法研究进展. 电工技术学报,40(9): 2897-2916 [DOI: 10.19595/j.cnki.1000-6753.tces.240610]
Liu L Z,Li P R,Wang D H and Zhu S Z. 2024a. A wind turbine damage detection algorithm designed based on YOLOv8. Applied Soft Computing,154: #111364 [DOI: 10.1016/j.asoc.2024.111364]
Liu Y H,Zheng Y Q,Wei T and Li Y B. 2024b. Lightweight algorithm based on you only look once version 5 for multiple class defect detection on wind turbine blade surfaces. Engineering Applications of Artificial Intelligence,138: #109422 [DOI: 10.1016/j.engappai.2024.109422]
Liu Y H,Zheng Y Q,Shao Z F,Wei T,Cui T C and Xu R. 2024. Defect detection of the surface of wind turbine blades combining attention mechanism. Advanced Engineering Informatics,59: #102292 [DOI: 10.1016/j.aei.2023.102292]
Liu Y J,Wang Z,Wu X L,Fang F and Saqlain A S. 2023. Cloud-edge-end cooperative detection of wind turbine blade surface damage based on lightweight deep learning network. IEEE Internet Computing,27(1): 43-51 [DOI: 10.1109/MIC.2022.3175935]
Lu J L,Gao Q B and Zhou K. 2024. Inspection of wind turbine blades using image deblurring and deep learning segmentation//Proceedings of SPIE 12951,Health Monitoring of Structural and Biological Systems XVIII. Long Beach,USA: SPIE: #129510L [DOI: 10.1117/12.3009721]
Lu L,Ichimura S,Yamagishi A and Rokunohe T. 2020. Oil film detection under solar irradiation and image processing. IEEE Sensors Journal,20(6): 3070-3077 [DOI: 10.1109/JSEN.2019.2955088]
Lu R S,Wu A,Zhang T D and Wang Y H. 2018. Review on automated optical (visual) inspection and its applications in defect detection. Acta Optica Sinica,38(8): #0815002
卢荣胜,吴昂,张腾达,王永红. 2018. 自动光学(视觉)检测技术及其在缺陷检测中的应用综述. 光学学报,38(8): #0815002 [DOI: 10.3788/AOS201838.0815002]
Lyu L,Yao Z Y,Wang E M,Ren X,Pang R,Wang H,et al. 2022. Efficient and accurate damage detector for wind turbine blade images. IEEE Access,10: 123378-123386 [DOI: 10.1109/ACCESS.2022.3224446]
Ma L M,Jiang X Y,Tang Z Z,Zhi S D and Wang T Y. 2024. Wind turbine blade defect detection algorithm based on lightweight MES-YOLOv8n. IEEE Sensors Journal,24(17): 28409-28418 [DOI: 10.1109/JSEN.2024.3430351]
Mao Y L,Wang S X,Yu D L and Zhao J C. 2021. Automatic image detection of multi-type surface defects on wind turbine blades based on cascade deep learning network. Intelligent Data Analysis,25(2): 463-482 [DOI: 10.3233/IDA-205143]
Memari M,Shekaramiz M,Masoum M A S and Seibi A C. 2024. Data fusion and ensemble learning for advanced anomaly detection using multi-spectral RGB and thermal imaging of small wind turbine blades. Energies,17(3): #673 [DOI: 10.3390/en17030673]
Moll J. 2019. Numerical and experimental analysis of defect detection in jointed electromagnetic waveguides//Proceedings of the 13th European Conference on Antennas and Propagation. Krakow,Poland: IEEE: 1-4
Moll J,Simon J,Malzer M,Krozer V,Pozdniakov D,Salman R,et al. 2018. Radar imaging system for in-service wind turbine blades inspections: initial results from a field installation at a 2 MW wind turbine. Progress in Electromagnetics Research,162: 51-60 [DOI: 10.2528/PIER18021905]
Moreno S,Peña M,Toledo A,Treviño R and Ponce H. 2018. A new vision-based method using deep learning for damage inspection in wind turbine blades//Proceedings of the 15th International Conference on Electrical Engineering,Computing Science and Automatic Control. Mexico City,Mexico: IEEE: 1-5 [DOI: 10.1109/ICEEE.2018.8533924]
Nakauchi S,Nishino K and Yamashita T. 2012. Selection of optimal combinations of band-pass filters for ice detection by hyperspectral imaging. Optics Express,20(2): 986-1000 [DOI: 10.1364/OE.20.000986]
National Energy Administration. 2025. National electricity industry statistics data of 2024 [EB/OL]. [2025-10-12]. https://www.nea.gov.cn/20250121/097bfd7c1cd3498897639857d86d5dac/c.html (
国家能源局. 2025. 国家能源局发布2024年全国电力工业统计数据 [EB/OL]. [2025-10-12]. https://www.nea.gov.cn/20250121/097bfd7c1cd3498897639857d86d5dac/c.html
Nijssen R and Manrique E. 2020. Literature Review of Structural and Non-Structural Wind Turbine Blade Damage. TNO
Pan X Y,Jia N X,Mu Y Z and Gao X R. 2023. Survey of small object detection. Journal of Image and Graphics,28(9): 2587-2615
潘晓英,贾凝心,穆元震,高炫蓉. 2023. 小目标检测研究综述. 中国图象图形学报,28(9): 2587-2615 [DOI: 10.11834/jig.220455]
Parent O and Ilinca A. 2011. Anti-icing and de-icing techniques for wind turbines: critical review. Cold Regions Science and Technology,65(1): 88-96 [DOI: 10.1016/j.coldregions.2010.01.005]
Peng Y P,Tang Z,Zhao G P,Cao G Z and Wu C. 2022a. Motion blur removal for UAV-based wind turbine blade images using synthetic datasets. Remote Sensing,14(1): #87 [DOI: 10.3390/rs14010087]
Peng Y P,Wang W J,Tang Z,Cao G Z and Zhou S X. 2022b. Non-uniform illumination image enhancement for surface damage detection of wind turbine blades. Mechanical Systems and Signal Processing,170: #108797 [DOI: 10.1016/j.ymssp.2021.108797]
Pratt R,Allen C,Masoum M A S and Seibi A. 2025. Defect detection and classification on wind turbine blades using deep learning with fuzzy voting. Machines,13(4): #283 [DOI: 10.3390/machines13040283]
Qiu Z,Wang S,Zeng Z and Yu D. 2019. Automatic visual defects inspection of wind turbine blades via YOLO-based small object detection approach. Journal of Electronic Imaging,28(4): #043023 [DOI: 10.1117/1.JEI.28.4.043023]
Ran X K,Zhang S,Wang H T and Zhang Z Y. 2022. An improved algorithm for wind turbine blade defect detection. IEEE Access,10: 122171-122181 [DOI: 10.1109/ACCESS.2022.3216613]
Reddy A,Indragandhi V,Ravi L and Subramaniyaswamy V. 2019. Detection of cracks and damage in wind turbine blades using artificial intelligence-based image analytics. Measurement,147: #106823 [DOI: 10.1016/j.measurement.2019.07.051]
Rizk P,Al Saleh N,Younes R,Ilinca A and Khoder J. 2020. Hyperspectral imaging applied for the detection of wind turbine blade damage and icing. Remote Sensing Applications: Society and Environment,18: #100291 [DOI: 10.1016/j.rsase.2020.100291]
Rizvi S Z,Jamil M and Huang W M. 2024a. Pixel U-Net: an improved version of U-Net for binary segmentation of wind turbine blades. Signal,Image and Video Processing,18(8): 6299-6307 [DOI: 10.1007/s11760-024-03317-8]
Rizvi S Z,Jamil M and Huang W M. 2024b. Enhanced defect detection on wind turbine blades using binary segmentation masks and YOLO. Computers and Electrical Engineering,120: #109615 [DOI: 10.1016/j.compeleceng.2024.109615]
Rubert T,McMillan D and Niewczas P. 2018. A decision support tool to assist with lifetime extension of wind turbines. Renewable Energy,120: 423-433 [DOI: 10.1016/j.renene.2017.12.064]
Sarkar D and Gunturi S K. 2021. Wind turbine blade structural state evaluation by hybrid object detector relying on deep learning models. Journal of Ambient Intelligence and Humanized Computing,12(8): 8535-8548 [DOI: 10.1007/s12652-020-02587-7]
Shihavuddin A and Chen X. 2018. DTU-drone inspection images of wind turbine. Mendeley Data,V2 [EB/OL]. [2025-10-12]. https://data.mendeley.com/datasets/hd96prn3nc/2
Shihavuddin A,Chen X,Fedorov V,Nymark Christensen A,Andre Brogaard Riis N,Branner K,et al. 2019. Wind turbine surface damage detection by deep learning aided drone inspection analysis. Energies,12(4): #676 [DOI: 10.3390/en12040676]
Sun H H,Zhang W T,Ma D L,Zhang Y M,Li D X and Gao X D. 2024. YOLO-WTDL: a lightweight wind turbine blades defect detection model based on YOLOv8//Proceedings of the 7th Asia Conference on Energy and Electrical Engineering. Chengdu,China: IEEE: 46-51 [DOI: 10.1109/ACEEE62329.2024.10651628]
Szeliski R. 2011. Computer vision: algorithms and applications. Springer. [DOI: 10.1007/978-1-84882-935-0]
Tiwari K A,Raisutis R and Samaitis S. 2017. Signal processing methods to improve the signal-to-noise ratio (SNR) in ultrasonic non-destructive testing of wind turbine blade. Procedia Structural Integrity,5: 1184-1191 [DOI: 10.1016/j.prostr.2017.07.036]
Vincent P,Larochelle H,Bengio Y and Manzagol P A. 2008. Extracting and composing robust features with denoising autoencoders//Proceedings of the 25th International Conference on Machine Learning. Helsinki,Finland: ACM: 1096-1103 [DOI: 10.1145/1390156.1390294]
Wang H J,Shen Q Z,Dai Q,Gao Y C,Gao J and Zhang T. 2024. Evolutionary variational YOLOv8 network for fault detection in wind turbines. Computers,Materials and Continua,80(1): 625-642 [DOI: 10.32604/cmc.2024.051757]
Wang L,Liu Y B,Teng W,Huang X W and Liu J T. 2023. Research and development of nondestructive detection technology for wind turbine blades. Electric Power,56(10): 80-95
王磊,柳亦兵,滕伟,黄心伟,刘剑韬. 2023b. 风电机组叶片无损检测技术研究与进展. 中国电力,56(10): 80-95 [DOI: 10.11930/j.issn.1004-9649.202303073]
Wang L and Zhang Z J. 2017. Automatic detection of wind turbine blade surface cracks based on UAV-taken images. IEEE Transactions on Industrial Electronics,64(9): 7293-7303 [DOI: 10.1109/TIE.2017.2682037]
Wang L,Zhang Z J and Lou X. 2019a. A two-stage data-driven approach for image-based wind turbine blade crack inspections. IEEE/ASME Transactions on Mechatronics,24(3): 1271-1281 [DOI: 10.1109/TMECH.2019.2908233]
Wang Q Z,Jin G S,Li Q,Wang K,Yang Z Y and Wang H. 2021. Industrial edge computing: vision and challenges. Information and Control,50(3): 257-274
王其朝,金光淑,李庆,王锴,杨祖业,王宏. 2021. 工业边缘计算研究现状与展望. 信息与控制,50(3): 257-274 [DOI: 10.13976/j.cnki.xk.2021.1030]
Wang W J,Xue Y,He C K and Zhao Y N. 2022. Review of the typical damage and damage-detection methods of large wind turbine blades. Energies,15(15): #5672 [DOI: 10.3390/en15155672]
Wang Y F,Lu Q and Ren B B. 2023. Wind turbine crack inspection using a quadrotor with image motion blur avoided. IEEE Robotics and Automation Letters,8(2): 1069-1076 [DOI: 10.1109/LRA.2023.3236576]
Wang Y N,Yoshihashi R,Kawakami R,You S D,Harano T,Ito M,et al. 2019b. Unsupervised anomaly detection with compact deep features for wind turbine blade images taken by a drone. IPSJ Transactions on Computer Vision and Applications,11(1): #3. [DOI: 10.1186/s41074-019-0056-0]
Xu D H,Wen C B and Liu J H. 2019. Wind turbine blade surface inspection based on deep learning and UAV-taken images. Journal of Renewable and Sustainable Energy,11(5): #053305 [DOI: 10.1063/1.5113532]
Xu G D,Ye H Y and Xie H B. 2018. The current state and future development of wind turbine technology. Strategic Study of CAE,20(3): 44-50
许国东,叶杭冶,解鸿斌. 2018. 风电机组技术现状及发展方向. 中国工程科学,20(3): 44-50 [DOI: 10.15302/J-SSCAE-2018.03.007]
Yang C,Liu X,Zhou H,Ke Y and See J. 2023. Towards accurate image stitching for drone-based wind turbine blade inspection. Renewable Energy,203: 267-279 [DOI: 10.1016/j.renene.2022.12.063]
Yang C C,Ding S H and Zhou G S. 2025a. Wind turbine blade damage detection based on acoustic signals. Scientific Reports,15(1): #3930 [DOI: 10.1038/s41598-025-88276-x]
Yang G Y,Xiong W,Lei J Q and Yang K. 2024. Blade-YOLOv8: improved YOLOv8 for wind turbine blade defect detection//Proceedings of the 4th Power System and Green Energy Conference. Shanghai,China: IEEE: 209-213 [DOI: 10.1109/PSGEC62376.2024.10721051]
Yang J W,Chen X C,Cao S X,Zhao B,Tang Z H,Wang G,et al. 2025b. Research on surface loss detection algorithm of wind turbine blade based on FRE-DETR network. Journal of Applied Science and Engineering,28(12): 2529-2539 [DOI: 10.6180/jase.202512_28(12).0002]
Yang X Y,Zhang Y F,Lyu W and Wang D. 2021. Image recognition of wind turbine blade damage based on a deep learning model with transfer learning and an ensemble learning classifier. Renewable Energy,163: 386-397 [DOI: 10.1016/j.renene.2020.08.125]
Yao Y,Wang G Z and Fan J H. 2023. WT-YOLOX: an efficient detection algorithm for wind turbine blade damage based on YOLOX. Energies,16(9): #3776 [DOI: 10.3390/en16093776]
Ye X Y,Wang L,Huang C and Lou X. 2024. Wind turbine blade defect detection with a semi-supervised deep learning framework. Engineering Applications of Artificial Intelligence,136: #108908 [DOI: 10.1016/j.engappai.2024.108908]
Yu H,Wang J G,Han Y X,Fan B and Zhang C. 2024. Research on an intelligent identification method for wind turbine blade damage based on CBAM-BiFPN-YOLOV8. Processes,12(1): #205 [DOI: 10.3390/pr12010205]
Yu J F,He Y Z,Liu H,Zhang F,Li J,Sun G S,et al. 2023a. An improved U-net model for infrared image segmentation of wind turbine blade. IEEE Sensors Journal,23(2): 1318-1327 [DOI: 10.1109/JSEN.2022.3224837]
Yu J F,He Y Z,Zhang F,Sun G S,Hou Y J,Liu H,et al. 2023b. An infrared image stitching method for wind turbine blade using UAV flight data and U-Net. IEEE Sensors Journal,23(8): 8727-8736 [DOI: 10.1109/JSEN.2023.3257824]
Yu Y J,Cao H,Yan X Y,Wang T and Ge S S. 2020. Defect identification of wind turbine blades based on defect semantic features with transfer feature extractor. Neurocomputing,376: 1-9 [DOI: 10.1016/j.neucom.2019.09.071]
Zhang C,Wen C B and Liu J H. 2020. Mask-MRNet: a deep neural network for wind turbine blade fault detection. Journal of Renewable and Sustainable Energy,12(5): #053302 [DOI: 10.1063/5.0014223]
Zhang C,Yang T and Yang J. 2022. Image recognition of wind turbine blade defects using attention-based MobileNetv1-YOLOv4 and transfer learning. Sensors,22(16): #6009 [DOI: 10.3390/s22166009]
Zhang H,Xu H,Tian X,Jiang J J and Ma J Y. 2021. Image fusion meets deep learning: a survey and perspective. Information Fusion,76: 323-336 [DOI: 10.1016/j.inffus.2021.06.008]
Zhang R and Wen C B. 2022. SOD-YOLO: a small target defect detection algorithm for wind turbine blades based on improved YOLOV5. Advanced Theory and Simulations,5(7): #2100631 [DOI: 10.1002/adts.202100631]
Zhang Y,Fang Y,Gao W W,Liu X T,Yang H,Tong Y M,et al. 2024. Attention mechanism based on deep learning for defect detection of wind turbine blade via multi-scale features. Measurement Science and Technology,35(10): #105408 [DOI: 10.1088/1361-6501/ad6024]
Zhang Y S,Yang Y L,Sun J Q,Ji R,Zhang P Y and Shan H L. 2023. Surface defect detection of wind turbine based on lightweight YOLOv5s model. Measurement,220: #113222 [DOI: 10.1016/j.measurement.2023.113222]
Zhang Z Y and Shu Z R. 2024. Unmanned aerial vehicle (UAV)-assisted damage detection of wind turbine blades: a review. Energies,17(15): #3731 [DOI: 10.3390/en17153731]
Zhao Y X,Qiu Z B,Zhang Y,Quan H,Wei Z H and Zhu X. 2024. Detection of defects in wind turbin blade based on cascaded adaptive hybrid attention network. IEEE Access,12: 99349-99361 [DOI: 10.1109/ACCESS.2024.3429632]
Zhong S and Wang L P. 2025. Review of research on object detection in UAV aerial images. Laser and Optoelectronics Progress,62(10): #1000005
钟帅,王丽萍. 2025. 无人机航拍图像目标检测技术研究综述. 激光与光电子学进展,62(10): #1000005 [DOI: 10.3788/LOP242120]
Zhou L M,Li S X,Zhu Z R,Chen F K,Liu C and Meng F R. 2023. A review of transmission line defect detection research based on deep learning. Computer Science and Application,13(6): 1235-1243
周立明,李士心,朱致仁,陈范凯,刘宸,孟范润. 2023. 基于深度学习的输电线路缺陷检测研究综述. 计算机科学与应用,13(6): 1235-1243 [DOI: 10.12677/CSA.2023.136121]
Zhou W J,Wang Z J,Zhang M S and Wang L W. 2023. Wind turbine actual defects detection based on visible and infrared image fusion. IEEE Transactions on Instrumentation and Measurement,72: #3509208 [DOI: 10.1109/TIM.2023.3251413]
Zhu X X,Hang X Y,Gao X X,Yang X,Xu Z X,Wang Y,et al. 2022. Research on crack detection method of wind turbine blade based on a deep learning method. Applied Energy,328: #120241 [DOI: 10.1016/j.apenergy.2022.120241]
Zou L,Chen A Q,Li C Z,Yang X H and Sun Y B. 2024. DCW-YOLO: an improved method for surface damage detection of wind turbine blades. Applied Sciences,14(19): #8763 [DOI: 10.3390/app14198763]
Zou L,Chen A Q,Yang X H and Sun Y B. 2025. An improved method of AUD-YOLO for surface damage detection of wind turbine blades. Scientific Reports,15(1): #5833 [DOI: 10.1038/s41598-025-89864-7]
Zou L and Cheng H,2022. Research on wind turbine blade surface damage identification based on improved convolution neural network. Applied Sciences,12(18): #9338 [DOI: 10.3390/app12189338]
2026年第31卷第8期
PDF下载
24
11
引用本文
BibTeX
文章信息
doi: 10.11834/jig.250458
  • 接收时间:2025-10-12
  • 首发时间:2026-10-30
  • 出版时间:2026-08-16
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2025-10-12
  • 修回日期:2025-11-26
基金
Fundamental Research Funds for the Central Universities (2024MS112) ;The S&T Program of Hebei (SZX2020034)
中央高校基本科研业务费专项资金资助(2024MS112);河北省省级科技计划资助(SZX2020034)
作者信息
    1 华北电力大学燕赵电力实验室,保定 071003
    2 华北电力大学控制与计算机工程学院,保定 071003
    3 河北省能源电力知识计算重点实验室,保定 071003
    4 复杂能源系统智能计算教育部工程研究中心,保定 071003
    5 河北省电力物联网技术重点实验室,保定 071003

通讯作者:

苏攀,通信作者,男,副教授,主要研究方向为机器学习与医学图像处理。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/zgtxtxxb/CN/10.11834/jig.250458
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
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
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