Article(id=1149769463668589536, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149769458706723113, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2404425, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1718208000000, receivedDateStr=2024-06-13, revisedDate=1740326400000, revisedDateStr=2025-02-24, acceptedDate=null, acceptedDateStr=null, onlineDate=1752056001822, onlineDateStr=2025-07-09, pubDate=1747497600000, pubDateStr=2025-05-18, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1752056001822, onlineIssueDateStr=2025-07-09, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1752056001822, creator=13701087609, updateTime=1752056001822, updator=13701087609, issue=Issue{id=1149769458706723113, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='14', pageStart='5705', pageEnd='6154', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1752056000638, creator=13701087609, updateTime=1768456798957, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1218559392753041779, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149769458706723113, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1218559392753041780, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149769458706723113, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=6044, endPage=6053, ext={EN=ArticleExt(id=1149769463848944610, articleId=1149769463668589536, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=A Road Crack Detection Algorithm Based on Improved YOLOv8n, columnId=1156262728772735295, journalTitle=Science Technology and Engineering, columnName=Papers·Traffics and Transportations, runingTitle=null, highlight=null, articleAbstract=

Addressing challenges such as large memory footprint, high computational complexity, and insufficient real-time detection speed in road crack detection models for complex scenarios, a highly efficient and precise algorithm named FCG-YOLO was proposed. Lightweight modules and attention mechanisms were integrated, and traditional feature fusion pyramids were enhanced.The algorithm incorporates PConv into the residual calculation module of YOLOv8n to introduce the improved C2f_Faster structure, thereby reducing model parameters and computational complexity. To enhance detection accuracy, GAM(global attention mechanism) was introduced into the backbone, and the Feature Fusion Pyramid SPPF was improved to SPPFCSPC module, enhancing the model’s ability to represent and fuse features of road cracks.The impact of each module on algorithm performance was verified through ablation experiments, identifying a lightweight and accurate model configuration. Furthermore, the robustness and generalization of the algorithm were explored in practical application scenarios.FCG-YOLO demonstrates outstanding detection efficiency, achieving a detection accuracy of 90.3% mAP50 and 74.4% mAP50-95 on the validation set, with a detection speed of 345 frames per second. These results highlight its high detection efficiency and significant practical value.

, correspAuthors=Mamat TURSUN, 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, authorCompany=null, fund=null, authors=null, authorsList=Mamat TURSUN, Jian-zhuo QIU, Jian LIU, Han-chen DU, Xing-lin ZHU, Li XU), CN=ArticleExt(id=1149769500804956890, articleId=1149769463668589536, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=基于改进YOLOv8n的道路裂缝检测算法, columnId=1156262730664366426, journalTitle=科学技术与工程, columnName=论文·交通运输, runingTitle=null, highlight=null, articleAbstract=

针对当前道路裂缝检测算法检测速度慢、且对于复杂道路条件下的路面裂缝识别精度低、误检漏检率高等问题,设计了一种基于YOLOv8的道路裂缝检测改进算法FCG-YOLO。首先,提出了改进的C2f_Faster结构以降低模型参数量和减少计算复杂度。其次,为了提高检测准确率,在骨干部分引入全局注意力机制GAM,增强算法对小目标裂缝的识别能力,同时改进特征融合金字塔SPPF为SPPFCSPC模块,用于增强算法的特征融合能力。最后,通过消融实验验证了各模块对算法性能的影响,找到了兼具轻量性与准确性的模型配置,并结合实际应用场景,探究了算法的鲁棒性与泛化性。FCG-YOLO在检测效率上表现出色,验证集上的检测精度mAP50与mAP50-95分别达到了90.3%、74.4%,且每秒检测速度提升13.8%,达到345帧,具备较高的检测效率与良好的实际应用价值。

, correspAuthors=吐尔逊·买买提, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=45IjIKLwz7ji7xVHo7KFlw==, magXml=UdvUQA4hk7sdKUUsvXpR/w==, pdfUrl=null, pdf=6XlVJQ9+Q7n5hvDyIcDwtA==, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=BVpaq4d548/Ap+5RD1acSw==, mapNumber=null, authorCompany=null, fund=null, authors=

吐尔逊·买买提(1975—),男,维吾尔族,新疆阿克苏人,副教授,博士。研究方向:交通工程,数据挖掘。E-mail:

, authorsList=吐尔逊·买买提, 邱建卓, 刘健, 杜含辰, 朱兴林, 徐粒)}, authors=[Author(id=1172929905504436995, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=tursun@xjau.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1172929905638654726, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, authorId=1172929905504436995, language=EN, stringName=Mamat TURSUN, firstName=Mamat, middleName=null, lastName=TURSUN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1. College of Transportation & Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1172929905785455367, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, authorId=1172929905504436995, language=CN, stringName=吐尔逊·买买提, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1. 新疆农业大学交通与物流工程学院, 乌鲁木齐 830052, bio={"content":"

吐尔逊·买买提(1975—),男,维吾尔族,新疆阿克苏人,副教授,博士。研究方向:交通工程,数据挖掘。E-mail:

"}, bioImg=null, bioContent=

吐尔逊·买买提(1975—),男,维吾尔族,新疆阿克苏人,副教授,博士。研究方向:交通工程,数据挖掘。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1172929905177281273, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, xref=null, ext=[AuthorCompanyExt(id=1172929905181475578, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905177281273, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. College of Transportation & Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China), AuthorCompanyExt(id=1172929905189864187, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905177281273, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. 新疆农业大学交通与物流工程学院, 乌鲁木齐 830052)])]), Author(id=1172929905844175625, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1172929905936450315, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, authorId=1172929905844175625, language=EN, stringName=Jian-zhuo QIU, firstName=Jian-zhuo, middleName=null, lastName=QIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1. College of Transportation & Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1172929905995170572, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, authorId=1172929905844175625, language=CN, stringName=邱建卓, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1. 新疆农业大学交通与物流工程学院, 乌鲁木齐 830052, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1172929905177281273, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, xref=null, ext=[AuthorCompanyExt(id=1172929905181475578, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905177281273, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. College of Transportation & Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China), AuthorCompanyExt(id=1172929905189864187, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905177281273, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. 新疆农业大学交通与物流工程学院, 乌鲁木齐 830052)])]), Author(id=1172929906053890830, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1172929906125194000, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, authorId=1172929906053890830, language=EN, stringName=Jian LIU, firstName=Jian, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2. Key Laboratory of Highway Engineering Technology in Arid Desert Areas of Transport Industry, Urumqi 830099, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1172929906196497169, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, authorId=1172929906053890830, language=CN, stringName=刘健, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2. 干旱荒漠区公路工程技术交通运输行业重点实验室, 乌鲁木齐 830099, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1172929905265361660, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, xref=null, ext=[AuthorCompanyExt(id=1172929905273750269, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905265361660, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2. Key Laboratory of Highway Engineering Technology in Arid Desert Areas of Transport Industry, Urumqi 830099, China), AuthorCompanyExt(id=1172929905277944574, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905265361660, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2. 干旱荒漠区公路工程技术交通运输行业重点实验室, 乌鲁木齐 830099)])]), Author(id=1172929906259411731, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1172929906318131989, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, authorId=1172929906259411731, language=EN, stringName=Han-chen DU, firstName=Han-chen, middleName=null, lastName=DU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3. Xinjiang Highway Bridge Testing Center Co., Ltd., Urumqi 830099, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1172929906393629462, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, authorId=1172929906259411731, language=CN, stringName=杜含辰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3. 新疆公路桥梁试验检测中心有限责任公司, 乌鲁木齐 830099, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1172929905319887615, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, xref=null, ext=[AuthorCompanyExt(id=1172929905340859136, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905319887615, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3. Xinjiang Highway Bridge Testing Center Co., Ltd., Urumqi 830099, China), AuthorCompanyExt(id=1172929905345053441, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905319887615, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3. 新疆公路桥梁试验检测中心有限责任公司, 乌鲁木齐 830099)])]), Author(id=1172929906464932632, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1172929906532041498, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, authorId=1172929906464932632, language=EN, stringName=Xing-lin ZHU, firstName=Xing-lin, middleName=null, lastName=ZHU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1. College of Transportation & Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1172929906599150363, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, authorId=1172929906464932632, language=CN, stringName=朱兴林, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1. 新疆农业大学交通与物流工程学院, 乌鲁木齐 830052, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1172929905177281273, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, xref=null, ext=[AuthorCompanyExt(id=1172929905181475578, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905177281273, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. College of Transportation & Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China), AuthorCompanyExt(id=1172929905189864187, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905177281273, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. 新疆农业大学交通与物流工程学院, 乌鲁木齐 830052)])]), Author(id=1172929906662064925, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, orderNo=5, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1172929906783699743, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, authorId=1172929906662064925, language=EN, stringName=Li XU, firstName=Li, middleName=null, lastName=XU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1. College of Transportation & Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1172929906901140256, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, authorId=1172929906662064925, language=CN, stringName=徐粒, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1. 新疆农业大学交通与物流工程学院, 乌鲁木齐 830052, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1172929905177281273, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, xref=null, ext=[AuthorCompanyExt(id=1172929905181475578, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905177281273, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. College of Transportation & Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China), AuthorCompanyExt(id=1172929905189864187, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905177281273, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. 新疆农业大学交通与物流工程学院, 乌鲁木齐 830052)])])], keywords=[Keyword(id=1172929907018580769, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, orderNo=1, keyword=road cracks), Keyword(id=1172929907094078242, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, orderNo=2, keyword=YOLOv8n), Keyword(id=1172929907198935843, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, orderNo=3, keyword=object detection), Keyword(id=1172929907278627620, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, orderNo=4, keyword=deep learning), Keyword(id=1172929907354125093, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, orderNo=5, keyword=lightweight models), Keyword(id=1172929907542868774, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, orderNo=1, keyword=路面裂缝), Keyword(id=1172929907605783335, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, orderNo=2, keyword=YOLOv8n), Keyword(id=1172929907672892200, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, orderNo=3, keyword=目标检测), Keyword(id=1172929907752583977, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, orderNo=4, keyword=深度学习), Keyword(id=1172929907941327658, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, orderNo=5, keyword=轻量化)], refs=[Reference(id=1172929910722151247, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2021, volume=37, issue=5, pageStart=29, pageEnd=34, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=任传亭, 任广海, 李森哲, journalName=公路交通技术, refType=null, unstructuredReference=任传亭, 任广海, 李森哲. 公路沥青路面检测技术发展综述[J]. 公路交通技术, 2021, 37(5): 29-34., articleTitle=公路沥青路面检测技术发展综述, refAbstract=null), Reference(id=1172929910793454416, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2021, volume=37, issue=5, pageStart=29, pageEnd=34, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=Ren Chuanting, Ren Guanghai, Li Senzhe, journalName=Highway Traffic Technology, refType=null, unstructuredReference=Ren Chuanting, Ren Guanghai, Li Senzhe. Review of development of asphalt pavement detection technology for highways[J]. Highway Traffic Technology, 2021, 37(5): 29-34., articleTitle=Review of development of asphalt pavement detection technology for highways, refAbstract=null), Reference(id=1172929910889923409, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2024, volume=37, issue=2, pageStart=1, pageEnd=15, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=刘宇飞, 冯楚乔, 陈伟乐, journalName=中国公路学报, refType=null, unstructuredReference=刘宇飞, 冯楚乔, 陈伟乐, 等. 基于机器视觉法的桥梁表观病害检测研究综述[J]. 中国公路学报, 2024, 37(2): 1-15., articleTitle=基于机器视觉法的桥梁表观病害检测研究综述, refAbstract=null), Reference(id=1172929910998975314, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2024, volume=37, issue=2, pageStart=1, pageEnd=15, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=Liu Yufei, Feng Chuqiao, Chen Weile, journalName=China Journal of Highway and Transport, refType=null, unstructuredReference=Liu Yufei, Feng Chuqiao, Chen Weile, et al. Review of research on bridge surface defect detection based on machine vision[J]. China Journal of Highway and Transport, 2024, 37(2): 1-15., articleTitle=Review of research on bridge surface defect detection based on machine vision, refAbstract=null), Reference(id=1172929911057695571, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=11, pageStart=362, pageEnd=365, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=铁瑛, 朱空军, 朱振伟, journalName=机械设计与制造, refType=null, unstructuredReference=铁瑛, 朱空军, 朱振伟, 等. 改进YOLOv3的无纺布表面缺陷检测研究[J]. 机械设计与制造, 2024(11): 362-365., articleTitle=改进YOLOv3的无纺布表面缺陷检测研究, refAbstract=null), Reference(id=1172929911141581652, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=11, pageStart=362, pageEnd=365, url=null, language=null, rfNumber=[3], rfOrder=5, authorNames=Tie Ying, Zhu Kongjun, Zhu Zhenwei, journalName=Mechanical Design and Manufacturing, refType=null, unstructuredReference=Tie Ying, Zhu Kongjun, Zhu Zhenwei, et al. Research on improved YOLOv3 for nonwoven fabric surface defect detection[J]. Mechanical Design and Manufacturing, 2024(11): 362-365., articleTitle=Research on improved YOLOv3 for nonwoven fabric surface defect detection, refAbstract=null), Reference(id=1172929911225467733, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=15, pageStart=6338, pageEnd=6345, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=王龙业, 黄鋆, 曾晓莉, journalName=科学技术与工程, refType=null, unstructuredReference=王龙业, 黄鋆, 曾晓莉. 基于YOLO-PCB的印刷电路板裸板缺陷检测[J]. 科学技术与工程, 2024, 24(15): 6338-6345., articleTitle=基于YOLO-PCB的印刷电路板裸板缺陷检测, refAbstract=null), Reference(id=1172929911292576598, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=15, pageStart=6338, pageEnd=6345, url=null, language=null, rfNumber=[4], rfOrder=7, authorNames=Wang Longye, Huang Jun, Zeng Xiaoli, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Wang Longye, Huang Jun, Zeng Xiaoli. Bare board defect detection of PCB based on YOLO-PCB[J]. Science Technology and Engineering, 2024, 24(15): 6338-6345., articleTitle=Bare board defect detection of PCB based on YOLO-PCB, refAbstract=null), Reference(id=1172929911401628503, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2025, volume=44, issue=1, pageStart=102, pageEnd=113, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=张在岩, 宋伟东, 邬嘉晨, journalName=辽宁工程技术大学学报(自然科学版), refType=null, unstructuredReference=张在岩, 宋伟东, 邬嘉晨. 改进YOLOv5的复杂场景下水泥路面病害检测[J]. 辽宁工程技术大学学报(自然科学版), 2025, 44(1): 102-113., articleTitle=改进YOLOv5的复杂场景下水泥路面病害检测, refAbstract=null), Reference(id=1172929911493903192, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2025, volume=44, issue=1, pageStart=102, pageEnd=113, url=null, language=null, rfNumber=[5], rfOrder=9, authorNames=Zhang Zaiyan, Song Weidong, Wu Jiachen, journalName=Journal of Liaoning University of Engineering and Technology (Natural Science Edition), refType=null, unstructuredReference=Zhang Zaiyan, Song Weidong, Wu Jiachen. Improved YOLOv5 for cement pavement disease detection in complex scenarios[J]. Journal of Liaoning University of Engineering and Technology (Natural Science Edition), 2025, 44(1): 102-113., articleTitle=Improved YOLOv5 for cement pavement disease detection in complex scenarios, refAbstract=null), Reference(id=1172929911569400665, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=28, pageStart=12159, pageEnd=12167, url=null, language=null, rfNumber=[6], rfOrder=10, authorNames=蒲玲玲, 杨柳, journalName=科学技术与工程, refType=null, unstructuredReference=蒲玲玲, 杨柳. 改进YOLOv5的多车辆目标实时检测及跟踪算法[J]. 科学技术与工程, 2023, 23(28): 12159-12167., articleTitle=改进YOLOv5的多车辆目标实时检测及跟踪算法, refAbstract=null), Reference(id=1172929911615538010, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=28, pageStart=12159, pageEnd=12167, url=null, language=null, rfNumber=[6], rfOrder=11, authorNames=Pu Lingling, Yang Liu, journalName=Science Techno-logy and Engineering, refType=null, unstructuredReference=Pu Lingling, Yang Liu. Improved real-time detection and tracking algorithm for multi vehicle targets in YOLOv5[J]. Science Techno-logy and Engineering, 2023, 23(28): 12159-12167., articleTitle=Improved real-time detection and tracking algorithm for multi vehicle targets in YOLOv5, refAbstract=null), Reference(id=1172929911678452571, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2023, volume=82, issue=17, pageStart=26063, pageEnd=26087, url=null, language=null, rfNumber=[7], rfOrder=12, authorNames=Zhang S, Che S, Liu Z, journalName=Multimedia Tools and Applications, refType=null, unstructuredReference=Zhang S, Che S, Liu Z, et al. A real-time and lightweight traffic sign detection method based on ghost-YOLO[J]. Multimedia Tools and Applications, 2023, 82(17): 26063-26087., articleTitle=A real-time and lightweight traffic sign detection method based on ghost-YOLO, refAbstract=null), Reference(id=1172929911737172828, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2023, volume=46, issue=14, pageStart=56, pageEnd=62, url=null, language=null, rfNumber=[8], rfOrder=13, authorNames=潘桂霞, 赖惠成, 王同官, journalName=现代电子技术, refType=null, unstructuredReference=潘桂霞, 赖惠成, 王同官, 等. 一种Yolov5颈部细化的小交通标志检测算法[J]. 现代电子技术, 2023, 46(14): 56-62., articleTitle=一种Yolov5颈部细化的小交通标志检测算法, refAbstract=null), Reference(id=1172929911791698781, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2023, volume=46, issue=14, pageStart=56, pageEnd=62, url=null, language=null, rfNumber=[8], rfOrder=14, authorNames=Pan Guixia, Lai Huicheng, Wang Tongguan, journalName=Modern Electronics Technique, refType=null, unstructuredReference=Pan Guixia, Lai Huicheng, Wang Tongguan, et al. A YOLOv5-based algorithm for fine neck refinement of small traffic signs detection[J]. Modern Electronics Technique, 2023, 46(14): 56-62., articleTitle=A YOLOv5-based algorithm for fine neck refinement of small traffic signs detection, refAbstract=null), Reference(id=1172929911850419038, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2023, volume=32, issue=6, pageStart=204, pageEnd=211, url=null, language=null, rfNumber=[9], rfOrder=15, authorNames=曹建荣, 庄园, 汪明, journalName=计算机系统应用, refType=null, unstructuredReference=曹建荣, 庄园, 汪明, 等. 基于ECA的YOLOv5水下鱼类目标检测[J]. 计算机系统应用, 2023, 32(6): 204-211., articleTitle=基于ECA的YOLOv5水下鱼类目标检测, refAbstract=null), Reference(id=1172929911896556383, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2023, volume=32, issue=6, pageStart=204, pageEnd=211, url=null, language=null, rfNumber=[9], rfOrder=16, authorNames=Cao Jianrong, Zhuang Yuan, Wang Ming, journalName=Computer Systems and Applications, refType=null, unstructuredReference=Cao Jianrong, Zhuang Yuan, Wang Ming, et al. Underwater fish target detection based on YOLOv5 with ECA[J]. Computer Systems and Applications, 2023, 32(6): 204-211., articleTitle=Underwater fish target detection based on YOLOv5 with ECA, refAbstract=null), Reference(id=1172929911955276640, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=17, authorNames=周丽芳, 邓广, 李伟生, journalName=一种基于SE-YOLOv3的轻量级遥感目标检测方法: CN202011310176.9, refType=null, unstructuredReference=周丽芳, 邓广, 李伟生, 等. 一种基于SE-YOLOv3的轻量级遥感目标检测方法: CN202011310176.9[P]. 2024-06-12., articleTitle=null, refAbstract=null), Reference(id=1172929912005608289, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=18, authorNames=Zhou Lifang, Deng Guang, Li Weisheng, journalName=Lightweight remote sensing object detection method based on SE-YOLOv3: CN202011310176.9, refType=null, unstructuredReference=Zhou Lifang, Deng Guang, Li Weisheng, et al. Lightweight remote sensing object detection method based on SE-YOLOv3: CN202011310176.9[P]. 2024-06-12., articleTitle=null, refAbstract=null), Reference(id=1172929912072717154, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=2022, pageEnd=2033, url=null, language=null, rfNumber=[11], rfOrder=19, authorNames=Zhou W, Ma Y, Xiao S, journalName=7th International Conference on Transportation Information and Safety (ICTIS), refType=null, unstructuredReference=Zhou W, Ma Y, Xiao S, et al. SE-yolov5 inland waterway vessel target recognition based on CGAN image enhancement[C]// 7th International Conference on Transportation Information and Safety (ICTIS). Xi'an: IEEE, 2023: 2022-2033., articleTitle=SE-yolov5 inland waterway vessel target recognition based on CGAN image enhancement, refAbstract=null), Reference(id=1172929912131437411, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2023, volume=26, issue=12, pageStart=46, pageEnd=51, url=null, language=null, rfNumber=[12], rfOrder=20, authorNames=雷源毅, 朱文球, 廖欢, journalName=软件工程, refType=null, unstructuredReference=雷源毅, 朱文球, 廖欢. 复杂场景下的改进YOLOv8n安全帽佩戴检测算法[J]. 软件工程, 2023, 26(12): 46-51., articleTitle=复杂场景下的改进YOLOv8n安全帽佩戴检测算法, refAbstract=null), Reference(id=1172929912185963364, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2023, volume=26, issue=12, pageStart=46, pageEnd=51, url=null, language=null, rfNumber=[12], rfOrder=21, authorNames=Lei Yuanyi, Zhu Wenqiu, Liao Huan, journalName=Software Engineering, refType=null, unstructuredReference=Lei Yuanyi, Zhu Wenqiu, Liao Huan. Improved YOLOv8n algorithm for safety helmet detection in complex scenes[J]. Software Engineering, 2023, 26(12): 46-51., articleTitle=Improved YOLOv8n algorithm for safety helmet detection in complex scenes, refAbstract=null), Reference(id=1172929912261460837, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2024, volume=60, issue=17, pageStart=167, pageEnd=178, url=null, language=null, rfNumber=[13], rfOrder=22, authorNames=李岩超, 史卫亚, 冯灿, journalName=计算机工程与应用, refType=null, unstructuredReference=李岩超, 史卫亚, 冯灿. 面向无人机航拍小目标检测的轻量级YOLOv8检测算法[J]. 计算机工程与应用, 2024, 60(17): 167-178., articleTitle=面向无人机航拍小目标检测的轻量级YOLOv8检测算法, refAbstract=null), Reference(id=1172929912341152614, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2024, volume=60, issue=17, pageStart=167, pageEnd=178, url=null, language=null, rfNumber=[13], rfOrder=23, authorNames=Li Yanchao, Shi Weiya, Feng Can, journalName=Computer Engineering and Applications, refType=null, unstructuredReference=Li Yanchao, Shi Weiya, Feng Can. Lightweight YOLOv8 detection algorithm for small target detection in UAV aerial imagery[J]. Computer Engineering and Applications, 2024, 60(17): 167-178., articleTitle=Lightweight YOLOv8 detection algorithm for small target detection in UAV aerial imagery, refAbstract=null), Reference(id=1172929912416650087, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=12021, pageEnd=12031, url=null, language=null, rfNumber=[14], rfOrder=24, authorNames=Chen J kao S, He H, journalName=2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), refType=null, unstructuredReference=Chen J kao S, He H, et al. Run, don’t walk: chasing higher FLOPS for faster neural networks[C]// 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Vancouver: IEEE, 2023: 12021-12031., articleTitle=null, refAbstract=null), Reference(id=1172929912504730472, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, doi=null, pmid=null, pmcid=null, year=2015, volume=37, issue=9, pageStart=1904, pageEnd=1916, url=null, language=null, rfNumber=[15], rfOrder=25, authorNames=He K, Zhang X, Ren S, journalName=IEEE transactions on Pattern Analysis and Machine Intelligence, refType=null, unstructuredReference=He K, Zhang X, Ren S, et al. Spatial pyramidpooling in deep convolutional networks for visual recognition[J]. IEEE transactions on Pattern Analysis and Machine Intelligence, 2015, 37(9): 1904-1916., articleTitle=Spatial pyramidpooling in deep convolutional networks for visual recognition, refAbstract=null)], funds=[Fund(id=1172929910516630349, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, awardId=51768071, language=CN, fundingSource=国家自然科学基金(51768071), fundOrder=null, country=null), Fund(id=1172929910587933518, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, awardId=JCZXXJAU2023001, language=CN, fundingSource=新疆交通投资(集团)有限责任公司资助项目(JCZXXJAU2023001), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1172929905177281273, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, xref=null, ext=[AuthorCompanyExt(id=1172929905181475578, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905177281273, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. College of Transportation & Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China), AuthorCompanyExt(id=1172929905189864187, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905177281273, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. 新疆农业大学交通与物流工程学院, 乌鲁木齐 830052)]), AuthorCompany(id=1172929905265361660, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, xref=null, ext=[AuthorCompanyExt(id=1172929905273750269, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905265361660, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2. Key Laboratory of Highway Engineering Technology in Arid Desert Areas of Transport Industry, Urumqi 830099, China), AuthorCompanyExt(id=1172929905277944574, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905265361660, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2. 干旱荒漠区公路工程技术交通运输行业重点实验室, 乌鲁木齐 830099)]), AuthorCompany(id=1172929905319887615, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, xref=null, ext=[AuthorCompanyExt(id=1172929905340859136, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905319887615, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3. Xinjiang Highway Bridge Testing Center Co., Ltd., Urumqi 830099, China), AuthorCompanyExt(id=1172929905345053441, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, companyId=1172929905319887615, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3. 新疆公路桥梁试验检测中心有限责任公司, 乌鲁木齐 830099)])], figs=[ArticleFig(id=1172929908159431468, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, label=Fig.1, caption=FCG-YOLO structure diagram, figureFileSmall=C5pyQgqBzGNFTKwDbKMH3A==, figureFileBig=13kkyjeQAUDwQB07pRgBWw==, tableContent=null), ArticleFig(id=1172929908222346029, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, label=图1, caption=FCG-YOLO结构图, figureFileSmall=C5pyQgqBzGNFTKwDbKMH3A==, figureFileBig=13kkyjeQAUDwQB07pRgBWw==, tableContent=null), ArticleFig(id=1172929908302037806, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, label=Fig.2, caption=PConv schematic diagram, figureFileSmall=abPl6EB4SdHvpGhxOE+WCA==, figureFileBig=NT8tUxboxI/6XnluBFeReQ==, tableContent=null), ArticleFig(id=1172929908390118191, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, label=图2, caption=PConv示意图

W为宽度;Cp为进行空间卷积部分的长度;Filters为卷积核;k为卷积核大小

, figureFileSmall=abPl6EB4SdHvpGhxOE+WCA==, figureFileBig=NT8tUxboxI/6XnluBFeReQ==, tableContent=null), ArticleFig(id=1172929908461421360, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, label=Fig.3, caption=C2f_Faster module, figureFileSmall=qNXez+vumXbS+gXRCeyAUg==, figureFileBig=KS+mybuxBpg0nYUQfpy35Q==, tableContent=null), ArticleFig(id=1172929908532724529, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, label=图3, caption=C2f_Faster模块, figureFileSmall=qNXez+vumXbS+gXRCeyAUg==, figureFileBig=KS+mybuxBpg0nYUQfpy35Q==, tableContent=null), ArticleFig(id=1172929908583056178, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, label=Fig.4, caption=SPPCSPC and SPPFCSPC structure diagram, figureFileSmall=o9ztpPzv9xiIafnM5RkhrQ==, figureFileBig=gtdAZLgQkFsKzj9S/qH5cw==, tableContent=null), ArticleFig(id=1172929908641776435, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, label=图4, caption=SPPCSPC与SPPFCSPC结构图, figureFileSmall=o9ztpPzv9xiIafnM5RkhrQ==, figureFileBig=gtdAZLgQkFsKzj9S/qH5cw==, tableContent=null), ArticleFig(id=1172929908721468212, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, label=Fig.5, caption=GAM attention module, figureFileSmall=diIMMEeD60srn/4eBC9M9w==, figureFileBig=hZNG9RRKqQnxhnMXhiMZDg==, tableContent=null), ArticleFig(id=1172929908784382773, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, label=图5, caption=GAM注意力模块, figureFileSmall=diIMMEeD60srn/4eBC9M9w==, figureFileBig=hZNG9RRKqQnxhnMXhiMZDg==, tableContent=null), ArticleFig(id=1172929908843103030, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, label=Fig.6, caption=Confusion matrix, figureFileSmall=5cY8S+wlUsr49UHjNwnj8g==, figureFileBig=uqxIwnerTul0zMJPYEVrSA==, tableContent=null), ArticleFig(id=1172929908910211895, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, label=图6, caption=混淆矩阵, figureFileSmall=5cY8S+wlUsr49UHjNwnj8g==, figureFileBig=uqxIwnerTul0zMJPYEVrSA==, tableContent=null), ArticleFig(id=1172929908973126456, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, label=Fig.7, caption=Attention results visualization, figureFileSmall=/90/QiA4txnaHCl1wh0Bmw==, figureFileBig=XYmwIevAHrSyv+J/8cGxvw==, tableContent=null), ArticleFig(id=1172929909040235321, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, label=图7, caption=注意力结果可视化, figureFileSmall=/90/QiA4txnaHCl1wh0Bmw==, figureFileBig=XYmwIevAHrSyv+J/8cGxvw==, tableContent=null), ArticleFig(id=1172929909094761274, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, label=Fig.8, caption=Detection results of different algorithms in normal scenarios, figureFileSmall=6gPk1sqVsYtOM0xgPgpe3Q==, figureFileBig=hG0ltHLpYOuNIlxm/eeP8Q==, tableContent=null), ArticleFig(id=1172929909149287228, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, label=图8, caption=正常场景下不同算法检测结果, figureFileSmall=6gPk1sqVsYtOM0xgPgpe3Q==, figureFileBig=hG0ltHLpYOuNIlxm/eeP8Q==, tableContent=null), ArticleFig(id=1172929909220590397, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, label=Fig.9, caption=Detection results of different algorithms in low-exposure scenarios, figureFileSmall=ke68rM6I3gOoLgisl32fcQ==, figureFileBig=AeI1awAOZblTEngU4ZCFtw==, tableContent=null), ArticleFig(id=1172929909283504958, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, label=图9, caption=曝光不足场景下不同算法检测结果, figureFileSmall=ke68rM6I3gOoLgisl32fcQ==, figureFileBig=AeI1awAOZblTEngU4ZCFtw==, tableContent=null), ArticleFig(id=1172929909338030911, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, label=Fig.10, caption=Detection results of different algorithms in scenes with light interference, figureFileSmall=jOdNPP/IYpmcwScJ1p/12g==, figureFileBig=+z41zTuE6aV+YXZA+K1Ylw==, tableContent=null), ArticleFig(id=1172929909388362560, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, label=图10, caption=光照干扰场景下不同算法检测结果, figureFileSmall=jOdNPP/IYpmcwScJ1p/12g==, figureFileBig=+z41zTuE6aV+YXZA+K1Ylw==, tableContent=null), ArticleFig(id=1172929909468054337, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, label=Table 1, caption=

Training parameter settings

, figureFileSmall=null, figureFileBig=null, tableContent=
参数名称 参数设置 参数名称 参数设置
Optimizer Adam workers 4
epochs 300 IoU 0.7
patience 50 lr0 0.001
batch 16 lrf 0.01
imgsz 640 momentum 0.937
), ArticleFig(id=1172929909535163202, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, label=表1, caption=

训练参数设置表

, figureFileSmall=null, figureFileBig=null, tableContent=
参数名称 参数设置 参数名称 参数设置
Optimizer Adam workers 4
epochs 300 IoU 0.7
patience 50 lr0 0.001
batch 16 lrf 0.01
imgsz 640 momentum 0.937
), ArticleFig(id=1172929909656798019, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, label=Table 2, caption=

Attention comparison results

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 mAP50/
%
mAP50-
95/%
帧率 参数
量/106
GFLOPs 权重/
MB
YOLOv8n 83.6 63.9 303 3.01 8.2 6.3
YOLOv8n+
ECA
81.4 61.3 244 3.16 8.3 6.3
YOLOv8n+
SA
86.0 69.0 286 3.01 8.1 5.99
YOLOv8n+
GAM
88.1 70.5 227 3.45 8.5 7.1
), ArticleFig(id=1172929909811987268, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, label=表2, caption=

注意力对比结果

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 mAP50/
%
mAP50-
95/%
帧率 参数
量/106
GFLOPs 权重/
MB
YOLOv8n 83.6 63.9 303 3.01 8.2 6.3
YOLOv8n+
ECA
81.4 61.3 244 3.16 8.3 6.3
YOLOv8n+
SA
86.0 69.0 286 3.01 8.1 5.99
YOLOv8n+
GAM
88.1 70.5 227 3.45 8.5 7.1
), ArticleFig(id=1172929909879096133, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, label=Table 3, caption=

Comparison of results with different positions of introducing C2fFaster module

, figureFileSmall=null, figureFileBig=null, tableContent=
方案 名称 mAP50/% mAP50-95/% 帧率 参数量/106 GFLOPs 权重/MB
无改进 YOLOv8n 83.6 63.9 303 3.01 8.2 6.3
方案一 替换检测头 80.0 61.3 323 2.67 7.5 5.6
方案二 替换骨干 78.8 57.1 286 2.69 7.4 5.3
方案三 替换全部 75.0 52.5 417 2.34 6.5 4.75
), ArticleFig(id=1172929909925233478, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, label=表3, caption=

C2fFaster模块引入不同位置的方案结果对比

, figureFileSmall=null, figureFileBig=null, tableContent=
方案 名称 mAP50/% mAP50-95/% 帧率 参数量/106 GFLOPs 权重/MB
无改进 YOLOv8n 83.6 63.9 303 3.01 8.2 6.3
方案一 替换检测头 80.0 61.3 323 2.67 7.5 5.6
方案二 替换骨干 78.8 57.1 286 2.69 7.4 5.3
方案三 替换全部 75.0 52.5 417 2.34 6.5 4.75
), ArticleFig(id=1172929909992342343, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, label=Table 4, caption=

Comparison of results with different feature fusion pyramids

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 mAP50/% mAP50-95/% 参数量/106 GFLOPs 帧率 权重/MB
SPPF 83.6 63.9 3.01 8.2 303 6.3
SPP 83.8 65.8 4.42 8.3 294 8.7
ASPP 86.3 66.1 5.07 9.7 37 10.4
SPPCSPC 84.3 68.5 4.61 9.4 256 9.5
SPPFCSPC 89.1 70.5 4.62 9.5 233 9.2
), ArticleFig(id=1172929910059451208, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, label=表4, caption=

不同特征融合金字塔结果对比

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 mAP50/% mAP50-95/% 参数量/106 GFLOPs 帧率 权重/MB
SPPF 83.6 63.9 3.01 8.2 303 6.3
SPP 83.8 65.8 4.42 8.3 294 8.7
ASPP 86.3 66.1 5.07 9.7 37 10.4
SPPCSPC 84.3 68.5 4.61 9.4 256 9.5
SPPFCSPC 89.1 70.5 4.62 9.5 233 9.2
), ArticleFig(id=1172929910113977161, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, label=Table 5, caption=

Ablation Study

, figureFileSmall=null, figureFileBig=null, tableContent=
GAM C2f_Faster SPPFCSPC mAP50/% mAP50-95/% 参数量/106 GFLOPs 帧率 权重/MB
× × × 83.6 63.9 3.01 8.2 303 6.3
P × × 88.1 70.5 3.45 8.5 244 7.1
× P × 75.0 52.5 2.34 6.5 417 4.8
× × P 89.1 70.5 4.61 9.4 233 9.2
P P × 84.0 61.3 2.82 7.1 385 5.9
P × P 92.1 76.6 5.05 9.8 227 10.4
× P P 86.5 67.3 3.96 7.9 370 7.8
P P P 90.3 74.7 4.42 8.3 345 9.1
), ArticleFig(id=1172929910172697418, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, label=表5, caption=

消融实验

, figureFileSmall=null, figureFileBig=null, tableContent=
GAM C2f_Faster SPPFCSPC mAP50/% mAP50-95/% 参数量/106 GFLOPs 帧率 权重/MB
× × × 83.6 63.9 3.01 8.2 303 6.3
P × × 88.1 70.5 3.45 8.5 244 7.1
× P × 75.0 52.5 2.34 6.5 417 4.8
× × P 89.1 70.5 4.61 9.4 233 9.2
P P × 84.0 61.3 2.82 7.1 385 5.9
P × P 92.1 76.6 5.05 9.8 227 10.4
× P P 86.5 67.3 3.96 7.9 370 7.8
P P P 90.3 74.7 4.42 8.3 345 9.1
), ArticleFig(id=1172929910248194891, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=EN, label=Table 6, caption=

Comparison of YOLO Series Algorithms

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 mAP50/% mAP50-90/% 参数量/106 GFLOPs 帧率 权重/MB
YOLOv3-tiny 50.6 21.1 8.68 12.9 208 17.5
YOLOv5n 75.4 48.5 1.77 4.1 222 3.9
YOLOv5l 87.8 78.2 46.13 107.7 143 92.9
YOLOv8n 83.6 63.9 3.01 8.2 303 6.3
YOLOv8s 86.3 70.7 11.13 28.4 256 22.6
YOLOv8l 85.7 70.2 43.61 164.8 222 87.7
YOLOv9-c 90.9 80.4 50.71 236.7 83 98
YOLOv10n 80.6 62.2 2.7 8.2 435 5.8
FCG-YOLO(ours) 90.3 74.7 4.42 8.3 345 9.1
), ArticleFig(id=1172929910374024012, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149769463668589536, language=CN, label=表6, caption=

YOLO系列算法比较结果

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 mAP50/% mAP50-90/% 参数量/106 GFLOPs 帧率 权重/MB
YOLOv3-tiny 50.6 21.1 8.68 12.9 208 17.5
YOLOv5n 75.4 48.5 1.77 4.1 222 3.9
YOLOv5l 87.8 78.2 46.13 107.7 143 92.9
YOLOv8n 83.6 63.9 3.01 8.2 303 6.3
YOLOv8s 86.3 70.7 11.13 28.4 256 22.6
YOLOv8l 85.7 70.2 43.61 164.8 222 87.7
YOLOv9-c 90.9 80.4 50.71 236.7 83 98
YOLOv10n 80.6 62.2 2.7 8.2 435 5.8
FCG-YOLO(ours) 90.3 74.7 4.42 8.3 345 9.1
)], attaches=null, journal=Journal(id=1146119176004939786, delFlag=0, nameCn=科学技术与工程, nameEn=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, issn=1671-1815, eissn=, cn=11-4688/T, coden=null, periodic=4, language=CN, oaType=是, 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=UKU/O7GSka5polgCTkbIIw==, journalPrice=null, startedYear=null, abbrevIsoEn=Sci Technol Eng, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754445529766, createdBy=null, updatedBy=13701087609, firstLetterCn=S, firstLetterEn=S, subjectCode=Natural Sciences, subjectName=自然科学, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=UKU/O7GSka5polgCTkbIIw==, picEn=5hwlULoNwcbj3xUmVi9MAQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1159791870395564357, language=CN, name=科学技术与工程, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529793, updatedTime=1754445529793, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.stae.com.cn/jsygc/site/menus/20090429150146001, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1159791870441701702, language=EN, name=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529804, updatedTime=1754445529804, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1146123166801305609, websiteList=[Website(id=1148243202391400884, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, 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/kxjsygc/CN, language=CN, createTime=1751692112777, createBy=18614031015, updateTime=1753520965431, updateBy=18614031015, name=科学技术与工程-中文站点, tplId=1146099689490845704, title=科学技术与工程, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148622798802673703, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=articleTextType, value=kx, createTime=1751782615614, updateTime=1751782615614, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798781702180, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=banner, value=null, createTime=1751782615609, updateTime=1751782615609, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798769119267, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1751782615606, updateTime=1751782615606, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798794285094, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751782615612, updateTime=1751782615612, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798790090789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751782615611, updateTime=1751782615611, creator=18614031015, updator=18614031015)]), Website(id=1155914124811976731, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, 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/kxjsygc/EN, language=EN, createTime=1753521003206, createBy=18614031015, updateTime=1753521003206, updateBy=18614031015, name=科学技术与工程-英文站点, tplId=1146101810881728533, title=Science Technology and Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155914371227308235, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=articleTextType, value=kx, createTime=1753521061952, updateTime=1753521061952, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371210531016, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=banner, value=null, createTime=1753521061947, updateTime=1753521061947, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371202142407, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1753521061945, updateTime=1753521061945, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371223113930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753521061950, updateTime=1753521061950, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371218919625, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753521061949, updateTime=1753521061949, creator=18614031015, updator=18614031015)])], journalTitle=科学技术与工程, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=Science Technology and Engineering, journalPhotoCn=UKU/O7GSka5polgCTkbIIw==, journalPhotoEn=5hwlULoNwcbj3xUmVi9MAQ==, journalFirstLetter=S, 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=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2404425, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2404425, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2404425, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2404425, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于改进YOLOv8n的道路裂缝检测算法
收藏切换
PDF下载
吐尔逊·买买提 1 , 邱建卓 1 , 刘健 2 , 杜含辰 3 , 朱兴林 1 , 徐粒 1
科学技术与工程 | 论文·交通运输 2025,25(14): 6044-6053
收起
收藏切换
科学技术与工程 | 论文·交通运输 2025, 25(14): 6044-6053
基于改进YOLOv8n的道路裂缝检测算法
全屏
吐尔逊·买买提1 , 邱建卓1, 刘健2, 杜含辰3, 朱兴林1, 徐粒1
作者信息
  • 1. 新疆农业大学交通与物流工程学院, 乌鲁木齐 830052
  • 2. 干旱荒漠区公路工程技术交通运输行业重点实验室, 乌鲁木齐 830099
  • 3. 新疆公路桥梁试验检测中心有限责任公司, 乌鲁木齐 830099
  • 吐尔逊·买买提(1975—),男,维吾尔族,新疆阿克苏人,副教授,博士。研究方向:交通工程,数据挖掘。E-mail:

A Road Crack Detection Algorithm Based on Improved YOLOv8n
Mamat TURSUN1 , Jian-zhuo QIU1, Jian LIU2, Han-chen DU3, Xing-lin ZHU1, Li XU1
Affiliations
  • 1. College of Transportation & Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China
  • 2. Key Laboratory of Highway Engineering Technology in Arid Desert Areas of Transport Industry, Urumqi 830099, China
  • 3. Xinjiang Highway Bridge Testing Center Co., Ltd., Urumqi 830099, China
出版时间: 2025-05-18 doi: 10.12404/j.issn.1671-1815.2404425
文章导航
收藏切换

针对当前道路裂缝检测算法检测速度慢、且对于复杂道路条件下的路面裂缝识别精度低、误检漏检率高等问题,设计了一种基于YOLOv8的道路裂缝检测改进算法FCG-YOLO。首先,提出了改进的C2f_Faster结构以降低模型参数量和减少计算复杂度。其次,为了提高检测准确率,在骨干部分引入全局注意力机制GAM,增强算法对小目标裂缝的识别能力,同时改进特征融合金字塔SPPF为SPPFCSPC模块,用于增强算法的特征融合能力。最后,通过消融实验验证了各模块对算法性能的影响,找到了兼具轻量性与准确性的模型配置,并结合实际应用场景,探究了算法的鲁棒性与泛化性。FCG-YOLO在检测效率上表现出色,验证集上的检测精度mAP50与mAP50-95分别达到了90.3%、74.4%,且每秒检测速度提升13.8%,达到345帧,具备较高的检测效率与良好的实际应用价值。

路面裂缝  /  YOLOv8n  /  目标检测  /  深度学习  /  轻量化

Addressing challenges such as large memory footprint, high computational complexity, and insufficient real-time detection speed in road crack detection models for complex scenarios, a highly efficient and precise algorithm named FCG-YOLO was proposed. Lightweight modules and attention mechanisms were integrated, and traditional feature fusion pyramids were enhanced.The algorithm incorporates PConv into the residual calculation module of YOLOv8n to introduce the improved C2f_Faster structure, thereby reducing model parameters and computational complexity. To enhance detection accuracy, GAM(global attention mechanism) was introduced into the backbone, and the Feature Fusion Pyramid SPPF was improved to SPPFCSPC module, enhancing the model’s ability to represent and fuse features of road cracks.The impact of each module on algorithm performance was verified through ablation experiments, identifying a lightweight and accurate model configuration. Furthermore, the robustness and generalization of the algorithm were explored in practical application scenarios.FCG-YOLO demonstrates outstanding detection efficiency, achieving a detection accuracy of 90.3% mAP50 and 74.4% mAP50-95 on the validation set, with a detection speed of 345 frames per second. These results highlight its high detection efficiency and significant practical value.

road cracks  /  YOLOv8n  /  object detection  /  deep learning  /  lightweight models
吐尔逊·买买提, 邱建卓, 刘健, 杜含辰, 朱兴林, 徐粒. 基于改进YOLOv8n的道路裂缝检测算法. 科学技术与工程, 2025 , 25 (14) : 6044 -6053 . DOI: 10.12404/j.issn.1671-1815.2404425
Mamat TURSUN, Jian-zhuo QIU, Jian LIU, Han-chen DU, Xing-lin ZHU, Li XU. A Road Crack Detection Algorithm Based on Improved YOLOv8n[J]. Science Technology and Engineering, 2025 , 25 (14) : 6044 -6053 . DOI: 10.12404/j.issn.1671-1815.2404425
随着城市化的快速推进,道路面临更高的使用频率和负荷,进而加剧了路面的磨损、裂缝及其他损伤,此外,气候变化和自然灾害等外部因素对道路状况也产生了重要影响。及时检测和修复道路裂缝不仅有助于防止更大规模的损坏,降低维护成本,还能延长道路的使用寿命,确保行车安全。截至2022年年底,全国公路养护里程已达535.03×104 km。公路养护的重心已从“建设”转向了“管理与养护”。然而,传统的人工巡检和目视观察方法存在效率低、主观性强及人力资源受限等问题,在此背景下,迫切需要开发自动化的道路裂缝检测方法,以提升道路养护的效率[1]
深度学习具备强大的特征学习能力,经过训练后能够识别复杂多样的图像。研究者们基于此开创了如Faster R-CNN(region-based convolutional neural network)、DETR(DEtection transformer)、YOLO(you only look once)、SSD(single shot multiBox detector)等基于深度学习的目标检测算法,在图像识别领域展现出巨大的应用前景[2]。YOLO系列具备较快的推理速度和高效的目标检测能力,成为目标检测领域的研究热点之一。文献[3-4]将YOLOv3和YOLOv5应用于无纺布表面缺陷与电路板缺陷检测,展现出了YOLO对于各种检测任务良好的适配性,但在检测速度与精度方面存在明显不足,难以满足实际应用要求。为了提升检测速度,文献[5-7]采用了空间卷积、跨阶段网络等方法对YOLO进行了改进,显著减少了运算复杂度,提高了检测速度,但在处理复杂图像时,易丢失部分细节特征,导致检测精度降低。为了提高检测精度,部分学者通过引入注意力的方式,从而增强卷积神经网络的信息提取能力与融合能力。传统注意力如CA(channel attention)、ECA(efficient channel attention)、SE(squeeze-and-excitation)等通过变形、分组、放缩等提高特征图的通道效率的方式增强网络的特征提取能力,解决了目标检测中特征采样不足的问题,获得了检测精度的提升,但导致了丢失跨维信息,对小目标经常出现漏检[8-11]。文献[12]引入Transformer的混合自注意机制,通过聚焦于全局性,对特征信息建立了长距离连接,获得了精度提升,但提升了计算复杂度,检测速度大幅度减少。
因此,现以YOLOv8n为蓝本,结合注意力、特征金字塔改进、空间卷积的思想,现提出一种改进的道路裂缝检测算法FCG-YOLO,旨在提高沥青路面裂缝的智能化检测性能和效率。首先将FasterNet中提出的PConv引入C2f中,全新构建C2f_Fasrer模块,获得更高的网络计算资源利用效率,以满足算法实时性要求。其次,舍弃原本的SPPF特征金字塔,通过加入跨阶段链接,改进为更高效的SPPFCSPC,以扩张模型的感受野,增强算法对更深层次裂缝特征的关注程度。最后,在主干网络中引入全局注意力(global attention mechanism,GAM),对裂缝特征建立作用于全局的长链接,用于提高模型对路面裂缝的特征提取、表达能力。
道路裂缝检测中需要兼顾检测的精度和速度。YOLOv8作为一种较新的目标检测算法,其轻量化变体YOLOv8n在精度和速度上表现优异。基于此,本文研究提出了一种特征提取能力更强的检测算法FCG-YOLO。整体框架如图1所示。
FCG-YOLO从集成注意力模块、改进特征融合网络、改进特征金字塔网络三个角度进行优化,在主干特征融合部分,考虑到前期提取特征的复杂性和数据集背景噪声多的特点,设计了具有更大感受野和表征能力的SPPFCSPC模块替换SPPF(spatial pyramid pooling)空间金字塔池化。另外,由于浅层信息包含的小目标信息更多,在空间金字塔输入部位添加GAM注意力机制,用于加强小目标信息的提取,并更加准确地提取从主干网络中得到的有效特征,从而实现对裂缝的准确检测。并且,为了提高YOLOv8n模型的性能和效率。本文研究优化了模型的特征融合网络,在这个优化过程中,引入FasterNet的空间卷积思路,对原始的C2f模块进行重构,引入了C2f_Faster模块作为替代,使得模型的计算复杂度和计算量大大减少。
高效的空间结构被认为是检测模型轻量化的有效方式,通过空间网络上对信息实现更高效的梯度流动,能够有效地减少目标检测模型的参数及运算量。FasterNet网络是一种轻量化方法[14],旨在通过简单、快速、有效的运算来消除计算差异,该网络提出了一种名为PConv的空间卷积方法,能够在降低浮点运算次数(floating point of operations,FLOPs)的同时保持高每秒浮点运算次数(floating-point operations per second,PLOPS),使得网络在保持高计算速度的同时有效地提取输入图像的特征。
图2中可知,PConv将输入沿着通道维度分割成进行部分卷积的部分(Cp,一般1/4)和保持不变的部分(x2),通过创建一个大小为k的卷积核,对Cp部分进行卷积操作,然后将结果与另一部分进行拼接,得到最终的输出特征,这种操作有效地减少了计算冗余,提高计算效率。PConv的FLOPs和内存访问量F计算公式为
F L O P s = k w k 2 c p 2
F = k 2 c p + k 2 c p 2
式中:wh为特征图的宽和高;k为卷积核的大小;cp为常规卷积作用的通道数;F为内存访问量。由于PConv只对其中的1/4进行普通卷积(cp=1/4),所以FLOPs仅仅为常规卷积的1/16。而F仅仅为常规卷积的1/4,其余的通道不参与计算。
在YOLOv8算法中,C2f模块作为残差计算模块,包含了大量的卷积操作,需要在每个特征输入通道上执行,因此会产生大量的网络计算参数量与内存占用。改进C2f模块是减少YOLO的计算量的一种有效方法。
Bottleneck模块中承担C2f模块中主要的计算任务,用于负责处理不同尺度的特征图信息,使用PConv改进Bottleneck部分的普通卷积,将具有更好的效率。如图3(a)所示,本文设计的全新模块Faster_Block结构简单,仅由一个PConv和一个多层感知机(multilayer perceptron,MLP)组成,前者用于减少模型的计算量, 后者通过增加模型的感知范围,从而提高算法对不同尺度特征信息的灵敏度。C2f_Faster如图3(b)所示,通过引入Faster_Block,在神经网络中建立了更为高效和灵活的空间结构,使得模型计算量大幅减少,信息梯度流动更加高效,算法具有更高的计算效率和更快的速度。
在传统卷积神经网络中,特定结构的网络需要输入固定大小的图像,导致在进行裁剪、翻转、缩放等其他操作时降低了检测精度。在路面裂缝识别检测任务中,不同类型和尺度的路面裂缝使得网络在感知能力和语义理解方面需要具备更高水平,以确保对目标进行准确识别和分类。空间金字塔池化(spatial pyramid pooling,SPP)是一种用于解决不同尺寸输入图像问题的方法,传统的卷积神经网络只能处理固定尺寸的输入图像[15],而 SPP 可以通过池化操作将不同尺寸的图像转换为固定尺寸的特征向量,从而保留各个尺度的特征信息。跨阶段连接(cross stage partial convolution,CSPC)用于提高特征传递和网络效率,通过在网络中引入跨阶段的部分连接,可以加快特征传递的速度,并且更好地利用了底层和高层特征之间的信息。YOLOv5将SPP不同窗口大小并联的池化操作改为相同窗口大小串联的池化操作,减少了模型的计算量,提高了模型速度,提出了SPPF,而YOLOv7在SPP基础上引入了CSPC操作,提出了SPPCSPC,通过在网络中引入跨阶段的部分连接,获得了更高的效率。本文结合YOLOv5的优化思想,如图3所示,将SPPCSPC不同大小的并联池化改为相同大小的串联池化的SPPFCSPC,其结构如图4所示。
图4所示,改进后的特征金字塔SPPCSPC首先进行一个CSP操作,将特征通道的数量减半,分别经过两个卷积操作得到特征图y1y2。接下来,特征图y1进行三次卷积操作得到处理后的特征图x1x1经过串联的三个最大池化层m,得到不同深度的下采样特征图x2x3x4,通过Concat将x1x2x3x4在通道维度上拼接,得到混合特征图y3,池化部分的计算公式为
x 2 = M a x P o o l k = 5 p = 2 ( R )
x 3 = M a x P o o l k = 5 p = 2 ( R )
x 5 = M a x P o o l k = 5 p = 2 ( R )
y 3 = x 1 x 2 x 3 x 4
SPPFCSPC通过串联三个同一大小的池化层,对捕获的特征信息执行最大池化操作,增强了网络的感受野,同时使用CSPC操作融合全连接层和语义金字塔卷积,增强了网络对目标的语义理解能力。将SPPFCSPC模块应用到YOLOv8n模型中,能够提高局部特征与全局特征的深度融合,增强感知范围、提高目标检测性能,并适应复杂场景。
实际道路检测场景中,图像背景占据大部分且裂缝多为小目标,在裂缝检测过程中有必要考虑到全局信息。针对此问题,本文研究引入了全局注意力模块GAM以提高模型对路面裂缝的特征表达能力。如图5所示,GAM在通道注意力模块基础上增加了空间子注意力模块,对于输入的特征图F1,首先经过通道注意力子模块(CA)处理包含关注图像中通道权重信息的Mc。之后再经过空间注意力子模块(SA),得到包含了通道与空间特征的加权组合并输出特征F3
通过在骨干部分加入GAM注意力机制模块,降低信息离散性的同时放大全局维度特征的交互,在路面裂缝检测图片的更深尺度上有效建立了全局依赖关系,同时降低了计算成本,提高了模型性能,使其更适用于不同尺度的道路裂缝检测任务。
本文使用的数据集是吐鲁番至小草湖路段的路面裂缝数据集,起点从吐峪沟收费站西侧起,终点至小草湖互通式立交东侧,路线全长 107.7 km。数据集共计4 581张图片,标注了4 950处共计5种类型的道路裂缝,分别为:条修(TiaoXiu)、块裂(KuaiLie)、龟裂(JunLie)、横裂(HengLie)、块修(KuaiXiu)。并按照8∶2的比例选择了3 664张照片作为训练集,917张照片作为验证集。
本实验采用Microsoft Windows 11专业版操作系统,搭配Intel(R) Core(TM) i7-10700CPU @2.90 GHz处理器和48 GB显存,同时配备NVIDIA GeForce RTX 4090 GPU,选用Python3.10作为编程语言,CUDA版本为12.1,训练基于PyTorch 2.1.2深度学习框架进行。训练参数设置如表1所示。
实验选择经过300轮训练的最佳模型,以确保模型在训练期间能够充分学习到数据的关键特征。综合选取准确率(Precision,P)、召回率(recall,R)、平均精度均值(mean average precision,mAP)、模型参数量(Parameters)、浮点运算量每秒(GFLOPs)、权重文件大小(Size)以及帧率(FPS)作为模型的评价指标,能够全面评估算法在检测精度和计算复杂度方面的表现。其中Parameters、GFLOPs、FPS可在YOLOv8推理过程中得到具体参数。PR和mAP的具体计算公式为
P = T P T P + F P
R = T P T P + F N
m A P = 1 n i = 1 n 0 1 p ( r ) d r
式中:TP为将正样本正确预测为正样本的数量;FP为将负样本错误预测为正样本的数量;AP为对P-R曲线的积分;mAP为各类预测对象精度的均值;mAP50为IoU阈值为0.5时的mAP;mAP50-95为IoU阈值为50~95时的mAP。
混淆矩阵,也称为误差矩阵,反映了分类器预测结果与实际类别之间的对应关系。其横轴代表预测类别,纵轴代表真实类别,对角线上的数值表示正确分类的样本数量,颜色越深表示样本数量越多。通过混淆矩阵的可视化,可以直观地统计每个类别的正确识别数,并了解错误分类的分布情况。在相同数据集中YOLOv8n和FCG-YOLO生成的混淆矩阵(confusion matrix)如图6所示。纵列为模型预测值,横列为真实值,行列的JunLie、KuaiLie等分别表示实际与预测的病害类别,矩阵中的数字为预测正确率,越接近1则表明预测类别的正确度越高。
图6可知,FCG-YOLO(右)的混淆矩阵的对角线区域颜色比YOLOv8深,表明该算法正确预测目标类别的能力得到了增强。改进后算法对5种病害类型的正确检测率分别提高了18.8%、8.1%、0.9%、-1.0%、11.0%,表明相对于原算法,FCG-YOLO具有更有效的特征提取机制,对道路裂缝能够更好地捕捉特征,在预测裂缝类别时更加可靠。
为了验证引入GAM注意力机制的有效性,实验分别在原始YOLOv8n的基础上增加ECA、SA注意力进行对比试验,结果如表2所示,GAM具有最高的检测精度与帧率(frame rate,FPS),相比原始的YOLOv8n,mAP50与mAP50-95分别提高了4.5%、6.6%。
图7所示,采用热力图直观评测注意力对路面裂缝的关注程度,红色表示对裂缝的关注情况。可以看到在实际检测过程中应对光照、背景的干扰问题上,GAM对裂缝的关注更精确,且如第一组所示,面对干扰时能够关注到阴影部分的裂缝,而ECA、SA等注意力机制则无法做到这一点。
综上所述,GAM的引入显著提升了检测精度,并提高了模型在不同层级和维度上的特征提取能力,增强了模型在各种复杂场景下的鲁棒性,尤其是在更严格的mAP50-95指标上表现出更大的提升幅度,证明了GAM在提高模型准确性方面的有效性。
为了验证在不同位置引入C2f_Faster模块与网络性能的关系,实验设置三种方案:方案1、方案2分别在YOLOv8n的检测头部分与骨干网络引入C2f_Faster,方案3则对全部C2f结构进行替换。结果如表3所示。
C2f_Faster在不同部位的引入均有效降低了模型的参数量、浮点运算量以及权重文件大小,同时提升了检测速度,这表明了模块引入对于提升检测速度的有效性。FCG-YOLO的使用的方案3具有最小的运算量与最快的检测速度,相对于原始YOLOv8n算法,参数量、浮点运算量、权重文件体积分别减少22.3%、20.7%、24.6%,而FPS增加114帧,检测速度提升37.6%。但同时应注意到,模块的引入牺牲了模型的检测精度,三种引入方式分别使得mAP50降低3.6%、4.8%、8.6%,在使用的过程中需要综合考虑应用场景,本文设计算法以轻量化为主要方向,因此选取了具备最快速度的方案3。
实验还使用了SPP、ASPP、SPPF、SPPCSPC等其他特征金字塔分别加入YOLOv8n原始模型上做对比,结果如表4所示。
相比其他模型,本文模型的SPPFCSPC具备最高的精度,同时具有较低的参数量,相比原始模型mAP50分别提升了5.5%、5.3%、2.8%,mAP50-95分别提升了6.6%、4.9%、4.4%,表明对于金字塔池化的改进增强了网络对裂缝特征的语义理解与特征融合能力,而结合CSPC操作有效提高了网络的感知范围和特征表示能力。应当注意到,相对于SPPCSPC,由于池化结构由并联改为了串联,加强了信息融合的同时也略微降低了运行速度,每秒检测速度降低了23帧。
为验证本文提出各项改进的有效性,在YOLOv8n网络基础上进行了消融实验,各组实验均设定相同的超参数并采用相同的训练策略。
消融实验如表5所示,相对于原始的YOLOv8n模型,GAM的加入使得精度mAP50提升4.5%,mAP50-95提升6.6%,同时导致参数量增加14.6%,在此基础增加SPPFCSPC后,精度进一步提升4.0%,6.1%,达到最高的92.1%、76.6%,表明GAM与SPPFCSPC均能有效提高算法精度,再加入C2f_Faster后,精度则略微降低,但参数量降低43.5%,运算量减少63.8%,FPS增加118帧,显示出优秀的轻量化特性。FCG-YOLO相对于原模型,mAP50、mAP50-95提高6.7%、10.8%,而每秒检测速度提高了13.8%,达到345帧,在检测精度与轻量化方向上得到了良好的平衡。
与YOLO系列的其他算法进行了比较,用于验证改进算法的优越性,结果如表6所示。从表6可知,FCG-YOLO综合表现出色,相对于YOLOv3、YOLOv5、YOLOv8系列,本文算法具有最高的精度与检测速度。另外应注意到,YOLOv9-c由于提出了更加复杂的模型结构,准确率最高,相对于FCG-YOLO,mAP50与mAP50-95分别提高了0.6%、5.7%,而浮点运算量增加了228.4×109,FPS降低了337帧/s。而YOLOv10由于消除了NMS的训练,降低了推理延迟,相对于本文算法检测速度提高了90 FPS,轻量化表现突出,但精度分别减少了9.7%、12.5%。综合而言,FCG-YOLO在检测速度与轻量化方向上得到了最佳的平衡,具备优越的性能。
另外,为了验证算法实际检测中的鲁棒性与泛化性,选取以上已训练好的检测模型(同系列取较轻量的模型),针对全新数据集中未经训练的正常、曝光不足、光照干扰等不同复杂场景下的图片进行预测推理,结果如图7~图9所示。
在对于正常场景下道路图片检测中(图8),YOLOv5、YOLOv8、YOLOv10n对条修出现漏检,YOLOv5n则出现误检,将条修与横裂识别为一个整体。在曝光不足场景下图片的检测中(图9),YOLOv5n、YOLOv8n、YOLOv9-c、YOLOv10n均对左侧小的横裂出现漏检,YOLOv3-tiny则识别区域过大。在检测光照干扰场景下的图片时(图10),YOLOv3-tiny、YOLOv8n、YOLOv10n出现漏检,而YOLOv5n、YOLOv9-c对右侧阴影部分出现误检。综上所述,在实际应用场景中面对小目标裂缝检测、背景与光照干扰等复杂情况,FCG-YOLO均准确检测到了各类裂缝,表现出良好的鲁棒性与可应用性。
基于YOLOv8n算法,提出了一种轻量化道路检测算法FCG-YOLO,通过引入空间卷积、改进特征融合金字塔以及增加注意力机制,有效提升了模型的检测精度和计算效率。探讨了不同模块对检测性能的影响,并通过与其他算法的比较,验证了所提出算法的优越性,最后探究了在复杂环境下模型的鲁棒性。得出如下结论。
(1)改进后的路面裂缝检测模型在验证集上的平均精度达到90.3%,与改进前相比,mAP50和mAP50-95分别提升了6.7%和10.8%。反映了GAM注意力机制和SPPFCSPC特征融合金字塔在全局信息提取与特征融合方向上具备显著优势。
(2)FCG-YOLO的检测速度达到345帧/s,较原模型提升13.8%。结合检测精度分析,表明所采用的C2f_Faster空间卷积结构有效促进了信息梯度的高效流动,能够显著提升检测算法的时间和空间效率。
(3)在复杂道路检测环境下的测试表明,FCG-YOLO相比其他YOLO系列算法,在小目标裂缝检测、背景和阴影干扰等复杂场景中表现出良好的鲁棒性和泛化能力,满足实时应用的需求。
总体而言,FCG-YOLO在道路裂缝检测精度和时间效率方面表现出色,表现出良好的可推广性与经济效益,具备广泛的工程应用前景,同时为后续YOLOv9、YOLOv10等版本的改进提供了新的思路和参考。
  • 国家自然科学基金(51768071)
  • 新疆交通投资(集团)有限责任公司资助项目(JCZXXJAU2023001)
参考文献 引证文献
排序方式:
[1]
任传亭, 任广海, 李森哲. 公路沥青路面检测技术发展综述[J]. 公路交通技术, 2021, 37(5): 29-34.
Ren Chuanting, Ren Guanghai, Li Senzhe. Review of development of asphalt pavement detection technology for highways[J]. Highway Traffic Technology, 2021, 37(5): 29-34.
[2]
刘宇飞, 冯楚乔, 陈伟乐, 等. 基于机器视觉法的桥梁表观病害检测研究综述[J]. 中国公路学报, 2024, 37(2): 1-15.
Liu Yufei, Feng Chuqiao, Chen Weile, et al. Review of research on bridge surface defect detection based on machine vision[J]. China Journal of Highway and Transport, 2024, 37(2): 1-15.
[3]
铁瑛, 朱空军, 朱振伟, 等. 改进YOLOv3的无纺布表面缺陷检测研究[J]. 机械设计与制造, 2024(11): 362-365.
Tie Ying, Zhu Kongjun, Zhu Zhenwei, et al. Research on improved YOLOv3 for nonwoven fabric surface defect detection[J]. Mechanical Design and Manufacturing, 2024(11): 362-365.
[4]
王龙业, 黄鋆, 曾晓莉. 基于YOLO-PCB的印刷电路板裸板缺陷检测[J]. 科学技术与工程, 2024, 24(15): 6338-6345.
Wang Longye, Huang Jun, Zeng Xiaoli. Bare board defect detection of PCB based on YOLO-PCB[J]. Science Technology and Engineering, 2024, 24(15): 6338-6345.
[5]
张在岩, 宋伟东, 邬嘉晨. 改进YOLOv5的复杂场景下水泥路面病害检测[J]. 辽宁工程技术大学学报(自然科学版), 2025, 44(1): 102-113.
Zhang Zaiyan, Song Weidong, Wu Jiachen. Improved YOLOv5 for cement pavement disease detection in complex scenarios[J]. Journal of Liaoning University of Engineering and Technology (Natural Science Edition), 2025, 44(1): 102-113.
[6]
蒲玲玲, 杨柳. 改进YOLOv5的多车辆目标实时检测及跟踪算法[J]. 科学技术与工程, 2023, 23(28): 12159-12167.
Pu Lingling, Yang Liu. Improved real-time detection and tracking algorithm for multi vehicle targets in YOLOv5[J]. Science Techno-logy and Engineering, 2023, 23(28): 12159-12167.
[7]
Zhang S, Che S, Liu Z, et al. A real-time and lightweight traffic sign detection method based on ghost-YOLO[J]. Multimedia Tools and Applications, 2023, 82(17): 26063-26087.
[8]
潘桂霞, 赖惠成, 王同官, 等. 一种Yolov5颈部细化的小交通标志检测算法[J]. 现代电子技术, 2023, 46(14): 56-62.
Pan Guixia, Lai Huicheng, Wang Tongguan, et al. A YOLOv5-based algorithm for fine neck refinement of small traffic signs detection[J]. Modern Electronics Technique, 2023, 46(14): 56-62.
[9]
曹建荣, 庄园, 汪明, 等. 基于ECA的YOLOv5水下鱼类目标检测[J]. 计算机系统应用, 2023, 32(6): 204-211.
Cao Jianrong, Zhuang Yuan, Wang Ming, et al. Underwater fish target detection based on YOLOv5 with ECA[J]. Computer Systems and Applications, 2023, 32(6): 204-211.
[10]
周丽芳, 邓广, 李伟生, 等. 一种基于SE-YOLOv3的轻量级遥感目标检测方法: CN202011310176.9[P]. 2024-06-12.
Zhou Lifang, Deng Guang, Li Weisheng, et al. Lightweight remote sensing object detection method based on SE-YOLOv3: CN202011310176.9[P]. 2024-06-12.
[11]
Zhou W, Ma Y, Xiao S, et al. SE-yolov5 inland waterway vessel target recognition based on CGAN image enhancement[C]// 7th International Conference on Transportation Information and Safety (ICTIS). Xi'an: IEEE, 2023: 2022-2033.
[12]
雷源毅, 朱文球, 廖欢. 复杂场景下的改进YOLOv8n安全帽佩戴检测算法[J]. 软件工程, 2023, 26(12): 46-51.
Lei Yuanyi, Zhu Wenqiu, Liao Huan. Improved YOLOv8n algorithm for safety helmet detection in complex scenes[J]. Software Engineering, 2023, 26(12): 46-51.
[13]
李岩超, 史卫亚, 冯灿. 面向无人机航拍小目标检测的轻量级YOLOv8检测算法[J]. 计算机工程与应用, 2024, 60(17): 167-178.
Li Yanchao, Shi Weiya, Feng Can. Lightweight YOLOv8 detection algorithm for small target detection in UAV aerial imagery[J]. Computer Engineering and Applications, 2024, 60(17): 167-178.
[14]
Chen J kao S, He H, et al. Run, don’t walk: chasing higher FLOPS for faster neural networks[C]// 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Vancouver: IEEE, 2023: 12021-12031.
[15]
He K, Zhang X, Ren S, et al. Spatial pyramidpooling in deep convolutional networks for visual recognition[J]. IEEE transactions on Pattern Analysis and Machine Intelligence, 2015, 37(9): 1904-1916.
2025年第25卷第14期
PDF下载
314
115
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2404425
  • 接收时间:2024-06-13
  • 首发时间:2025-07-09
  • 出版时间:2025-05-18
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-06-13
  • 修回日期:2025-02-24
基金
国家自然科学基金(51768071)
新疆交通投资(集团)有限责任公司资助项目(JCZXXJAU2023001)
作者信息
    1. 新疆农业大学交通与物流工程学院, 乌鲁木齐 830052
    2. 干旱荒漠区公路工程技术交通运输行业重点实验室, 乌鲁木齐 830099
    3. 新疆公路桥梁试验检测中心有限责任公司, 乌鲁木齐 830099
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2404425
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数

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

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