Article(id=1190284125123744229, tenantId=1146029695717560320, journalId=1189918454225211397, issueId=1190290597995970659, articleNumber=null, orderNo=null, doi=10.20104/j.cnki.1674-6546.20240263, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=null, receivedDateStr=null, revisedDate=1725724800000, revisedDateStr=2024-09-08, acceptedDate=null, acceptedDateStr=null, onlineDate=1761715449835, onlineDateStr=2025-10-29, pubDate=1755187200000, pubDateStr=2025-08-15, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1761716993436, onlineIssueDateStr=2025-10-29, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=1761715449835, onlineFirstDateStr=2025-10-29, sourceXml=null, magXml=null, createTime=1761715449835, creator=13701087609, updateTime=1761715449835, updator=13701087609, issue=Issue{id=1190290597995970659, tenantId=1146029695717560320, journalId=1189918454225211397, year='2025', volume='', issue='8', pageStart='1', pageEnd='48', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=1, specialIssue=null, createTime=1761716993092, creator=13701087609, updateTime=1761717119370, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1190291127711404379, tenantId=1146029695717560320, journalId=1189918454225211397, issueId=1190290597995970659, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1190291127711404380, tenantId=1146029695717560320, journalId=1189918454225211397, issueId=1190290597995970659, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=22, endPage=28, ext={EN=ArticleExt(id=1190284125295710694, articleId=1190284125123744229, tenantId=1146029695717560320, journalId=1189918454225211397, language=EN, title=Research on Traffic Sign Recognition Based on the Improved YOLOv5s-CBAM-ASFF Algorithm, columnId=null, journalTitle=Automotive Engineer, columnName=null, runingTitle=null, highlight=null, articleAbstract=

To achieve more efficient detection of small traffic sign targets under complex urban street background conditions, this paper proposes an improved YOLOv5s algorithm. This enhancement is achieved by incorporating a Convolution Block Attention Module (CBAM) Spatial Channel Attention Mechanism, an Adaptive Spatial Feature Fusion (ASFF) module, and an improved loss function for detection boxes. The validation results on the TT100K traffic sign dataset demonstrate that the proposed algorithm achieves a mean Average Precision (mAP) of 84.5% in traffic sign recognition.

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, authorCompany=null, fund=null, authors=null, authorsList=Rongping Fu, Jiansheng Fu, Wangyang Liang), CN=ArticleExt(id=1190284373103580014, articleId=1190284125123744229, tenantId=1146029695717560320, journalId=1189918454225211397, language=CN, title=基于改进YOLOv5s-CBAM-ASFF算法的交通标志识别研究, columnId=1190284124830142945, journalTitle=汽车工程师, columnName=智能车辆环境感知与智能决策技术专刊, runingTitle=null, highlight=null, articleAbstract=

为在复杂的城市街道背景条件下实现更为高效的交通标志小目标识别,提出了一种改进的YOLOv5s算法,通过引入卷积块注意模块(CBAM)空间通道注意力机制、自适应空间特征融合(ASFF)模块和改进检测框的损失函数进一步提升网络性能,TT100K交通标志数据集上的验证结果表明,所提出的改进算法的交通标志识别平均精度均值(mAP)达到84.5%。

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=IRaQqJ14lUx2OG2O65Xl/A==, magXml=RsFTywWyhQ/GNj8o+KT/fQ==, pdfUrl=null, pdf=yPA72HPKHpsogE5eoEXTCA==, pdfFileSize=4310126, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=53noo6XYbLoyhbOUybXYGQ==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=0+b42/50p1ZLh89vXJeTdA==, mapNumber=null, authorCompany=null, fund=null, authors=null, authorsList=付蓉萍, 付建胜, 梁旺阳)}, authors=[Author(id=1194610998289076321, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, orderNo=0, 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=1194610998368768099, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, authorId=1194610998289076321, language=EN, stringName=Rongping Fu, firstName=Rongping, middleName=null, lastName=Fu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Guilin University of Electronic Technology, Guilin 541000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1194610998427488356, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, authorId=1194610998289076321, language=CN, stringName=付蓉萍, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=桂林电子科技大学, 桂林 541000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1194610998205190237, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, xref=null, ext=[AuthorCompanyExt(id=1194610998213578846, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, companyId=1194610998205190237, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Guilin University of Electronic Technology, Guilin 541000), AuthorCompanyExt(id=1194610998221967455, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, companyId=1194610998205190237, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=桂林电子科技大学, 桂林 541000)])]), Author(id=1194610998498791526, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, 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=1194610998565900392, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, authorId=1194610998498791526, language=EN, stringName=Jiansheng Fu, firstName=Jiansheng, middleName=null, lastName=Fu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Guilin University of Electronic Technology, Guilin 541000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1194610998628814953, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, authorId=1194610998498791526, language=CN, stringName=付建胜, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=桂林电子科技大学, 桂林 541000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1194610998205190237, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, xref=null, ext=[AuthorCompanyExt(id=1194610998213578846, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, companyId=1194610998205190237, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Guilin University of Electronic Technology, Guilin 541000), AuthorCompanyExt(id=1194610998221967455, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, companyId=1194610998205190237, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=桂林电子科技大学, 桂林 541000)])]), Author(id=1194610998695923819, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, 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=1194610998763032685, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, authorId=1194610998695923819, language=EN, stringName=Wangyang Liang, firstName=Wangyang, middleName=null, lastName=Liang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Guilin University of Electronic Technology, Guilin 541000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1194610998821752942, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, authorId=1194610998695923819, language=CN, stringName=梁旺阳, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=桂林电子科技大学, 桂林 541000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1194610998205190237, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, xref=null, ext=[AuthorCompanyExt(id=1194610998213578846, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, companyId=1194610998205190237, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Guilin University of Electronic Technology, Guilin 541000), AuthorCompanyExt(id=1194610998221967455, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, companyId=1194610998205190237, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=桂林电子科技大学, 桂林 541000)])])], keywords=[Keyword(id=1194610998960164975, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=EN, orderNo=1, keyword=Deep learning), Keyword(id=1194610999010496624, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=EN, orderNo=2, keyword=Traffic sign recognition), Keyword(id=1194610999065022577, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=EN, orderNo=3, keyword=Attention mechanism), Keyword(id=1194610999127937138, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=EN, orderNo=4, keyword=Multi-scale feature fusion), Keyword(id=1194610999207628915, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=EN, orderNo=5, keyword=YOLOv5s), Keyword(id=1194610999278932084, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=CN, orderNo=1, keyword=深度学习), Keyword(id=1194610999341846645, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=CN, orderNo=2, keyword=交通标志识别), Keyword(id=1194610999400566902, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=CN, orderNo=3, keyword=注意力机制), Keyword(id=1194610999459287159, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=CN, orderNo=4, keyword=多尺度特征融合), Keyword(id=1194610999522201720, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=CN, orderNo=5, keyword=YOLOv5s)], refs=[Reference(id=1194611002202361999, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=160, pageEnd=166, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=WILLIAM M M, ZAKI P S, SOLIMAN B K, journalName=Proceedings of the 2019 Ninth International Conference on Intelligent Computing and Information Systems (ICICIS), refType=null, unstructuredReference=WILLIAM M M, ZAKI P S, SOLIMAN B K, et al. Traffic Signs Detection and Recognition System Using Deep Learning[C]// Proceedings of the 2019 Ninth International Conference on Intelligent Computing and Information Systems (ICICIS). Cairo, Egypt: IEEE, 2019: 160-166., articleTitle=Traffic Signs Detection and Recognition System Using Deep Learning, refAbstract=null), Reference(id=1194611002273665168, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=1963, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=ROBERTS L, journalName=Massachusetts, refType=null, unstructuredReference=ROBERTS L. Machine Perception of Three-Dimensional Solids[D]. Massachusetts, USA: Massachusetts Institute of Technology, 1963., articleTitle=null, refAbstract=null), Reference(id=1194611002344968337, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2002, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=LIENHART R, MAYDT J, journalName=Proceedings of the 2002 International Conference on Image Processing, refType=null, unstructuredReference=LIENHART R, MAYDT J. An Extended Set of Haar-Like Features for Rapid Object Detection[C]// Proceedings of the 2002 International Conference on Image Processing. Rochester, NY, USA: IEEE, 2002., articleTitle=An Extended Set of Haar-Like Features for Rapid Object Detection, refAbstract=null), Reference(id=1194611002445631634, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2005, volume=null, issue=null, pageStart=886, pageEnd=893, url=null, language=null, rfNumber=[4], rfOrder=3, authorNames=DALAL N, TRIGGS B, journalName=Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, refType=null, unstructuredReference=DALAL N, TRIGGS B. Histograms of Oriented Gradients for Human Detection[C]// Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Diego, California, USA: IEEE, 2005: 886-893., articleTitle=Histograms of Oriented Gradients for Human Detection, refAbstract=null), Reference(id=1194611002504351891, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=null, pageEnd=null, url=https://arxiv.org/abs/1804.02767, language=null, rfNumber=[5], rfOrder=4, authorNames=REDMON J, FARHADI A, journalName=null, refType=null, unstructuredReference=REDMON J, FARHADI A. YOLOv3:An Incremental Improvement[EB/OL]. (2018-04-08) [2022-11-20]. https://arxiv.org/abs/1804.02767, articleTitle=YOLOv3:An Incremental Improvement, refAbstract=null), Reference(id=1194611002579849364, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=21, pageEnd=null, url=null, language=null, rfNumber=[6], rfOrder=5, authorNames=LIU W, ANGUELOV D, ERHAN D, journalName=SSD: Single Shot MultiBox Detector, refType=null, unstructuredReference=LIU W, ANGUELOV D, ERHAN D, et al. SSD: Single Shot MultiBox Detector[M]// LEIBE B, MATAS J, SEBE N, et al. Computer Vision-ECCV 2016. Lecture Notes in Computer Science. Cham: Springer, 2016: 21-37., articleTitle=null, refAbstract=null), Reference(id=1194611002634375317, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2020, volume=56, issue=10, pageStart=20, pageEnd=26, url=null, language=null, rfNumber=[7], rfOrder=6, authorNames=伍晓晖, 田启川, journalName=计算机工程与应用, refType=null, unstructuredReference=伍晓晖, 田启川. 交通标志识别方法综述[J]. 计算机工程与应用, 2020, 56(10): 20-26., articleTitle=交通标志识别方法综述, refAbstract=null), Reference(id=1194611002701484182, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2020, volume=56, issue=10, pageStart=20, pageEnd=26, url=null, language=null, rfNumber=[7], rfOrder=7, authorNames=WU X H, TIAN Q C, journalName=Computer Engineering and Applications, refType=null, unstructuredReference=WU X H, TIAN Q C. Survey of Traffic Sign Recognition[J]. Computer Engineering and Applications, 2020, 56(10): 20-26., articleTitle=Survey of Traffic Sign Recognition, refAbstract=null), Reference(id=1194611002764398743, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2021, volume=42, issue=6, pageStart=1262, pageEnd=1266, url=null, language=null, rfNumber=[8], rfOrder=8, authorNames=廖璐明, 张伟, 王亚刚, journalName=小型微型计算机系统, refType=null, unstructuredReference=廖璐明, 张伟, 王亚刚. 基于注意力机制的交通标志识别[J]. 小型微型计算机系统, 2021, 42(6): 1262-1266., articleTitle=基于注意力机制的交通标志识别, refAbstract=null), Reference(id=1194611002827313304, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2021, volume=42, issue=6, pageStart=1262, pageEnd=1266, url=null, language=null, rfNumber=[8], rfOrder=9, authorNames=LIAO L M, ZHANG W, WANG Y G, journalName=Journal of Miniaturization for Microcomputers, refType=null, unstructuredReference=LIAO L M, ZHANG W, WANG Y G. Traffic Sign Recognition Based on Attention Mechanism[J]. Journal of Miniaturization for Microcomputers, 2021, 42(6): 1262-1266., articleTitle=Traffic Sign Recognition Based on Attention Mechanism, refAbstract=null), Reference(id=1194611002890227865, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2020, volume=40, issue=8, pageStart=2472, pageEnd=2478, url=null, language=null, rfNumber=[9], rfOrder=10, authorNames=江金洪, 鲍胜利, 史文旭, journalName=计算机应用, refType=null, unstructuredReference=江金洪, 鲍胜利, 史文旭, 等. 基于YOLOv3算法改进的交通标志识别算法[J]. 计算机应用, 2020, 40(8): 2472-2478., articleTitle=基于YOLOv3算法改进的交通标志识别算法, refAbstract=null), Reference(id=1194611002957336730, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2020, volume=40, issue=8, pageStart=2472, pageEnd=2478, url=null, language=null, rfNumber=[9], rfOrder=11, authorNames=JIANG J H, BAO S L, SHI W X, journalName=Computer Applications, refType=null, unstructuredReference=JIANG J H, BAO S L, SHI W X, et al. Improvement of Traffic Sign Recognition Algorithm Based on YOLOv3[J]. Computer Applications, 2020, 40(8): 2472-2478., articleTitle=Improvement of Traffic Sign Recognition Algorithm Based on YOLOv3, refAbstract=null), Reference(id=1194611003058000027, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2022, volume=42, issue=2, pageStart=93, pageEnd=102, url=null, language=null, rfNumber=[10], rfOrder=12, authorNames=黄志强, 李军, journalName=南京邮电大学学报(自然科学版), refType=null, unstructuredReference=黄志强, 李军. 基于空间通道注意力机制与多尺度融合的交通标志识别研究[J]. 南京邮电大学学报(自然科学版), 2022, 42(2): 93-102., articleTitle=基于空间通道注意力机制与多尺度融合的交通标志识别研究, refAbstract=null), Reference(id=1194611003116720284, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2022, volume=42, issue=2, pageStart=93, pageEnd=102, url=null, language=null, rfNumber=[10], rfOrder=13, authorNames=HUANG Z Q, LI J, journalName=Journal of Nanjing University of Posts and Telecommunications (Natural Science Edition), refType=null, unstructuredReference=HUANG Z Q, LI J. Research on Traffic Sign Recognition Based on Spatial Channel Attention Mechanism and Multi-Scale Fusion[J]. Journal of Nanjing University of Posts and Telecommunications (Natural Science Edition), 2022, 42(2): 93-102., articleTitle=Research on Traffic Sign Recognition Based on Spatial Channel Attention Mechanism and Multi-Scale Fusion, refAbstract=null), Reference(id=1194611003179634845, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2022, volume=43, issue=7, pageStart=1471, pageEnd=1476, url=null, language=null, rfNumber=[11], rfOrder=14, authorNames=郭继峰, 孙文博, 庞志奇, journalName=小型微型计算机系统, refType=null, unstructuredReference=郭继峰, 孙文博, 庞志奇, 等. 一种改进YOLOv4的交通标志识别算法[J]. 小型微型计算机系统, 2022, 43(7): 1471-1476., articleTitle=一种改进YOLOv4的交通标志识别算法, refAbstract=null), Reference(id=1194611003288686750, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2022, volume=43, issue=7, pageStart=1471, pageEnd=1476, url=null, language=null, rfNumber=[11], rfOrder=15, authorNames=GUO J F, SUN W B, PANG Z Q, journalName=Journal of Miniaturization for Microcomputers, refType=null, unstructuredReference=GUO J F, SUN W B, PANG Z Q, et al. An Improved YOLOv4 Algorithm for Traffic Sign Recognition[J]. Journal of Miniaturization for Microcomputers, 2022, 43(7): 1471-1476., articleTitle=An Improved YOLOv4 Algorithm for Traffic Sign Recognition, refAbstract=null), Reference(id=1194611003376767135, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2023, volume=37, issue=11, pageStart=20, pageEnd=31, url=null, language=null, rfNumber=[12], rfOrder=16, authorNames=查超能, 罗素云, 何佳, journalName=重庆理工大学学报(自然科学), refType=null, unstructuredReference=查超能, 罗素云, 何佳. 雨天下基于注意力机制与特征融合的交通标志识别[J]. 重庆理工大学学报(自然科学), 2023, 37(11): 20-31., articleTitle=雨天下基于注意力机制与特征融合的交通标志识别, refAbstract=null), Reference(id=1194611003456458912, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2023, volume=37, issue=11, pageStart=20, pageEnd=31, url=null, language=null, rfNumber=[12], rfOrder=17, authorNames=ZHA C N, LUO S Y, HE J, journalName=Journal of Chongqing University of Technology (Natural Science), refType=null, unstructuredReference=ZHA C N, LUO S Y, HE J. Traffic Sign Recognition Based on Attention Mechanism and Feature Fusion under Rainy Conditions[J]. Journal of Chongqing University of Technology (Natural Science), 2023, 37(11): 20-31., articleTitle=Traffic Sign Recognition Based on Attention Mechanism and Feature Fusion under Rainy Conditions, refAbstract=null), Reference(id=1194611003540344993, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2022, volume=42, issue=9, pageStart=2876, pageEnd=2884, url=null, language=null, rfNumber=[13], rfOrder=18, authorNames=尹靖涵, 瞿绍军, 姚泽楷, journalName=计算机应用, refType=null, unstructuredReference=尹靖涵, 瞿绍军, 姚泽楷, 等. 基于YOLOv5的雾霾天气下交通标志识别模型[J]. 计算机应用, 2022, 42(9): 2876-2884., articleTitle=基于YOLOv5的雾霾天气下交通标志识别模型, refAbstract=null), Reference(id=1194611003603259554, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2022, volume=42, issue=9, pageStart=2876, pageEnd=2884, url=null, language=null, rfNumber=[13], rfOrder=19, authorNames=YIN J H, QU S J, YAO Z K, journalName=Computer Applications, refType=null, unstructuredReference=YIN J H, QU S J, YAO Z K, et al. A Traffic Sign Recognition Model Based on YOLOv5 in Hazy Weather[J]. Computer Applications, 2022, 42(9): 2876-2884., articleTitle=A Traffic Sign Recognition Model Based on YOLOv5 in Hazy Weather, refAbstract=null), Reference(id=1194611003657785507, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=1446, pageEnd=1451, url=null, language=null, rfNumber=[14], rfOrder=20, authorNames=SHI X H, HU J, LEI X Y, journalName=Proceedings of the 6th International Conference on Intelligent Computing and Signal Processing, refType=null, unstructuredReference=SHI X H, HU J, LEI X Y, et al. Detection of Flying Birds in Airport Monitoring Based on Improved YOLOv5[C]// Proceedings of the 6th International Conference on Intelligent Computing and Signal Processing. Xi'an, China: IEEE, 2021: 1446-1451., articleTitle=Detection of Flying Birds in Airport Monitoring Based on Improved YOLOv5, refAbstract=null), Reference(id=1194611003716505764, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=3, pageEnd=19, url=null, language=null, rfNumber=[15], rfOrder=21, authorNames=WOO S, PARK J, LEE J Y, journalName=Proceedings of the European Conference on Computer Vision (ECCV), refType=null, unstructuredReference=WOO S, PARK J, LEE J Y, et al. CBAM: Convolutional Block Attention Module[C]// Proceedings of the European Conference on Computer Vision (ECCV). Munich, Germany: Springer, 2018: 3-19., articleTitle=CBAM: Convolutional Block Attention Module, refAbstract=null), Reference(id=1194611003779420325, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=null, pageEnd=null, url=https://arxiv.org/abs/1911.09516, language=null, rfNumber=[16], rfOrder=22, authorNames=LIU S T, HUANG D, WANG Y H, journalName=null, refType=null, unstructuredReference=LIU S T, HUANG D, WANG Y H. Learning Spatial Fusion for Single-Shot Object Detection[EB/OL]. (2019-11-25) [2024-09-08]. https://arxiv.org/abs/1911.09516, articleTitle=Learning Spatial Fusion for Single-Shot Object Detection, refAbstract=null), Reference(id=1194611003838140582, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=658, pageEnd=666, url=null, language=null, rfNumber=[17], rfOrder=23, authorNames=REZATOFIGHI H, TSOI N, GWAK J, journalName=Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, refType=null, unstructuredReference=REZATOFIGHI H, TSOI N, GWAK J, et al. Generalized Intersection Over Union: A Metric and a Loss for Bounding Box Regression[C]// Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach, CA, USA: IEEE, 2019: 658-666., articleTitle=Generalized Intersection Over Union: A Metric and a Loss for Bounding Box Regression, refAbstract=null), Reference(id=1194611003909443751, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=516, pageEnd=520, url=null, language=null, rfNumber=[18], rfOrder=24, authorNames=YU J H, JIANG Y N, WANG Z Y, journalName=Proceedings of the 24th ACM International Conference on Multimedia, refType=null, unstructuredReference=YU J H, JIANG Y N, WANG Z Y, et al. UnitBox: An Advanced Object Detection Network[C]// Proceedings of the 24th ACM International Conference on Multimedia. Amsterdam, The Netherlands: ACM, 2016: 516-520., articleTitle=UnitBox: An Advanced Object Detection Network, refAbstract=null), Reference(id=1194611003972358312, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2021, volume=9, issue=null, pageStart=105686, pageEnd=105695, url=null, language=null, rfNumber=[19], rfOrder=25, authorNames=WANG X F, SONG J, journalName=IEEE Access, refType=null, unstructuredReference=WANG X F, SONG J. ICIoU: Improved Loss Based on Complete Intersection Over Union for Bounding Box Regression[J]. IEEE Access, 2021, 9: 105686-105695., articleTitle=ICIoU: Improved Loss Based on Complete Intersection Over Union for Bounding Box Regression, refAbstract=null), Reference(id=1194611004026884265, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[20], rfOrder=26, authorNames=LIU W, ANGUELOV D, ERHAN D, journalName=European Conference on Computer Vision, refType=null, unstructuredReference=LIU W, ANGUELOV D, ERHAN D, et al. SSD: Single Shot MultiBox Detector[C]// European Conference on Computer Vision. Cham, Switzerland: Springer, 2016., articleTitle=SSD: Single Shot MultiBox Detector, refAbstract=null), Reference(id=1194611004089798826, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=null, pageEnd=null, url=https://arxiv.org/abs/1904.07850, language=null, rfNumber=[21], rfOrder=27, authorNames=ZHOU X Y, WANG D Q, KRÄHENBÜHL P, journalName=null, refType=null, unstructuredReference=ZHOU X Y, WANG D Q, KRÄHENBÜHL P. Objects as Points[EB/OL]. (2019-04-25) [2024-09-08]. https://arxiv.org/abs/1904.07850, articleTitle=Objects as Points, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1194610998205190237, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, xref=null, ext=[AuthorCompanyExt(id=1194610998213578846, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, companyId=1194610998205190237, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Guilin University of Electronic Technology, Guilin 541000), AuthorCompanyExt(id=1194610998221967455, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, companyId=1194610998205190237, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=桂林电子科技大学, 桂林 541000)])], figs=[ArticleFig(id=1194610999677390969, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=EN, label=null, caption=null, figureFileSmall=w31Ho+jnlqhRDs6zPHJ+kw==, figureFileBig=Ma0jZeBCYTQ1uEagG5gMwQ==, tableContent=null), ArticleFig(id=1194610999731916922, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=CN, label=图1, caption=YOLOv5网络结构示意, figureFileSmall=w31Ho+jnlqhRDs6zPHJ+kw==, figureFileBig=Ma0jZeBCYTQ1uEagG5gMwQ==, tableContent=null), ArticleFig(id=1194610999811608699, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=EN, label=null, caption=null, figureFileSmall=2tMLdm80p+guDPoDl/H4Gg==, figureFileBig=Lh8CpszCgp9mAXPdhb56xg==, tableContent=null), ArticleFig(id=1194610999878717564, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=CN, label=图2, caption=YOLOv5s-CBAM-ASFF整体框架示意, figureFileSmall=2tMLdm80p+guDPoDl/H4Gg==, figureFileBig=Lh8CpszCgp9mAXPdhb56xg==, tableContent=null), ArticleFig(id=1194610999941632125, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=EN, label=null, caption=null, figureFileSmall=dpIAirv+o2CORIfCRLfgBA==, figureFileBig=wZMp4z3pzpxKtqr1hOwRxQ==, tableContent=null), ArticleFig(id=1194610999996158078, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=CN, label=图3, caption=CBAM空间通道注意力模块示意, figureFileSmall=dpIAirv+o2CORIfCRLfgBA==, figureFileBig=wZMp4z3pzpxKtqr1hOwRxQ==, tableContent=null), ArticleFig(id=1194611000059072639, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=EN, label=null, caption=null, figureFileSmall=WJzhB4rqnMVkrJGarYA7dQ==, figureFileBig=emG1L3WW+pSNBDn06qthEw==, tableContent=null), ArticleFig(id=1194611000117792896, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=CN, label=图4, caption=ASFF结构示意, figureFileSmall=WJzhB4rqnMVkrJGarYA7dQ==, figureFileBig=emG1L3WW+pSNBDn06qthEw==, tableContent=null), ArticleFig(id=1194611000172318849, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=EN, label=null, caption=null, figureFileSmall=HgDZlhIRGBxHrmIUWi5ckA==, figureFileBig=l2e50OhN9NMMz0qYEFT8sw==, tableContent=null), ArticleFig(id=1194611000235233410, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=CN, label=图5, caption=TT100K数据集样本, figureFileSmall=HgDZlhIRGBxHrmIUWi5ckA==, figureFileBig=l2e50OhN9NMMz0qYEFT8sw==, tableContent=null), ArticleFig(id=1194611000302342275, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=EN, label=null, caption=null, figureFileSmall=+OfOWFo6X8EJhzSdLgNXhA==, figureFileBig=52rif1aIIPPBa5bzOapRVw==, tableContent=null), ArticleFig(id=1194611000373645444, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=CN, label=图6, caption=45类交通标志, figureFileSmall=+OfOWFo6X8EJhzSdLgNXhA==, figureFileBig=52rif1aIIPPBa5bzOapRVw==, tableContent=null), ArticleFig(id=1194611000449142917, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=EN, label=null, caption=null, figureFileSmall=HrWdW/d7kwWdZeKD3v2Wpw==, figureFileBig=M6Atqpg61uMFPwY0XxrvcQ==, tableContent=null), ArticleFig(id=1194611000512057478, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=CN, label=图7, caption=交通标志种类数量分布情况, figureFileSmall=HrWdW/d7kwWdZeKD3v2Wpw==, figureFileBig=M6Atqpg61uMFPwY0XxrvcQ==, tableContent=null), ArticleFig(id=1194611000579166343, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=EN, label=null, caption=null, figureFileSmall=z3i0vu0m5dvRFuvSywjO6A==, figureFileBig=sDtmZ2UCgjm6kTD/8zLu8Q==, tableContent=null), ArticleFig(id=1194611000637886600, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=CN, label=图8, caption=改进算法模型验证结果, figureFileSmall=z3i0vu0m5dvRFuvSywjO6A==, figureFileBig=sDtmZ2UCgjm6kTD/8zLu8Q==, tableContent=null), ArticleFig(id=1194611000700801161, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
参数 配置
中央处理器(CPU) 英特尔酷睿i7-11700
图形处理器(GPU) 英伟达GeForce GTX 3090
系统 Linux-Ubuntu20.04
编程语言 Python
随机存取内存(RAM)/GB 24
统一计算设备架构(CUDA)版本 CUDA11.3
PyTorch版本 PyTorch1.11
), ArticleFig(id=1194611000767910026, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=CN, label=表1, caption=

验证环境参数

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 配置
中央处理器(CPU) 英特尔酷睿i7-11700
图形处理器(GPU) 英伟达GeForce GTX 3090
系统 Linux-Ubuntu20.04
编程语言 Python
随机存取内存(RAM)/GB 24
统一计算设备架构(CUDA)版本 CUDA11.3
PyTorch版本 PyTorch1.11
), ArticleFig(id=1194611000839213195, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型 P R mAP@0.5 F1分数
YOLOv5s 85.4 86.1 74.8 69.3
YOLOv5s-CBAM 91.8 95.4 82.1 79.0
YOLOv5s-ASFF 92.0 96.2 83.2 80.1
YOLOv5s-CBAM-ASFF 93.8 97.0 84.5 82.2
), ArticleFig(id=1194611001904566412, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=CN, label=表2, caption=

消融试验结果 %

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型 P R mAP@0.5 F1分数
YOLOv5s 85.4 86.1 74.8 69.3
YOLOv5s-CBAM 91.8 95.4 82.1 79.0
YOLOv5s-ASFF 92.0 96.2 83.2 80.1
YOLOv5s-CBAM-ASFF 93.8 97.0 84.5 82.2
), ArticleFig(id=1194611001975869581, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型 P R mAP@0.5 F1分数
SSD 86.4 70.9 57.8 55.9
CenterNet 89.9 79.1 81.7 66.4
YOLOv3 65.6 67.2 51.1 41.2
YOLOv4 79.7 68.0 54.4 44.0
YOLOv5s 85.4 86.1 74.8 69.3
YOLOv5s-CBAM-ASFF 93.8 97.0 84.5 82.2
), ArticleFig(id=1194611002047172750, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1190284125123744229, language=CN, label=表3, caption=

对比试验结果 %

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型 P R mAP@0.5 F1分数
SSD 86.4 70.9 57.8 55.9
CenterNet 89.9 79.1 81.7 66.4
YOLOv3 65.6 67.2 51.1 41.2
YOLOv4 79.7 68.0 54.4 44.0
YOLOv5s 85.4 86.1 74.8 69.3
YOLOv5s-CBAM-ASFF 93.8 97.0 84.5 82.2
)], attaches=null, journal=Journal(id=1189918244568731652, delFlag=0, nameCn=汽车工程师, nameEn=Automotive Engineer, nameHistory1=null, nameHistory2=null, issn=1674-6546, eissn=null, cn=22-1432/U, coden=null, periodic=0, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=+bJsKkKt/pjz9u6EwhnksQ==, journalPrice=null, startedYear=null, abbrevIsoEn=null, journalRemark=null, publicationField=null, createdTime=1761628217121, updatedTime=1761735708780, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=A, firstLetterEn=A, subjectCode=Engineering, subjectName=Engineering, subjectCodeEn=Engineering, subjectNameEn=null, picCn=+bJsKkKt/pjz9u6EwhnksQ==, picEn=O3Sn3tnYYrh/jm6emnnMWA==, jcr=null, cjcr=null, exts=[JournalExt(id=1190369097415233706, language=CN, name=汽车工程师, 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=1761735708812, updatedTime=1761735708812, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=1, submissionEditorUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=3, submissionReviewUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=2, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1190369097553645739, language=EN, name=Automotive Engineer, 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=1761735708845, updatedTime=1761735708845, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=1, submissionEditorUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=3, submissionReviewUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=2, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1189918454225211397, websiteList=[Website(id=1189918982430847716, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1189918454225211397, 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/qcgcs/CN, language=CN, createTime=1761628393037, createBy=18614031015, updateTime=1761628422913, updateBy=18614031015, name=汽车工程师-中文, tplId=1146099689490845704, title=汽车工程师, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1189919800185917791, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=articleTextType, value=kx, createTime=1761628588005, updateTime=1761628588005, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800164946268, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=banner, value=null, createTime=1761628588000, updateTime=1761628588000, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800211083618, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=grayFlag, value=0, createTime=1761628588011, updateTime=1761628588011, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800156557659, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=logo, value=https://castjournals.cast.org.cn/joweb/qcgcs/CN/file/pic?fileId=yiZ96RYoYcnGnRMuWdmkWA==, createTime=1761628587998, updateTime=1761628587998, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800223666532, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=minRunFlag, value=0, createTime=1761628588014, updateTime=1761628588014, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800181723486, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/qcgcs/CN/file/pic, createTime=1761628588004, updateTime=1761628588004, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800215277923, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=silenceFlag, value=0, createTime=1761628588012, updateTime=1761628588012, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800173334877, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1761628588002, updateTime=1761628588002, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800194306400, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=themeColor, value=null, createTime=1761628588007, updateTime=1761628588007, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800202695009, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=themeStyle, value=null, createTime=1761628588009, updateTime=1761628588009, creator=18614031015, updator=18614031015)]), Website(id=1189918982527316711, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1189918454225211397, 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/qcgcs/EN, language=EN, createTime=1761628393061, createBy=18614031015, updateTime=1761628543075, updateBy=18614031015, name=汽车工程师-英文, tplId=1146101810881728533, title=Automotive Engineer, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1189919837561352952, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=articleTextType, value=kx, createTime=1761628596916, updateTime=1761628596916, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837540381429, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=banner, value=null, createTime=1761628596911, updateTime=1761628596911, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837582324475, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=grayFlag, value=0, createTime=1761628596921, updateTime=1761628596921, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837527798516, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=logo, value=https://castjournals.cast.org.cn/joweb/qcgcs/EN/file/pic?fileId=yiZ96RYoYcnGnRMuWdmkWA==, createTime=1761628596908, updateTime=1761628596908, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837594907389, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=minRunFlag, value=0, createTime=1761628596924, updateTime=1761628596924, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837557158647, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/qcgcs/EN/file/pic, createTime=1761628596915, updateTime=1761628596915, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837586518780, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=silenceFlag, value=0, createTime=1761628596922, updateTime=1761628596922, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837548770038, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1761628596913, updateTime=1761628596913, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837569741561, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=themeColor, value=null, createTime=1761628596918, updateTime=1761628596918, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837573935866, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=themeStyle, value=null, createTime=1761628596919, updateTime=1761628596919, creator=18614031015, updator=18614031015)])], journalTitle=汽车工程师, weixinUrl=null, journalUrl=https://tjqc.cbpt.cnki.net/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Automotive Engineer, journalPhotoCn=+bJsKkKt/pjz9u6EwhnksQ==, journalPhotoEn=O3Sn3tnYYrh/jm6emnnMWA==, journalFirstLetter=A, 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), detailUrlCn=https://castjournals.cast.org.cn/joweb/qcgcs/CN/10.20104/j.cnki.1674-6546.20240263, detailUrlEn=https://castjournals.cast.org.cn/joweb/qcgcs/EN/10.20104/j.cnki.1674-6546.20240263, pdfUrlCn=https://castjournals.cast.org.cn/joweb/qcgcs/CN/PDF/10.20104/j.cnki.1674-6546.20240263, pdfUrlEn=https://castjournals.cast.org.cn/joweb/qcgcs/EN/PDF/10.20104/j.cnki.1674-6546.20240263, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于改进YOLOv5s-CBAM-ASFF算法的交通标志识别研究
收藏切换
PDF下载
付蓉萍 , 付建胜 , 梁旺阳
汽车工程师 | 智能车辆环境感知与智能决策技术专刊 2025,(8): 22-28
收起
收藏切换
汽车工程师 | 智能车辆环境感知与智能决策技术专刊 2025, (8): 22-28
基于改进YOLOv5s-CBAM-ASFF算法的交通标志识别研究
全屏
付蓉萍, 付建胜, 梁旺阳
作者信息
  • 桂林电子科技大学, 桂林 541000
Research on Traffic Sign Recognition Based on the Improved YOLOv5s-CBAM-ASFF Algorithm
Rongping Fu, Jiansheng Fu, Wangyang Liang
Affiliations
  • Guilin University of Electronic Technology, Guilin 541000
出版时间: 2025-08-15 doi: 10.20104/j.cnki.1674-6546.20240263
文章导航
收藏切换

为在复杂的城市街道背景条件下实现更为高效的交通标志小目标识别,提出了一种改进的YOLOv5s算法,通过引入卷积块注意模块(CBAM)空间通道注意力机制、自适应空间特征融合(ASFF)模块和改进检测框的损失函数进一步提升网络性能,TT100K交通标志数据集上的验证结果表明,所提出的改进算法的交通标志识别平均精度均值(mAP)达到84.5%。

深度学习  /  交通标志识别  /  注意力机制  /  多尺度特征融合  /  YOLOv5s

To achieve more efficient detection of small traffic sign targets under complex urban street background conditions, this paper proposes an improved YOLOv5s algorithm. This enhancement is achieved by incorporating a Convolution Block Attention Module (CBAM) Spatial Channel Attention Mechanism, an Adaptive Spatial Feature Fusion (ASFF) module, and an improved loss function for detection boxes. The validation results on the TT100K traffic sign dataset demonstrate that the proposed algorithm achieves a mean Average Precision (mAP) of 84.5% in traffic sign recognition.

Deep learning  /  Traffic sign recognition  /  Attention mechanism  /  Multi-scale feature fusion  /  YOLOv5s
付蓉萍, 付建胜, 梁旺阳. 基于改进YOLOv5s-CBAM-ASFF算法的交通标志识别研究. 汽车工程师, 2025 , (8) : 22 -28 . DOI: 10.20104/j.cnki.1674-6546.20240263
Rongping Fu, Jiansheng Fu, Wangyang Liang. Research on Traffic Sign Recognition Based on the Improved YOLOv5s-CBAM-ASFF Algorithm[J]. Automotive Engineer, 2025 , (8) : 22 -28 . DOI: 10.20104/j.cnki.1674-6546.20240263
在城市街道场景中,交通标志的识别往往存在复杂的背景干扰和标志被部分遮挡等因素[1],严重影响检测速度和准确性。传统的手工特征提取方法,如索贝尔(Sobel)边缘检测特征[2]、哈尔(Haar)特征[3]、方向梯度直方图(Histogram of Oriented Gradient,HOG)特征[4]等,因其泛化能力有限且计算复杂度高,在实时交通标志识别场景中表现不佳。
深度学习技术的兴起与发展使其在目标检测领域的应用日益广泛。单阶段检测算法如YOLO(You Only Look Once)[5]和单步多框检测器(Single Shot MultiBox Detector,SSD)[6]在交通标志检测任务中取得了显著成果。深度学习有助于解决城市街道复杂背景条件下的交通标志识别问题,是自动驾驶和无人驾驶技术性能提升的主要途径[7]
廖璐明等[8]提出一种融合空间变化网络与注意力机制的交通标志识别系统,有效提升了识别性能;江金洪等[9]在YOLOv3基础上引入深度可分离卷积技术,改进损失函数,进而提高检测精度;黄志强等[10]提出轻量化的YOLOv3-tiny模型,通过引入注意力机制提高了识别的准确性;郭继峰等[11]针对YOLOv4算法进行改进,采用深度可分离卷积和焦点(Focal)损失函数,解决模型参数量大和样本不均衡问题,进而提升了检测精度;查超能等[12]在YOLOv5中引入Transformer和自适应空间特征融合(Adaptive Spatial Feature Fusion,ASFF)机制,有效解决了雨天环境下的交通标志识别问题;尹靖涵等[13]通过数据增强、K-均值(K-Means)先验框和全局非极大值抑制(Global Non-Maximum Suppression,GNMS)机制来提升YOLOv5模型的识别精度。
尽管前述算法在提升交通标志识别精度方面取得了一定进展,但在城市街道的复杂环境下实现交通标志的识别仍面临诸多挑战。本文致力于解决城市街道复杂背景下的交通标志小目标识别问题,提出一种基于YOLOv5s的改进算法。通过在颈部(Neck)网络中引入卷积块注意模块(Convolutional Block Attention Module,CBAM)空间通道注意力机制,增强对细节特征的识别精度。在检测头网络中添加ASFF机制模块,实现特征优化组合。此外,采用完全交并比(Complete Intersection over Union,CIoU)损失函数替代广义交并比(Generalized Intersection over Union,GIoU)损失函数以提高模型训练的收敛速度和识别能力。最后,在交通标志数据集TT100K(Tsinghua-Tencent 100K)上进行模型的训练与测试。
YOLOv5采用模块化思想,提供了不同规模的模型来适应不同的应用需求,YOLOv5s是该系列中模型规模最小的版本,具有相对较窄的特征映射通道。随着模型复杂度的逐级提升,其他3个变体在深度和宽度上均有所增加,从而在一定程度上提高了检测精度,但这是以增加计算资源消耗为代价的。为了在确保网络轻量化的同时实现交通标志的快速检测与识别,本文采用YOLOv5s作为基础模型。
YOLOv5网络结构主要分为4个部分:输入层(Input)、主干层(Backbone)、颈部层(Neck)和预测层(Prediction)[14],如图1所示。Input负责接收输入图像并将其标准化,为后续处理提供统一的数据格式;Backbone利用深度卷积神经网络从输入图像中有效提取关键特征;Neck网络采用特征金字塔网络(Feature Pyramid Network,FPN)结构,通过多尺度特征融合策略增强模型对不同尺寸目标的检测能力;Prediction模块负责在特征映射上生成锚框(Anchor Box),并输出用于目标检测的边界框,同时提供检测对象的类别名称和相应的置信度。这种分层和模块化的设计,使YOLOv5能够在保持高检测速度的同时,实现对目标的快速准确识别。
其中,Backbone网络中的Focus为聚焦模块,CBL模块集成了卷积(Conv)模块、批归一化(Batch Normalization,BN)模块和线性修正单元(Rectified Linear Unit,ReLU)激活函数,Conv模块中的Act为激活函数(Activation Function),Res Unit为残差单元(Residual Unit)。进一步地,跨阶段部分(Cross Stage Partial,CSP)网络细分为CSP1和CSP2两个阶段:CSP1常用于模型的Backbone部分,CSP2常用于模型的Neck部分。此外,空间金字塔池化(Spatial Pyramid Pooling,SPP)模块常采用最大池化(Max Pooling)操作。Neck网络中的Concat为拼接模块,Upsample为上采样操作模块。
本文基于注意力机制和多尺度特征融合方法从3个方面对YOLOv5s网络进行交通标志识别的性能改进,如图2所示:在Neck网络中的CSP2模块引进CBAM空间通道注意力机制;在检测头网络中各层级部分加入ASFF模块;将检测框损失函数由GIoU改进为CIoU混合函数。
本文通过引入注意力机制增强模型对交通标志小目标的细节提取能力。CBAM空间通道注意力机制[15]由通道注意力模块(Channel Attention Module,CAM)和空间注意力模块(Spatial Attention Module,SAM)组成,使神经网络能够自适应地聚焦于输入数据中最重要的部分。
本文将空间通道注意力机制模块加入YOLOv5s中Neck网络的CSP2模块,以增强模型对输入特征的敏感度。CBAM空间通道注意力模块如图3所示。通道注意力模块:模型首先分别通过全局平均池化(Global Average Pooling,GAP)和全局最大池化(Global Max Pooling,GMP)处理特征图的通道维度,随后,通过1×1的卷积操作将综合特征表示转化为初始权重,再引入Sigmoid非线性激活函数对初始权重进行变换,从而实现对通道特征的有效加权。空间注意力模块:模型首先对特征图进行通道池化操作,以提取空间的代表性特征;然后,将经过池化的特征图与原始特征图进行通道维度堆叠,形成新的特征表示;随后,模型再通过一个7×7卷积核(Convolutional Kernel),以生成每个空间位置的权重;最后,通过Sigmoid激活函数对空间权重进行非线性变换,以增强模型对重要空间特征的识别能力。其表达式为:
$W=\sigma ({f}_{MLP}{w}_{1}+{f}_{MLP}{w}_{2})$
${W}^{\text{'}}=\sigma \left[{f}_{conv}{f}_{cat}\right({w}_{3},{w}_{4}\left)\right]$
式中:W为通道注意力权重,W′为空间注意力权重,σ为Sigmoid激活函数,fMLP为1×1卷积函数,w1w2w3w4为池化的权重向量,fconv为7×7卷积函数,fcat为向量堆叠函数。
YOLOv5s结构中路径聚合网络(Path Aggregation Network,PANet)模块常被采纳为特征融合的内核。PANet通过自底向上与自顶向下的双路径机制实现跨层级融合,通过不同层级特征图之间的语义关联来增强目标检测的准确性,但该方法在处理特征图时并未深入挖掘利用多尺度特征的潜力。由此,本文在检测头网络中引入一种ASFF[16]机制,通过在多尺度特征图之间进行融合,来增强模型对各种尺寸目标的识别性能。
ASFF的结构如图4所示,其中Level 1、Level 2和Level 3分别代表不同层级的融合策略:Level 1专注于最精细的尺度,利用高分辨率特征图来捕捉小尺寸目标的细节;Level 2采用中等分辨率,旨在识别中等尺寸的目标,同时降低计算成本;Level 3关注最粗的尺度,通过最低分辨率的特征图来识别大尺寸目标的整体轮廓。在ASFF框架下,特征融合的权重会根据各层级对任务的贡献进行自适应调整,从而优化融合效果。此外,ASFF通过空间域的筛选机制,有效排除了不协调的特征响应,提升了特征融合的一致性。
本文通过引入ASFF模块,使网络能够直接在不同层级对特征进行空间滤波,从而只保留有用的信息进行融合,进而筛选出对目标检测有益的特征,实现特征的高效整合。通过这种方式,ASFF不仅提升了特征的表达能力,还增强了模型对复杂场景的适应性。
YOLOv5算法采用GIoU[17]损失函数,该函数通过综合考量预测框与真实框的重叠面积、形状和大小,为模型提供更为全面的优化目标。位置回归损失的评价指标交并比(Intersection over Union,IoU)[18]通过计算2个边界框的交集与并集的比值衡量其相似性。GIoU损失函数进一步优化了边界框的位置预测,使得模型能够实现更高的精度和鲁棒性,基于GIoU的损失函数LGIoU的计算公式为:
${L}_{GIoU}=1-{I}_{IoU}+\frac{\left|B\right.-(A\bigcap {A}_{1} ) |}{\left|B\right|}$
其中:
${I}_{IoU}=\frac{\left|A\right.\bigcap {A}_{1}|}{\left|A\right.\bigcup {A}_{1}|}$
式中:A为真实框,A1为预测框,B为包含AA1的最小凸集,IIoU为交集与并集的比值。
GIoU损失函数虽然在处理预测框和真实框不相交的情况时表现出色,但预测框完全包含在真实框内时,GIoU损失函数会退化为传统的IoU损失函数,无法进一步区分预测框和真实框的相对位置关系,从而限制其优化潜力。因此,本文引入CIoU[19]损失函数和焦点损失函数相融合来替代原有GIoU函数,进而实现对交通标志锚点框的精确定位。CIoU损失函数优化了模型对目标形状和位置的识别能力,特别是在交通标志识别领域的街道复杂场景下。首先,通过式(4)计算出预测框和真实框之间的交集和并集区,再计算出中心点距离ρ(b,bgt)、对角线距离c;其次,通过IoU损失函数值计算权重参数αν,衡量预测框和真实框的纵横比的一致性;最后,结合上述因素计算CIoU损失函数值。此外,焦点损失函数通过定义预测概率Pt有效缓解各样本类别不平衡的问题,确保模型对预测框的置信度正确可信,进一步提升模型对小目标或稀有类别的检测准确性,使模型在处理具有高度相似性或低频出现的交通标志时,表现出卓越的性能。因此,CIoU损失函数在目标检测任务中的应用,不仅增强了模型的泛化能力,也提高了其在实际应用中的可靠性,基于CIoU的损失函数LGIoU的计算公式为:
${L}_{CIoU}=1-{I}_{IoU}+\frac{{\rho }^{2}\left(b,{b}_{gt}\right)}{{c}^{2}}+\alpha \nu $
其中:
$\alpha =\frac{\nu }{\left(1-{I}_{IoU}\right)+\nu }$
$\nu =\frac{4}{{\pi }^{2}}{(arctan\frac{{W}_{gt}}{{h}_{gt}}-arctan\frac{w}{h})}^{2}$
最终损失值为:
${L}_{FL}=-{\alpha }_{t}{\left(1-{P}_{t}\right)}^{\gamma }log \left({P}_{t}\right)$
式中:bbgt分别为预测框和真实框的中心点坐标,c为最小包围框的对角线长度,wh分别为预测框的宽度、高度,Wgthgt分别为真实框的宽度、高度,αt为平衡因子,γ为调节因子。
本文所选用的数据集为中国交通标志开源数据集TT100K。TT100K具有庞大的数据量和丰富的语义信息,其数据样本是通过高清摄像头在真实街道环境中拍摄获得的,能够真实还原复杂的城市街道背景。图5展示了TT100K数据集中的样本图像,包含多样的城市街道元素。在TT100K数据集中,本文进一步筛选出45类具有重要语义信息的交通标志,如图6所示,包括限速指示牌、禁止通行指示牌和禁止停车指示牌等。本文将增强后的11 584份数据样本按照8∶1∶1的比例划分为训练集、验证集和测试集,分别包含8 901张、1 341张和1 342张图像,图7所示为45类交通标志在增强后数据集中的数量分布情况。
为了验证模型对交通标志的识别性能,本文采用精确率(Precision)、召回率(Recall)、平均精度均值(mean Average Precision,mAP)和F1分数(F1 Score)作为评价指标评估网络性能:
$P=\frac{{G}_{TP}}{{G}_{TP}+{G}_{FP}}\times 100$%
$R=\frac{{G}_{TP}}{{G}_{TP}+{G}_{FN}}\times 100$%
${H}_{mAP}=\frac{1}{M}\sum _{k=1}^{M}\left({H}_{AP}\right(k\left)\right)\times 100$%
${F}_{1}=\frac{2PR}{P+R}\times 100$%
式中:GTPGFPGFN分别为真正例、假正例、假负例数量,M为检测目标的类别总数,HAP(k)$={\int }_{0}^{1}P\left(R\right)dR\times 100$%为第k个类别的平均精度(Average Precision,AP)。
本文选择PyTorch深度学习框架,并采用Python作为编程语言,验证环境如表1所示,在模型训练阶段,设置轮次(epoch)为200次,批量大小(Batch_size)为16。选择随机梯度下降(Stochastic Gradient Descent,SGD)法,并对其进行了学习率优化,初始学习率设定为0.01。此外,为评估模型性能,将置信度阈值统一设定为0.5。为提高网络模型的训练效率,本文采用LabelImg软件对交通标志图像进行标注。
改进算法验证结果如图8所示。由图8可知,本文提出的YOLOv5s-CBAM-ASFF算法模型在多个关键性能指标上均有显著提升:mAP达到84.5%,与YOLOv5s算法相比,提升了9.7百分点,F1分数由69.3%提升至82.2%,而且在精确率和召回率方面均有较好的表现。这充分说明YOLOv5s-CBAM-ASFF算法能更准确地对交通标志进行识别。
为了验证不同策略对网络性能的影响,本文选择在颈部网络引进CBAM空间通道注意力机制模块,并在检测头网络添加ASFF模块开展消融试验。在相同环境下,构建YOLOv5s、YOLOv5s-CBAM、YOLOv5s-ASFF和YOLOv5s-CBAM-ASFF算法模型并比较其识别结果,如表2所示。
结果表明,4种模型均可完成复杂环境下的交通标志检测识别任务。其中,YOLOv5s-CBAM-ASFF的交并比阈值为0.5时的mAP(mAP@0.5)和F1分数相较于YOLOv5s网络模型分别提高了9.7百分点和12.9百分点,同时,精确率和召回率也高于其他3种模型。
为了验证本文改进算法的可行性,将其与SSD算法[20]、中心点检测网络(CenterNet)[21]、YOLOv3、YOLOv4、YOLOv5s等主流算法进行对比验证,试验参数与本文模型参数保持一致,差异参数保持默认设置,结果如表3所示。
表3可知,改进算法各性能参数保持在相对较高的水平,充分展现了其优化后的性能优势。
针对复杂的城市街道背景条件下交通标志的小目标识别问题,本文基于YOLOv5s网络模型提出了一种改进的YOLOv5s-CBAM-ASFF算法。通过引入空间通道注意力机制和添加多尺度特征融合模块,并采用CIoU损失函数替代GIoU损失函数进行优化,进而去除复杂的背景干扰和有效融合多尺度特征信息,进一步提高对小目标的识别能力。在TT100K数据集上的验证结果表明,与原始YOLOv5s模型相比,YOLOv5s-CBAM-ASFF算法的精确率P、召回率R、平均精度均值mAP和F1分数分别实现了8.4百分点、10.9百分点、9.7百分点和12.9百分点的显著提升。
在未来的工作中,考虑到自然场景的复杂性和多样性,需要进一步研究雨雪天气、遮挡和不同光照强度等环境因子的影响,并丰富交通标志数据集中的交通标志种类和数量,训练和增强模型性能,以满足更多场景应用需求。
参考文献 引证文献
排序方式:
[1]
WILLIAM M M, ZAKI P S, SOLIMAN B K, et al. Traffic Signs Detection and Recognition System Using Deep Learning[C]// Proceedings of the 2019 Ninth International Conference on Intelligent Computing and Information Systems (ICICIS). Cairo, Egypt: IEEE, 2019: 160-166.
[2]
ROBERTS L. Machine Perception of Three-Dimensional Solids[D]. Massachusetts, USA: Massachusetts Institute of Technology, 1963.
[3]
LIENHART R, MAYDT J. An Extended Set of Haar-Like Features for Rapid Object Detection[C]// Proceedings of the 2002 International Conference on Image Processing. Rochester, NY, USA: IEEE, 2002.
[4]
DALAL N, TRIGGS B. Histograms of Oriented Gradients for Human Detection[C]// Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Diego, California, USA: IEEE, 2005: 886-893.
[5]
REDMON J, FARHADI A. YOLOv3:An Incremental Improvement[EB/OL]. (2018-04-08) [2022-11-20]. https://arxiv.org/abs/1804.02767 https://arxiv.org/abs/1804.02767
[6]
LIU W, ANGUELOV D, ERHAN D, et al. SSD: Single Shot MultiBox Detector[M]// LEIBE B, MATAS J, SEBE N, et al. Computer Vision-ECCV 2016. Lecture Notes in Computer Science. Cham: Springer, 2016: 21-37.
[7]
伍晓晖, 田启川. 交通标志识别方法综述[J]. 计算机工程与应用, 2020, 56(10): 20-26.
WU X H, TIAN Q C. Survey of Traffic Sign Recognition[J]. Computer Engineering and Applications, 2020, 56(10): 20-26.
[8]
廖璐明, 张伟, 王亚刚. 基于注意力机制的交通标志识别[J]. 小型微型计算机系统, 2021, 42(6): 1262-1266.
LIAO L M, ZHANG W, WANG Y G. Traffic Sign Recognition Based on Attention Mechanism[J]. Journal of Miniaturization for Microcomputers, 2021, 42(6): 1262-1266.
[9]
江金洪, 鲍胜利, 史文旭, 等. 基于YOLOv3算法改进的交通标志识别算法[J]. 计算机应用, 2020, 40(8): 2472-2478.
JIANG J H, BAO S L, SHI W X, et al. Improvement of Traffic Sign Recognition Algorithm Based on YOLOv3[J]. Computer Applications, 2020, 40(8): 2472-2478.
[10]
黄志强, 李军. 基于空间通道注意力机制与多尺度融合的交通标志识别研究[J]. 南京邮电大学学报(自然科学版), 2022, 42(2): 93-102.
HUANG Z Q, LI J. Research on Traffic Sign Recognition Based on Spatial Channel Attention Mechanism and Multi-Scale Fusion[J]. Journal of Nanjing University of Posts and Telecommunications (Natural Science Edition), 2022, 42(2): 93-102.
[11]
郭继峰, 孙文博, 庞志奇, 等. 一种改进YOLOv4的交通标志识别算法[J]. 小型微型计算机系统, 2022, 43(7): 1471-1476.
GUO J F, SUN W B, PANG Z Q, et al. An Improved YOLOv4 Algorithm for Traffic Sign Recognition[J]. Journal of Miniaturization for Microcomputers, 2022, 43(7): 1471-1476.
[12]
查超能, 罗素云, 何佳. 雨天下基于注意力机制与特征融合的交通标志识别[J]. 重庆理工大学学报(自然科学), 2023, 37(11): 20-31.
ZHA C N, LUO S Y, HE J. Traffic Sign Recognition Based on Attention Mechanism and Feature Fusion under Rainy Conditions[J]. Journal of Chongqing University of Technology (Natural Science), 2023, 37(11): 20-31.
[13]
尹靖涵, 瞿绍军, 姚泽楷, 等. 基于YOLOv5的雾霾天气下交通标志识别模型[J]. 计算机应用, 2022, 42(9): 2876-2884.
YIN J H, QU S J, YAO Z K, et al. A Traffic Sign Recognition Model Based on YOLOv5 in Hazy Weather[J]. Computer Applications, 2022, 42(9): 2876-2884.
[14]
SHI X H, HU J, LEI X Y, et al. Detection of Flying Birds in Airport Monitoring Based on Improved YOLOv5[C]// Proceedings of the 6th International Conference on Intelligent Computing and Signal Processing. Xi'an, China: IEEE, 2021: 1446-1451.
[15]
WOO S, PARK J, LEE J Y, et al. CBAM: Convolutional Block Attention Module[C]// Proceedings of the European Conference on Computer Vision (ECCV). Munich, Germany: Springer, 2018: 3-19.
[16]
LIU S T, HUANG D, WANG Y H. Learning Spatial Fusion for Single-Shot Object Detection[EB/OL]. (2019-11-25) [2024-09-08]. https://arxiv.org/abs/1911.09516 https://arxiv.org/abs/1911.09516
[17]
REZATOFIGHI H, TSOI N, GWAK J, et al. Generalized Intersection Over Union: A Metric and a Loss for Bounding Box Regression[C]// Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach, CA, USA: IEEE, 2019: 658-666.
[18]
YU J H, JIANG Y N, WANG Z Y, et al. UnitBox: An Advanced Object Detection Network[C]// Proceedings of the 24th ACM International Conference on Multimedia. Amsterdam, The Netherlands: ACM, 2016: 516-520.
[19]
WANG X F, SONG J. ICIoU: Improved Loss Based on Complete Intersection Over Union for Bounding Box Regression[J]. IEEE Access, 2021, 9: 105686-105695.
[20]
LIU W, ANGUELOV D, ERHAN D, et al. SSD: Single Shot MultiBox Detector[C]// European Conference on Computer Vision. Cham, Switzerland: Springer, 2016.
[21]
ZHOU X Y, WANG D Q, KRÄHENBÜHL P. Objects as Points[EB/OL]. (2019-04-25) [2024-09-08]. https://arxiv.org/abs/1904.07850 https://arxiv.org/abs/1904.07850
2025年第卷第8期
PDF下载
437
176
引用本文
BibTeX
文章信息
doi: 10.20104/j.cnki.1674-6546.20240263
  • 首发时间:2025-10-29
  • 出版时间:2025-08-15
补充材料
相关文章
文章信息
作者
出版历史
  • 修回日期:2024-09-08
基金
作者信息
    桂林电子科技大学, 桂林 541000
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/qcgcs/CN/10.20104/j.cnki.1674-6546.20240263
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