Article(id=1148106722369069281, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1148106708670477182, articleNumber=1003-3033(2025)03-0069-08, orderNo=null, doi=10.16265/j.cnki.issn1003-3033.2025.03.1181, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1729526400000, receivedDateStr=2024-10-22, revisedDate=1734969600000, revisedDateStr=2024-12-24, acceptedDate=null, acceptedDateStr=null, onlineDate=1751659573404, onlineDateStr=2025-07-05, pubDate=1743091200000, pubDateStr=2025-03-28, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1751659573404, onlineIssueDateStr=2025-07-05, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1751659573404, creator=13701087609, updateTime=1751659573404, updator=13701087609, issue=Issue{id=1148106708670477182, tenantId=1146029695717560320, journalId=1146031787341344770, year='2025', volume='35', issue='3', pageStart='1', pageEnd='268', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1751659570138, creator=13701087609, updateTime=1757401518130, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1172190184155238915, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1148106708670477182, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1172190184155238916, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1148106708670477182, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=69, endPage=76, ext={EN=ArticleExt(id=1149767560096301593, articleId=1148106722369069281, tenantId=1146029695717560320, journalId=1146031787341344770, language=EN, title=Research on tunnel fire detection based on improved YOLOv8s model, columnId=1149733269173878863, journalTitle=China Safety Science Journal, columnName=Safety engineering technology, runingTitle=null, highlight=null, articleAbstract=

To accurately and efficiently detect fires in complex tunnel environments,an enhanced YOLOv8s-based tunnel fire detection algorithm was proposed. Firstly,the Cross-Stage Partial Transformer Block (CSP-PTB) module was introduced to reconstruct the backbone network structure,thereby reducing computational complexity while preserving feature extraction capabilities. Secondly,CBAM was integrated to enhance the perception of the model of key areas and improve the discriminative power of feature representation. Finally,the Normalized Wasserstein Distance (NWD) loss function was employed to optimize the training process,effectively addressing the issue of insufficient detection accuracy for small targets. Experimental results demonstrate that the improved YOLOv8s model achieves a mean average precision (mAP) of 0.848,representing a 2% improvement over the original YOLOv8s model. The recall rate reachs 0.812,marking a significant increase of 9.3% compared to the original model. Additionally,the computational cost (GFLOPS) of the improved model is reduced by 6.7%,achieving dual objectives of performance enhancement and efficiency optimization. Compared with mainstream object detection models such as Faster R-CNN(Faster Region-based Convolutional Neural Network),SSD(Single Shot MultiBox Detector),and YOLOv5s,the improved model exhibits superior performance,with mAP improvements of 7.3%,10.1%,and 4.2%,respectively,thus meeting the stringent requirements for tunnel fire detection.

, correspAuthors=Quanjie LIU, 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=Chunyuan WANG, Quanjie LIU), CN=ArticleExt(id=1148106729386140595, articleId=1148106722369069281, tenantId=1146029695717560320, journalId=1146031787341344770, language=CN, title=基于改进YOLOv8s模型的隧道火灾检测, columnId=1149733269727526997, journalTitle=中国安全科学学报, columnName=安全工程技术, runingTitle=null, highlight=null, articleAbstract=

为准确高效地检测复杂环境隧道火灾,提出一种基于改进YOLOv8s的隧道火灾检测算法。首先,引入跨阶段部分变换器模块(CSP-PTB)重构主干网络结构,在降低计算复杂度的同时保持特征提取能力;其次,融入卷积注意力(CBAM)增强模型对关键区域的感知能力,提升特征表征的判别性;最后,采用归一化高斯瓦瑟斯坦距离(NWD)损失函数优化训练过程,有效解决小目标检测精度不足的问题。检测试验结果表明:改进后的YOLOv8s模型平均精度均值(mAP)为0.848,比原版YOLOv8s模型提升2%;召回率为0.812,较原模型大幅提升9.3%;同时模型计算量(GFLOPS)减少6.7%,实现性能提升与效率优化的双重目标。与主流目标检测模型比,改进模型的mAP较快速区域卷积神经网络(Faster R-CNN)、单发多框检测(SSD)和YOLOv5s分别提升7.3%、10.1%和4.2%。

, correspAuthors=刘权捷, authorNote=null, correspAuthorsNote=
** 刘权捷(2000—),男,辽宁抚顺人,硕士研究生,研究方向为智能交通安全及目标检测跟踪算法。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=KkZYtTr+sMrAAVu0ONYt1g==, magXml=1lc0G9Xh6yiYX47Ri1JNng==, pdfUrl=null, pdf=lZONhloG3tKponZye3+JMg==, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=FPl4KoMdg/F4dgO4gfmTgQ==, mapNumber=null, authorCompany=null, fund=null, authors=

王春源 (1980—),男,山东诸城人,硕士,副教授,主要从事地下工程智能减灾防灾技术和安全管理方面的研究。E-mail:

, authorsList=王春源, 刘权捷)}, authors=[Author(id=1165678457322221573, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=chunyuanmail@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1165678457380941831, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, authorId=1165678457322221573, language=EN, stringName=Chunyuan WANG, firstName=Chunyuan, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=School of Mechanical and Automotive Engineering,Qingdao University of Technology,Qingdao Shandong 266520,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1165678457469022216, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, authorId=1165678457322221573, language=CN, stringName=王春源, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=青岛理工大学 机械与汽车工程学院,山东 青岛 266520, bio={"img":"Vwqv7pWtiE60EmIMAOqyXA==","content":"

王春源 (1980—),男,山东诸城人,硕士,副教授,主要从事地下工程智能减灾防灾技术和安全管理方面的研究。E-mail:

"}, bioImg=Vwqv7pWtiE60EmIMAOqyXA==, bioContent=

王春源 (1980—),男,山东诸城人,硕士,副教授,主要从事地下工程智能减灾防灾技术和安全管理方面的研究。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1165678457250918401, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, xref=null, ext=[AuthorCompanyExt(id=1165678457259307010, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, companyId=1165678457250918401, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Mechanical and Automotive Engineering,Qingdao University of Technology,Qingdao Shandong 266520,China), AuthorCompanyExt(id=1165678457263501315, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, companyId=1165678457250918401, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=青岛理工大学 机械与汽车工程学院,山东 青岛 266520)])]), Author(id=1165678457527742474, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=lqj11150218@163.com, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1165678457590657036, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, authorId=1165678457527742474, language=EN, stringName=Quanjie LIU, firstName=Quanjie, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=**, address=School of Mechanical and Automotive Engineering,Qingdao University of Technology,Qingdao Shandong 266520,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1165678457645182989, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, authorId=1165678457527742474, language=CN, stringName=刘权捷, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=**, address=青岛理工大学 机械与汽车工程学院,山东 青岛 266520, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1165678457250918401, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, xref=null, ext=[AuthorCompanyExt(id=1165678457259307010, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, companyId=1165678457250918401, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Mechanical and Automotive Engineering,Qingdao University of Technology,Qingdao Shandong 266520,China), AuthorCompanyExt(id=1165678457263501315, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, companyId=1165678457250918401, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=青岛理工大学 机械与汽车工程学院,山东 青岛 266520)])])], keywords=[Keyword(id=1165678457762623502, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, orderNo=1, keyword=YOLOv8 model), Keyword(id=1165678457808760847, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, orderNo=2, keyword=tunnel fire detection), Keyword(id=1165678457875869712, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, orderNo=3, keyword=convolutional neural network(CNN)), Keyword(id=1165678457922007057, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, orderNo=4, keyword=convolutional block attention module (CBAM)), Keyword(id=1165678457972338706, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, orderNo=5, keyword=loss function), Keyword(id=1165678458026864659, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, orderNo=1, keyword=YOLOv8模型), Keyword(id=1165678458089779220, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, orderNo=2, keyword=隧道火灾检测), Keyword(id=1165678458148499477, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, orderNo=3, keyword=卷积神经网络(CNN)), Keyword(id=1165678458207219734, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, orderNo=4, keyword=卷积注意力(CBAM)), Keyword(id=1165678458257551383, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, orderNo=5, keyword=损失函数)], refs=[Reference(id=1165678460308566067, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2023, volume=46, issue=3, pageStart=118, pageEnd=120, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=李东平, journalName=黑龙江交通科技, refType=null, unstructuredReference=李东平. 公路隧道的服役现状安全性评价管理体系构建[J]. 黑龙江交通科技, 2023, 46(3): 118-120., articleTitle=公路隧道的服役现状安全性评价管理体系构建, refAbstract=null), Reference(id=1165678460371480628, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2023, volume=46, issue=3, pageStart=118, pageEnd=120, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=LI Dongping, journalName=Communications Science and Technology Heilongjiang, refType=null, unstructuredReference=LI Dongping. Construction of safety evaluation management system for service status of highway tunnels[J]. Communications Science and Technology Heilongjiang, 2023, 46(3): 118-120., articleTitle=Construction of safety evaluation management system for service status of highway tunnels, refAbstract=null), Reference(id=1165678460434395189, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2021, volume=37, issue=增1, pageStart=83, pageEnd=86, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=卢毅, 田芳, 陈建忠, journalName=公路交通技术, refType=null, unstructuredReference=卢毅, 田芳, 陈建忠, 等. 公路隧道火灾灾情监测技术现状及展望[J]. 公路交通技术, 2021, 37(增1): 83-86, 92., articleTitle=公路隧道火灾灾情监测技术现状及展望, refAbstract=null), Reference(id=1165678460493115446, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2021, volume=37, issue=S1, pageStart=83, pageEnd=86, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=LU Yi, TIAN Fang, CHEN Jianzhong, journalName=Technology of Highway and Transport, refType=null, unstructuredReference=LU Yi, TIAN Fang, CHEN Jianzhong, et al. Studies and prospects of technologies for highway tunnel fire monitoring[J]. Technology of Highway and Transport, 2021, 37(S1): 83-86, 92., articleTitle=Studies and prospects of technologies for highway tunnel fire monitoring, refAbstract=null), Reference(id=1165678460551835703, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=朱学慧, journalName=基于图像处理的公路隧道火灾检测技术研究, refType=null, unstructuredReference=朱学慧. 基于图像处理的公路隧道火灾检测技术研究[D]. 杭州: 中国计量大学, 2021., articleTitle=null, refAbstract=null), Reference(id=1165678460606361656, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=5, authorNames=ZHU Xuehui, journalName=Research on fire detection technology of highway tunnel based on image processing, refType=null, unstructuredReference=ZHU Xuehui. Research on fire detection technology of highway tunnel based on image processing[D]. Hangzhou: China University of Metrology, 2021., articleTitle=null, refAbstract=null), Reference(id=1165678460673470521, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2022, volume=42, issue=2, pageStart=291, pageEnd=302, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=邓实强, 丁浩, 杨孟, journalName=隧道建设:中英文, refType=null, unstructuredReference=邓实强, 丁浩, 杨孟, 等. 基于视频图像的公路隧道火灾烟雾检测[J]. 隧道建设:中英文, 2022, 42(2): 291-302., articleTitle=基于视频图像的公路隧道火灾烟雾检测, refAbstract=null), Reference(id=1165678460732190778, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2022, volume=42, issue=2, pageStart=291, pageEnd=302, url=null, language=null, rfNumber=[4], rfOrder=7, authorNames=DENG Shiqiang, DING Hao, YANG Meng, journalName=Tunnel Construction, refType=null, unstructuredReference=DENG Shiqiang, DING Hao, YANG Meng, et al. Fire smoke detection in highway tunnels based on video images[J]. Tunnel Construction, 2022, 42(2): 291-302., articleTitle=Fire smoke detection in highway tunnels based on video images, refAbstract=null), Reference(id=1165678460786716731, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=11, pageStart=155, pageEnd=157, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=吕宇, 张子俊, journalName=中国设备工程, refType=null, unstructuredReference=吕宇, 张子俊. 红外热成像技术在隧道火灾检测中的研究与应用[J]. 中国设备工程, 2023(11): 155-157., articleTitle=红外热成像技术在隧道火灾检测中的研究与应用, refAbstract=null), Reference(id=1165678460849631292, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=11, pageStart=155, pageEnd=157, url=null, language=null, rfNumber=[5], rfOrder=9, authorNames=LYU Yu, journalName=China Plant Engineering, refType=null, unstructuredReference=LYU Yu, ZHANG Zijun. Research and application of infrared thermal imaging technology in tunnel fire detection[J]. China Plant Engineering, 2023(11): 155-157., articleTitle=Research and application of infrared thermal imaging technology in tunnel fire detection, refAbstract=null), Reference(id=1165678460899962941, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=21, pageEnd=37, url=null, language=null, rfNumber=[6], rfOrder=10, authorNames=LIU Wei, ANGUELOV D, ERHAN D, journalName=Computer Vision-ECCV 2016: 14th European Conference, refType=null, unstructuredReference=LIU Wei, ANGUELOV D, ERHAN D, et al. SSD: single shot multibox detector[C]. Computer Vision-ECCV 2016: 14th European Conference, 2016: 21-37., articleTitle=SSD: single shot multibox detector, refAbstract=null), Reference(id=1165678460979654718, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2016, volume=39, issue=6, pageStart=1137, pageEnd=1149, url=null, language=null, rfNumber=[7], rfOrder=11, authorNames=REN Shaoqing, HE Kaiming, GIRSHICK R, journalName=IEEE Transactions on Pattern Analysis and Machine Intelligence, refType=null, unstructuredReference=REN Shaoqing, HE Kaiming, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 39(6): 1137-1149., articleTitle=Faster R-CNN: towards real-time object detection with region proposal networks, refAbstract=null), Reference(id=1165678461050957887, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=2961, pageEnd=2969, url=null, language=null, rfNumber=[8], rfOrder=12, authorNames=HE Kaiming, GKIOXARI G, DOLLAR P, journalName=Proceedings of the IEEE International Conference on Computer Vision, refType=null, unstructuredReference=HE Kaiming, GKIOXARI G, DOLLAR P, et al. Mask R-CNN[C]. Proceedings of the IEEE International Conference on Computer Vision, 2017: 2961-2969., articleTitle=Mask R-CNN, refAbstract=null), Reference(id=1165678461097095232, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=6105, pageEnd=6114, url=null, language=null, rfNumber=[9], rfOrder=13, authorNames=TAN Ming, LE Qing, journalName=International Conference on Machine Learning, refType=null, unstructuredReference=TAN Ming, LE Qing. Efficientnet: rethinking model scaling for convolutional neural networks[C]. International Conference on Machine Learning, 2019: 6105-6114., articleTitle=Efficientnet: rethinking model scaling for convolutional neural networks, refAbstract=null), Reference(id=1165678461151621185, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=14, authorNames=唐小垚, journalName=基于深度学习的公路隧道火灾检测研究, refType=null, unstructuredReference=唐小垚. 基于深度学习的公路隧道火灾检测研究[D]. 重庆: 重庆交通大学, 2023., articleTitle=null, refAbstract=null), Reference(id=1165678461206147138, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=15, authorNames=TANG Xiaoyao, journalName=Research on highway tunnel fire detection based on deep learning, refType=null, unstructuredReference=TANG Xiaoyao. Research on highway tunnel fire detection based on deep learning[D]. Chongqing: Chongqing Jiaotong University, 2023., articleTitle=null, refAbstract=null), Reference(id=1165678461260673091, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2023, volume=33, issue=10, pageStart=214, pageEnd=223, url=null, language=null, rfNumber=[11], rfOrder=16, authorNames=马庆禄, 孙枭, 唐小垚, journalName=中国安全科学学报, refType=null, unstructuredReference=马庆禄, 孙枭, 唐小垚, 等. 基于改进YOLOv5s算法的隧道初期火灾检测模型[J]. 中国安全科学学报, 2023, 33(10): 214-223., articleTitle=基于改进YOLOv5s算法的隧道初期火灾检测模型, refAbstract=null), Reference(id=1165678461315199044, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2023, volume=33, issue=10, pageStart=214, pageEnd=223, url=null, language=null, rfNumber=[11], rfOrder=17, authorNames=MA Qinglu, SUN Xiao, TANG Xiaoyao, journalName=China Safety Science Journal, refType=null, unstructuredReference=MA Qinglu, SUN Xiao, TANG Xiaoyao, et al. Optimization method for tunnel initial fire detection based on YOLOv5s algorithm[J]. China Safety Science Journal, 2023, 33(10): 214-223., articleTitle=Optimization method for tunnel initial fire detection based on YOLOv5s algorithm, refAbstract=null), Reference(id=1165678461369724997, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2023, volume=59, issue=2, pageStart=222, pageEnd=231, url=null, language=null, rfNumber=[12], rfOrder=18, authorNames=张晋瑞, 宋焕生, 孙士杰, journalName=计算机工程与应用, refType=null, unstructuredReference=张晋瑞, 宋焕生, 孙士杰, 等. 改进YOLOv5的隧道火灾帧差检测网络与应用方法[J]. 计算机工程与应用, 2023, 59(2): 222-231., articleTitle=改进YOLOv5的隧道火灾帧差检测网络与应用方法, refAbstract=null), Reference(id=1165678461424250950, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2023, volume=59, issue=2, pageStart=222, pageEnd=231, url=null, language=null, rfNumber=[12], rfOrder=19, authorNames=ZHANG Jinrui, SONG Huansheng, SUN Shijie, journalName=Computer Engineering and Applications, refType=null, unstructuredReference=ZHANG Jinrui, SONG Huansheng, SUN Shijie, et al. Tunnel fire frame difference detection network by improving YOLOv5 and application method[J]. Computer Engineering and Applications, 2023, 59(2): 222-231., articleTitle=Tunnel fire frame difference detection network by improving YOLOv5 and application method, refAbstract=null), Reference(id=1165678461474582599, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2023, volume=44, issue=5, pageStart=978, pageEnd=987, url=null, language=null, rfNumber=[13], rfOrder=20, authorNames=宋焕生, 文雅, 孙士杰, journalName=图学学报, refType=null, unstructuredReference=宋焕生, 文雅, 孙士杰, 等. 基于改进教师学生网络的隧道火灾检测[J]. 图学学报, 2023, 44(5): 978-987., articleTitle=基于改进教师学生网络的隧道火灾检测, refAbstract=null), Reference(id=1165678461533302856, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2023, volume=44, issue=5, pageStart=978, pageEnd=987, url=null, language=null, rfNumber=[13], rfOrder=21, authorNames=SONG Huansheng, WEN Ya, SUN Shijie, journalName=Journal of Graphics, refType=null, unstructuredReference=SONG Huansheng, WEN Ya, SUN Shijie, et al. Tunnel fire detection based on improved student-teacher network[J]. Journal of Graphics, 2023, 44(5): 978-987., articleTitle=Tunnel fire detection based on improved student-teacher network, refAbstract=null), Reference(id=1165678461583634505, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2022, volume=31, issue=4, pageStart=244, pageEnd=251, url=null, language=null, rfNumber=[14], rfOrder=22, authorNames=马庆禄, 马恋, 孔国英, journalName=火灾科学, refType=null, unstructuredReference=马庆禄, 马恋, 孔国英, 等. 基于红外热成像的隧道火焰检测技术研究[J]. 火灾科学, 2022, 31(4): 244-251., articleTitle=基于红外热成像的隧道火焰检测技术研究, refAbstract=null), Reference(id=1165678461642354762, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2022, volume=31, issue=4, pageStart=244, pageEnd=251, url=null, language=null, rfNumber=[14], rfOrder=23, authorNames=MA Qinglu, MA Lian, KONG Guoying, journalName=Fire Safety Science, refType=null, unstructuredReference=MA Qinglu, MA Lian, KONG Guoying, et al. Research on tunnel flame detection technology based on infrared thermal imaging[J]. Fire Safety Science, 2022, 31(4): 244-251., articleTitle=Research on tunnel flame detection technology based on infrared thermal imaging, refAbstract=null), Reference(id=1165678461696880715, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=1, pageEnd=6, url=null, language=null, rfNumber=[15], rfOrder=24, authorNames=VARGHESE R, SAMBATH M, journalName=2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS), refType=null, unstructuredReference=VARGHESE R, SAMBATH M. YOLOv8: a novel object detection algorithm with enhanced performance and robustness[C]. 2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS), 2024: 1-6., articleTitle=YOLOv8: a novel object detection algorithm with enhanced performance and robustness, refAbstract=null), Reference(id=1165678461751406668, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=3, pageEnd=19, url=null, language=null, rfNumber=[16], rfOrder=25, 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), 2018: 3-19., articleTitle=CBAM: convolutional block attention module, refAbstract=null), Reference(id=1165678461810126925, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=26, authorNames=WANG Jinwang, XU Chang, YANG Wen, journalName=arXiv Preprint, refType=null, unstructuredReference=WANG Jinwang, XU Chang, YANG Wen, et al. A normalized gaussian wasserstein distance for tiny object detection[J]. arXiv Preprint, 2022: DOI: 10.48550/arXiv.2110.13389., articleTitle=A normalized gaussian wasserstein distance for tiny object detection, refAbstract=null), Reference(id=1165678461868847182, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2024, volume=52, issue=10, pageStart=11, pageEnd=18, url=null, language=null, rfNumber=[18], rfOrder=27, authorNames=李耀, 徐红伟, 柯海森, journalName=棉纺织技术, refType=null, unstructuredReference=李耀, 徐红伟, 柯海森, 等. 基于改进YOLOv8n的织物疵点检测[J]. 棉纺织技术, 2024, 52(10): 11-18., articleTitle=基于改进YOLOv8n的织物疵点检测, refAbstract=null), Reference(id=1165678461919178831, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, doi=null, pmid=null, pmcid=null, year=2024, volume=52, issue=10, pageStart=11, pageEnd=18, url=null, language=null, rfNumber=[18], rfOrder=28, authorNames=LI Yao, XU Hongwei, KE Haisen, journalName=Cotton Textile Technology, refType=null, unstructuredReference=LI Yao, XU Hongwei, KE Haisen, et al. Fabric defect detection based on improved YOLOv8n[J]. Cotton Textile Technology, 2024, 52(10): 11-18., articleTitle=Fabric defect detection based on improved YOLOv8n, refAbstract=null)], funds=[Fund(id=1165678460182736946, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, awardId=52474238, language=CN, fundingSource=国家自然科学基金资助(52474238), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1165678457250918401, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, xref=null, ext=[AuthorCompanyExt(id=1165678457259307010, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, companyId=1165678457250918401, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Mechanical and Automotive Engineering,Qingdao University of Technology,Qingdao Shandong 266520,China), AuthorCompanyExt(id=1165678457263501315, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, companyId=1165678457250918401, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=青岛理工大学 机械与汽车工程学院,山东 青岛 266520)])], figs=[ArticleFig(id=1165678458400157720, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, label=Fig.1, caption=CSP-PTB structure, figureFileSmall=S/TbsvD9xaBYyM0ZpgrbdA==, figureFileBig=tQ3QseP2DJV/IN7U04a9eQ==, tableContent=null), ArticleFig(id=1165678458467266585, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, label=图1, caption=CSP-PTB结构, figureFileSmall=S/TbsvD9xaBYyM0ZpgrbdA==, figureFileBig=tQ3QseP2DJV/IN7U04a9eQ==, tableContent=null), ArticleFig(id=1165678458530181146, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, label=Fig.2, caption=Structure of AFEB block, figureFileSmall=i9wd0Koch4yAaPJL9HLN9Q==, figureFileBig=2YxQxAHyIqJt3ilZVbWrcA==, tableContent=null), ArticleFig(id=1165678458584707099, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, label=图2, caption=AFEB结构, figureFileSmall=i9wd0Koch4yAaPJL9HLN9Q==, figureFileBig=2YxQxAHyIqJt3ilZVbWrcA==, tableContent=null), ArticleFig(id=1165678458714730524, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, label=Fig.3, caption=CBAM structure, figureFileSmall=K1HDLfDL57sleoehUaXbGw==, figureFileBig=ZPbc4LCxCBvycVO2onCtYA==, tableContent=null), ArticleFig(id=1165678458790227997, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, label=图3, caption=CBAM结构, figureFileSmall=K1HDLfDL57sleoehUaXbGw==, figureFileBig=ZPbc4LCxCBvycVO2onCtYA==, tableContent=null), ArticleFig(id=1165678458848948254, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, label=Fig.4, caption=Improved YOLOv8s structure, figureFileSmall=AgfWrgggnkMGHoqWoOkccA==, figureFileBig=lcc0cXAdMN5o2SWxP8uPmw==, tableContent=null), ArticleFig(id=1165678458907668511, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, label=图4, caption=改进后YOLOv8s结构, figureFileSmall=AgfWrgggnkMGHoqWoOkccA==, figureFileBig=lcc0cXAdMN5o2SWxP8uPmw==, tableContent=null), ArticleFig(id=1165678458978971680, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, label=Fig.5, caption=Training set original image, figureFileSmall=qzLl2qsTpelzAKbuiv7gMA==, figureFileBig=jqfVeWjlbUZ59ExBgAp/JA==, tableContent=null), ArticleFig(id=1165678459041886241, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, label=图5, caption=训练集原图片, figureFileSmall=qzLl2qsTpelzAKbuiv7gMA==, figureFileBig=jqfVeWjlbUZ59ExBgAp/JA==, tableContent=null), ArticleFig(id=1165678459092217890, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, label=Fig.6, caption=Training set image after data enhancement, figureFileSmall=RiPhfVKJhxSFTJ3Pq6Dz+g==, figureFileBig=vcKkrIPRtU1lT0omb/wyhA==, tableContent=null), ArticleFig(id=1165678459155132451, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, label=图6, caption=数据增强后训练集图片, figureFileSmall=RiPhfVKJhxSFTJ3Pq6Dz+g==, figureFileBig=vcKkrIPRtU1lT0omb/wyhA==, tableContent=null), ArticleFig(id=1165678459213852708, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, label=Fig.7, caption=Comparison of training results between improved YOLOv8s and original YOLOv8s, figureFileSmall=IzvBGupyhCggXShvTNYM5Q==, figureFileBig=R2MhYnhbRgQRVvIrATVRlA==, tableContent=null), ArticleFig(id=1165678459280961573, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, label=图7, caption=改进YOLOv8s与原版YOLOv8s训练结果对比, figureFileSmall=IzvBGupyhCggXShvTNYM5Q==, figureFileBig=R2MhYnhbRgQRVvIrATVRlA==, tableContent=null), ArticleFig(id=1165678459352264742, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, label=Fig.8, caption=Modified YOLOv8s receptive field, figureFileSmall=2xXB3Qnj9DczGMBMxVB+6g==, figureFileBig=4dhC6/ZgTi1bwzUtFq1b2Q==, tableContent=null), ArticleFig(id=1165678459419373607, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, label=图8, caption=改进YOLOv8s模型感受野, figureFileSmall=2xXB3Qnj9DczGMBMxVB+6g==, figureFileBig=4dhC6/ZgTi1bwzUtFq1b2Q==, tableContent=null), ArticleFig(id=1165678459486482472, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, label=Fig.9, caption=Original YOLOv8s receptive field, figureFileSmall=5QZsm3c3Ohi0gASZN56R2w==, figureFileBig=kXxDgzmQTVBxzOA5razc6A==, tableContent=null), ArticleFig(id=1165678459549397033, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, label=图9, caption=原版YOLOv8s模型感受野, figureFileSmall=5QZsm3c3Ohi0gASZN56R2w==, figureFileBig=kXxDgzmQTVBxzOA5razc6A==, tableContent=null), ArticleFig(id=1165678459603922986, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, label=Fig.10, caption=Original YOLOv8s model heat map, figureFileSmall=2vTyC91cUSYpFhrQ37CCeg==, figureFileBig=kiBElVu4+oC+k+sdPzQlFQ==, tableContent=null), ArticleFig(id=1165678459662643243, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, label=图10, caption=原版YOLOv8s模型热力图, figureFileSmall=2vTyC91cUSYpFhrQ37CCeg==, figureFileBig=kiBElVu4+oC+k+sdPzQlFQ==, tableContent=null), ArticleFig(id=1165678459721363500, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, label=Fig.11, caption=Improved YOLOv8s model heat map, figureFileSmall=eh6xyVIdNOaQ+bvkgdtDQQ==, figureFileBig=rOemouNZlLdKP1EdnU2d9w==, tableContent=null), ArticleFig(id=1165678459788472365, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, label=图11, caption=改进YOLOv8s热力图, figureFileSmall=eh6xyVIdNOaQ+bvkgdtDQQ==, figureFileBig=rOemouNZlLdKP1EdnU2d9w==, tableContent=null), ArticleFig(id=1165678459847192622, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, label=Table 1, caption=

Ablation experiment results

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 APF APS mAP RF RS R GFLOPS FPS/(帧·s-1)
YOLOv8s 0.899 0.765 0.832 0.838 0.648 0.743 28.4 58.33
YOLOv8s+CSP-PTB 0.903 0.781 0.842 0.833 0.626 0.730 26.5 59.47
YOLOv8s+CSP-PTB+CBAM 0.895 0.798 0.846 0.844 0.648 0.746 26.5 58.52
YOLOv8s+CSP-PTB+CBAM+NWD 0.890 0.807 0.848 0.866 0.758 0.812 26.5 58.39
), ArticleFig(id=1165678459926884399, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, label=表1, caption=

消融试验结果

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 APF APS mAP RF RS R GFLOPS FPS/(帧·s-1)
YOLOv8s 0.899 0.765 0.832 0.838 0.648 0.743 28.4 58.33
YOLOv8s+CSP-PTB 0.903 0.781 0.842 0.833 0.626 0.730 26.5 59.47
YOLOv8s+CSP-PTB+CBAM 0.895 0.798 0.846 0.844 0.648 0.746 26.5 58.52
YOLOv8s+CSP-PTB+CBAM+NWD 0.890 0.807 0.848 0.866 0.758 0.812 26.5 58.39
), ArticleFig(id=1165678459989798960, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=EN, label=Table 2, caption=

Performance comparison of different models on test set

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 mAP mAP@
0.75
mAP@
0.5:0.95
R
SSD 0.762 0.563 0.476 0.551
YOLOv5s 0.805 0.601 0.520 0.618
FR-CNN 0.782 0.534 0.478 0.548
YOLOv8s+CSP-
PTB+CBAM+NWD
0.839 0.605 0.530 0.619
), ArticleFig(id=1165678460052713521, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106722369069281, language=CN, label=表2, caption=

不同模型在测试集上的性能对比

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 mAP mAP@
0.75
mAP@
0.5:0.95
R
SSD 0.762 0.563 0.476 0.551
YOLOv5s 0.805 0.601 0.520 0.618
FR-CNN 0.782 0.534 0.478 0.548
YOLOv8s+CSP-
PTB+CBAM+NWD
0.839 0.605 0.530 0.619
)], attaches=null, journal=Journal(id=1123942128916217864, delFlag=0, nameCn=中国安全科学学报, nameEn=China Safety Science Journal, nameHistory1=null, nameHistory2=null, issn=1003-3033, eissn=, cn=11-2865/X, coden=null, periodic=0, language=CN, oaType=0, 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=fkqsFM6VKlHC4gCtS5XqTw==, journalPrice=null, startedYear=null, abbrevIsoEn=Chin Saf Sci J, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754269350027, createdBy=null, updatedBy=13701087609, firstLetterCn=C, firstLetterEn=C, subjectCode=Engineering, subjectName=工程, subjectCodeEn=Engineering, subjectNameEn=null, picCn=fkqsFM6VKlHC4gCtS5XqTw==, picEn=SHn9HgqSxtJrOcAxqD++4Q==, jcr=null, cjcr=null, exts=[JournalExt(id=1159052918994595848, language=CN, name=中国安全科学学报, 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.cssjj.com.cn/, createdTime=1754269350050, updatedTime=1754269350050, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.cssjj.com.cn/CN/column/item15.shtml, submissionAuthorUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionReviewUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionCeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionAeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, option={"copyright":""}), JournalExt(id=1159052919040733193, language=EN, name=China Safety Science Journal, 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.cssjj.com.cn/EN/1003-3033/home.shtml, createdTime=1754269350061, updatedTime=1754269350061, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=https://synbioj.cip.com.cn/EN/column/column3.shtml, submissionAuthorUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionReviewUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionCeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionAeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, option={"copyright":""})], databaseList=null, tenantJournalId=1146031787341344770, websiteList=[Website(id=1148243202345263519, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146031787341344770, 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/zgaqkxxb/CN, language=CN, createTime=1751692112766, createBy=18614031015, updateTime=1753502583634, updateBy=18614031015, name=《中国安全科学学报》中文站点, tplId=1146099689490845704, title=中国安全科学学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148618794941046792, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=articleTextType, value=kx, createTime=1751781661020, updateTime=1751781661020, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794911686661, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=banner, value=null, createTime=1751781661012, updateTime=1751781661012, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794894909444, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=tui0IVO9FMwB61HHtX5scg==, createTime=1751781661008, updateTime=1751781661008, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794932658183, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751781661017, updateTime=1751781661017, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794924269574, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751781661015, updateTime=1751781661015, creator=18614031015, updator=18614031015)]), Website(id=1155836763751993353, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146031787341344770, 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/zgaqkxxb/EN, language=EN, createTime=1753502558893, createBy=18614031015, updateTime=1753524450387, updateBy=18614031015, name=《中国安全科学学报》英文站点, tplId=1146101810881728533, title=China Safety Science Journal, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155895925743669425, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=articleTextType, value=kx, createTime=1753516664205, updateTime=1753516664205, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925722697902, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=banner, value=null, createTime=1753516664200, updateTime=1753516664200, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925714309293, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=tui0IVO9FMwB61HHtX5scg==, createTime=1753516664198, updateTime=1753516664198, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925735280816, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753516664203, updateTime=1753516664203, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925731086511, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753516664202, updateTime=1753516664202, creator=18614031015, updator=18614031015)])], journalTitle=中国安全科学学报, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=China Safety Science Journal, journalPhotoCn=fkqsFM6VKlHC4gCtS5XqTw==, journalPhotoEn=SHn9HgqSxtJrOcAxqD++4Q==, journalFirstLetter=C, journalRecommend=null, journalNew=null, journalCollection=1, 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/zgaqkxxb/CN/10.16265/j.cnki.issn1003-3033.2025.03.1181, detailUrlEn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/EN/10.16265/j.cnki.issn1003-3033.2025.03.1181, pdfUrlCn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN/PDF/10.16265/j.cnki.issn1003-3033.2025.03.1181, pdfUrlEn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/EN/PDF/10.16265/j.cnki.issn1003-3033.2025.03.1181, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于改进YOLOv8s模型的隧道火灾检测
收藏切换
PDF下载
王春源 , 刘权捷 **
中国安全科学学报 | 安全工程技术 2025,35(3): 69-76
收起
收藏切换
中国安全科学学报 | 安全工程技术 2025, 35(3): 69-76
基于改进YOLOv8s模型的隧道火灾检测
全屏
王春源 , 刘权捷**
作者信息
  • 青岛理工大学 机械与汽车工程学院,山东 青岛 266520
  • 王春源 (1980—),男,山东诸城人,硕士,副教授,主要从事地下工程智能减灾防灾技术和安全管理方面的研究。E-mail:

通讯作者:

** 刘权捷(2000—),男,辽宁抚顺人,硕士研究生,研究方向为智能交通安全及目标检测跟踪算法。E-mail:
Research on tunnel fire detection based on improved YOLOv8s model
Chunyuan WANG , Quanjie LIU**
Affiliations
  • School of Mechanical and Automotive Engineering,Qingdao University of Technology,Qingdao Shandong 266520,China
出版时间: 2025-03-28 doi: 10.16265/j.cnki.issn1003-3033.2025.03.1181
文章导航
收藏切换

为准确高效地检测复杂环境隧道火灾,提出一种基于改进YOLOv8s的隧道火灾检测算法。首先,引入跨阶段部分变换器模块(CSP-PTB)重构主干网络结构,在降低计算复杂度的同时保持特征提取能力;其次,融入卷积注意力(CBAM)增强模型对关键区域的感知能力,提升特征表征的判别性;最后,采用归一化高斯瓦瑟斯坦距离(NWD)损失函数优化训练过程,有效解决小目标检测精度不足的问题。检测试验结果表明:改进后的YOLOv8s模型平均精度均值(mAP)为0.848,比原版YOLOv8s模型提升2%;召回率为0.812,较原模型大幅提升9.3%;同时模型计算量(GFLOPS)减少6.7%,实现性能提升与效率优化的双重目标。与主流目标检测模型比,改进模型的mAP较快速区域卷积神经网络(Faster R-CNN)、单发多框检测(SSD)和YOLOv5s分别提升7.3%、10.1%和4.2%。

YOLOv8模型  /  隧道火灾检测  /  卷积神经网络(CNN)  /  卷积注意力(CBAM)  /  损失函数

To accurately and efficiently detect fires in complex tunnel environments,an enhanced YOLOv8s-based tunnel fire detection algorithm was proposed. Firstly,the Cross-Stage Partial Transformer Block (CSP-PTB) module was introduced to reconstruct the backbone network structure,thereby reducing computational complexity while preserving feature extraction capabilities. Secondly,CBAM was integrated to enhance the perception of the model of key areas and improve the discriminative power of feature representation. Finally,the Normalized Wasserstein Distance (NWD) loss function was employed to optimize the training process,effectively addressing the issue of insufficient detection accuracy for small targets. Experimental results demonstrate that the improved YOLOv8s model achieves a mean average precision (mAP) of 0.848,representing a 2% improvement over the original YOLOv8s model. The recall rate reachs 0.812,marking a significant increase of 9.3% compared to the original model. Additionally,the computational cost (GFLOPS) of the improved model is reduced by 6.7%,achieving dual objectives of performance enhancement and efficiency optimization. Compared with mainstream object detection models such as Faster R-CNN(Faster Region-based Convolutional Neural Network),SSD(Single Shot MultiBox Detector),and YOLOv5s,the improved model exhibits superior performance,with mAP improvements of 7.3%,10.1%,and 4.2%,respectively,thus meeting the stringent requirements for tunnel fire detection.

YOLOv8 model  /  tunnel fire detection  /  convolutional neural network(CNN)  /  convolutional block attention module (CBAM)  /  loss function
王春源, 刘权捷. 基于改进YOLOv8s模型的隧道火灾检测. 中国安全科学学报, 2025 , 35 (3) : 69 -76 . DOI: 10.16265/j.cnki.issn1003-3033.2025.03.1181
Chunyuan WANG, Quanjie LIU. Research on tunnel fire detection based on improved YOLOv8s model[J]. China Safety Science Journal, 2025 , 35 (3) : 69 -76 . DOI: 10.16265/j.cnki.issn1003-3033.2025.03.1181
目前,我国公路隧道的数量和总长度在全球处于领先地位[1],然而,近年来,公路隧道火灾时有发生,及时准确地识别隧道火灾已成为当今社会亟待解决的问题。公路隧道火灾检测技术主要包括传统探测器和视频图像处理技术2类,可监测火灾的发生并发出报警,但存在火灾检测信息不全面、准确性不高和泛化能力不强的不足[2]
为完善隧道火灾检测技术,学者开展了广泛且深入的研究。朱学慧[3]提出一种基于支持向量机(Support Vector Machine,SVM)的隧道火灾检测方法,通过提取火焰图像的颜色特征和纹理特征实现对火焰的准确识别。邓实强等[4]对比分析了BP神经网络、SVM和随机森林在隧道烟雾识别中的效果,探讨了机器学习在火灾检测中的应用可能性。吕宇等[5]结合红外热成像处理技术与SVM,开发了一套隧道火灾检测系统。随着人工智能技术的发展,深度学习已应用于火灾检测,卷积神经网络(Convolutional Neural Network,CNN)被引入了隧道火灾检测领域。深度学习目标检测算法有单阶段算法YOLO和单发多框检测(Single Shot MultiBox Detector,SSD)[6];2阶段算法有快速区域CNN(Faster Region-based CNN,FR-CNN)[7]和掩码区域CNN(Mask Region-based CNN,MR-CNN)[8]等;基于高效CNN网络(Efficient Convolutional Neural Network,E-Net)[9]等主干网络也逐渐应用于火灾检测领域。唐小垚[10]引入卷积注意力(Convolutional Block Attention Module,CBAM)、改进骨干网络和优化损失函数优化YOLOv5,提高了对于隧道火灾的检测能力。马庆禄等[11]在YOLOv5s中引入变换器预测头,并新增160×160的检测头,增强了原模型的多尺度识别能力。张晋瑞等[12]引入3D卷积的帧差网络,改进YOLOv5算法,提升了检测的精度和速度。宋焕生等[13]基于改进教师学生网络的隧道火灾检测方法,结合无监督学习策略、注意力、Mish激活函数和对称正定卷积(Symmetric Positive Definite Convolution,SPD-Conv)模块,解决了隧道火灾数据集匮乏和检测精度低的问题。YOLOv8由YOLOv5迭代更新而来[15],能够在处理能力有限的设备上以较高速度运行,并保持较高的检测精度。但YOLOv8系列模型仍存在不足之处,比如卷积到特征模块(Convolutional to Feature,C2f)计算量大、易过拟合,检测框损失函数(Bbox Loss,BL)对边界框完全不重叠情况处理不佳和对小目标的处理仍不够理想等问题。
综上,隧道火灾检测技术经历了从传统机器学习方法向深度学习方法的转变,并逐步演化出针对性的模型优化和改进策略,但仍存在实时性和高精度之间仍难以达到平衡、对于早期火灾和微小火焰的检测上准确性不高的问题。为更好地满足实时预测火灾的发展态势并且具有高准确性的隧道火灾检测要求[14],笔者拟以YOLOv8s模型为基础进行改进,引入跨阶段部分变换器模块(Cross Stage Partial-Partially Transformer Block,CSP-PTB)结构,替代YOLOv8主干网络中部分C2f结构,对输入通道进行部分分配来优化计算效率和特征提取能力,并在主干网络添加CBAM提升神经网络的表征能力[16],同时,使用归一化高斯瓦瑟斯坦距离(Normalized Gaussian Wasserstein Distance,NWD)损失函数[17]提升目标检测的精度。
CSP-PTB结构增强全局特征提取能力和在复杂背景下小目标的识别效果如图1所示。首先通过1×1卷积将输入特征图,随后采用CSP结构将特征分为保持原始特征信息的直通分支和进行特征增强的转换分支,其中,变换分支包含n个串联的PTB模块。
在PTB模块中采用特征部分变换器化策略,即将输入特征在通道维度上分为变换器分支和CNN分支。变换器分支采用自适应特征增强模块(Adaptive Feature Enhancement Block,AFEB),如图2所示。其包含多头极化自注意力和卷积门控线性单元,通过二维层归一化和残差连接来保持特征的稳定性与有效性;CNN分支则采用经典的瓶颈层结构来提取局部特征。模块通过卷积融合对2个分支的特征进行自适应融合,并将所有分支(包括直通分支和n个PTB模块的输出)在通道维度上拼接,经过最后卷积整合输出。
将CBAM[16]集成到YOLOv8s模型中,以增强模型对重要特征的捕捉能力和其在复杂环境中的鲁棒性和稳定性。CBAM分为通道注意力和空间注意力2个子模块,其整体结构如图3所示。通道注意力模块采用降维-升维的瓶颈结构,在保持参数效率的同时实现通道间依赖关系的建模;空间注意力模块则通过7×7卷积核构建空间维度上的非局部交互关系,强化模型对目标几何结构的感知能力。
YOLOv8中的BL是用于训练目标检测模型的边界框回归损失函数,BL采用完全交并比(Complete Intersection over Union,CIoU)反映边界框之间的几何关系:
L C = 1 - I o U + d 2 e 2 + α υ
υ = 4 π 2 a r c t a n w g h g - a r c t a n w h 2
α = υ ( 1 - I o U ) + υ
式中:LC为CIoU计算函数;IoU为预测框与真实框的交并比(Intersection over Union,IoU);d为预测框和真实框中心点之间的欧氏距离;e为包围2个框的最小闭合框的对角线长度; α υ为调整参数,用来平衡宽高比;wh分别为预测框的宽和高;wghg分别为真实框的宽和高。
由于CIoU对隧道初始火灾小目标的检测效果不佳[18],且仅依赖于边框的几何信息没有考虑框内部的像素内容,易影响训练结果的精度。因此,使用NWD损失函数替代CIoU损失函数来增强鲁棒性,对边界框进行高斯分布建模,计算瓦瑟斯坦距离W并归一化,得到NWD值[17]。预测框的高斯分布G(μp,Σp)和真实框的高斯分布G(μt,Σt)可表示为:
μ = μ x   μ y Σ = σ x 2 ρ σ x σ y   ρ σ x σ y σ y 2
W 2 [ G ( μ p p ) G ( μ t t ) ] = μ p - μ t 2 + T r p + t - 2 (   t 1 2 p   t 1 2 ) 1 2
N W D [ G ( μ p p ) G ( μ t t ) ] =   e x p - W 2 [ G ( μ p p ) G ( μ t t ) ] C
式中:μ为均值向量;Σ为协方差矩阵;μxμy分别为在xy维度上的均值; σ x 2 σ y 2分别为xy的方差;ρxy之间的相关性系数,取值范围为[-1,1]; μ p - μ t 为2个均值向量间的欧氏距离;Tr为矩阵的迹;C为与数据集相关的常数。
改进后的YOLOv8s模型结构如图4所示。网络的主干部分使用标准的卷积层提取特征,通过C2f模块进一步优化特征融合,C2f模块通过分裂、瓶颈结构、拼接等操作提升特征的表达能力;采用CSP-PTB结构实现更高效的特征提取,并加入快速空间金字塔池化(Spatial Pyramid Pooling Fast,SPPF)模块和CBAM。网络的颈部和头部通过特征的拼接和上采样操作有效地融合不同尺度的信息,使目标检测的精度和定位能力加强。此外,通过多个检测模块和损失函数的设计,进一步优化隧道火焰和烟雾的检测和分类效果。
试验基于Windows 10操作系统,中央处理器为Intel i5-12400F@2.50 GHz,图形处理器为GeForce RTX 3090(24 GB),内存为16 GB。深度学习框架为PyTorch 2.3.1,Python 3.9.19,计算统一设备架构12.1,训练过程设置批次大小为16,轮次为200。
在网络上获取公开隧道火灾数据集图片1 439张,并在隧道内实地采集火焰和烟雾视频抽帧处理,得到1 775张图片,其中,昏暗灯光496张、烟雾遮挡2 508张、正常光照94张、无烟雾遮挡116张,共计3 214张图片作为数据集,并按照7︰2︰1的比例划分训练集、验证集和测试集。对训练集进行随机锐化、水平翻转、添加高斯噪声和对比度限制自适应直方图均衡(Contrast Limited Adaptive Histogram Equalization,CLAHE),增强局部对比度。验证集和测试集的主要作用是评估模型在未见数据上的性能,未对其进行数据增强。图5图6为训练集图片数据增强前后对比。
为准确评估模型对隧道环境下火焰和烟雾的检测效果,选择平均精度均值(mean Average Precision,mAP)、召回率R、模型计算量(Giga Floating-point Operations Per Second,GFLOPS)和每秒检测图片帧数(Frames Per Second,FPS)评价模型。
选择原始YOLOv8s模型作为基线模型,依次进行消融试验,包括引入CSP-PTB结构替代部分C2f模块、在主干网络中加入CBAM以及采用NWD损失函数,结果见表1
使用测试集中的325张图片开展消融试验,包括昏暗灯光46张、烟雾遮挡236张、正常光照12张、无烟雾遮挡23张。结果表明:与原始YOLOv8s模型相比,引入CSP-PTB结构替换部分C2f结构后,火焰的平均精度(APF)和烟雾的平均精度(APS)均显著提升,mAP提高1.2%。同时,模型的计算量从28.4减少到26.5,FPS也有所提升,模型的效率和响应速度显著提高。结合CBAM后,模型的总体平均精度提升1.7%,同时保持了模型的轻量化特性。此外,采用NWD损失函数后,模型的总体平均精度提升2%,火焰的召回率(RF)、烟雾的召回率(RS)和总体R均大幅提高,有效减少了隧道火灾漏报的情况。
进一步对比分析mAP、mAP@0.5:0.95、R、边界损失、感受野和热力图,在200轮次的训练过程中改进模型和原版模型的训练结果如图7所示。可以看出:改进后的模型在定位准确度、检测精度和全面性上均有大幅提升,且可使训练过程加速,并显著提高实际隧道火灾的检测效果。
感受野描述网络中某一层的神经元或过滤器能够接收到的输入图像区域的大小,较大的感受野能够捕捉到更大范围的信息,有助于识别图像中的大物体或全局特征。选取主干网络的最后1层以及80×80、40×40和20×20的检测头进行对比,可代表模型的特征提取能力以及对不同尺度目标的敏感性。改进后YOLOv8s和原版YOLOv8s的感受野分别如图8图9所示。可以看出,改进后的YOLOv8s模型显示出更大的覆盖范围且图像在细节方面更加清晰,增强了特征的捕捉能力,在提取深层次特征时表现得更加出色;在专门适用于小目标检测的80×80检测头中,改进YOLOv8s模型感受野明显更加集中,中心区域的激活强度更高,覆盖范围更合适,在提升模型的检测效率的同时避免对背景的过度关注,从而提升了小目标的检测准确率。改进YOLOv8s模型在其他尺度检测头中也展现了更加清晰、更细致的特征信息,显示出更强的噪声抑制能力,从而提升了检测结果的可靠性和稳定性。
在目标检测算法中,热力图通常用于表示目标在图像中的位置和置信度,颜色越深,表示模型对该区域属于目标的置信度越高。对比并分析原版YOLOv8s热力图(图10)和改进后的热力图(图11)可知:改进后的热力图在火灾区域的覆盖范围更大而且定位火灾的核心区域更准确,即可以捕捉到更多潜在的火灾区域并提供精确的火源位置,有助于早期检测和预防火灾的蔓延。同时,改进后的热力图在颜色分布上更加均匀,使得火灾区域更加突出,能够更快地识别和定位火灾,提升监控效率。可见:改进后的YOLOv8s模型在训练指标和特征可视化方面表现均优于原版YOLOv8s。
选取YOLOv5s、FR-CNN和SSD进行对比,使用测试集针对所有场景进行横向比较,结果见表2
SSD模型在各个指标上的表现相对较低,尤其在mAP@0.75和mAP@0.5:0.95上的表现(分别为0.563和0.476),显示出其在更严格的IoU阈值下效果不佳,仅在mAP达到0.762,召回率为0.551,表明其检测能力存在一定局限性。相比之下,YOLOv5s模型在所有指标上均表现较为出色,尤其是在mAP和mAP@0.75上分别达到0.805和0.601,显示其在不同IoU阈值下的检测能力较强,此外其在mAP@0.5:0.95为0.520,召回率达到0.618,整体性能优异。FR-CNN模型在mAP为0.782,略逊于YOLOv5s,但在mAP@0.75和mAP@0.5:0.95上的表现较弱,分别为0.534和0.478,显示出其在高IOU阈值下的性能不足。
改进后的YOLOv8s模型在各项指标上的表现都优于其他模型,在mAP、mAP@0.75和mAP@0.5:0.95上分别达到0.839、0.605和0.530,即该模型能够在不同的IOU阈值下保持高精度,能够更准确地检测到火灾和烟雾等关键目标。同时,召回率达到0.619,在检测火灾和烟雾时,能够识别出更多的真实目标,减少漏检的可能性,确保隧道火灾被及时发现。
1) 在YOLOv8s模型中引入CSP-PTB结构、加入CBAM,并使用NWD损失函数,改进后的YOLOv8s模型平均精度达到0.848,召回率达到0.812。与原版模型相比,平均精度提高2%,召回率提高9.3%,准确性和可靠性均有较大提高。
2) 改进后的YOLOv8s模型相较于原始版本,计算量减少6.7%,检测速度得到提升,实时检测性能更强。同时,加入CBAM可增强特征提取能力,提高对早期火灾和微小火焰的感知和识别,使模型在实际隧道火灾检测中更加精准和高效。
3) 改进后的YOLOv8s的mAP达到0.839,相较于SSD、YOLOv5s和F R-CNN,分别提升10.10%、4.20%和7.29%。同时,模型在mAP75和mAP50-95关键指标上分别达到0.605和0.530,召回率为0.619,均优于对比模型。
  • 国家自然科学基金资助(52474238)
参考文献 引证文献
排序方式:
[1]
李东平. 公路隧道的服役现状安全性评价管理体系构建[J]. 黑龙江交通科技, 2023, 46(3): 118-120.
LI Dongping. Construction of safety evaluation management system for service status of highway tunnels[J]. Communications Science and Technology Heilongjiang, 2023, 46(3): 118-120.
[2]
卢毅, 田芳, 陈建忠, 等. 公路隧道火灾灾情监测技术现状及展望[J]. 公路交通技术, 2021, 37(增1): 83-86, 92.
LU Yi, TIAN Fang, CHEN Jianzhong, et al. Studies and prospects of technologies for highway tunnel fire monitoring[J]. Technology of Highway and Transport, 2021, 37(S1): 83-86, 92.
[3]
朱学慧. 基于图像处理的公路隧道火灾检测技术研究[D]. 杭州: 中国计量大学, 2021.
ZHU Xuehui. Research on fire detection technology of highway tunnel based on image processing[D]. Hangzhou: China University of Metrology, 2021.
[4]
邓实强, 丁浩, 杨孟, 等. 基于视频图像的公路隧道火灾烟雾检测[J]. 隧道建设:中英文, 2022, 42(2): 291-302.
DENG Shiqiang, DING Hao, YANG Meng, et al. Fire smoke detection in highway tunnels based on video images[J]. Tunnel Construction, 2022, 42(2): 291-302.
[5]
吕宇, 张子俊. 红外热成像技术在隧道火灾检测中的研究与应用[J]. 中国设备工程, 2023(11): 155-157.
LYU Yu, ZHANG Zijun. Research and application of infrared thermal imaging technology in tunnel fire detection[J]. China Plant Engineering, 2023(11): 155-157.
[6]
LIU Wei, ANGUELOV D, ERHAN D, et al. SSD: single shot multibox detector[C]. Computer Vision-ECCV 2016: 14th European Conference, 2016: 21-37.
[7]
REN Shaoqing, HE Kaiming, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 39(6): 1137-1149.
[8]
HE Kaiming, GKIOXARI G, DOLLAR P, et al. Mask R-CNN[C]. Proceedings of the IEEE International Conference on Computer Vision, 2017: 2961-2969.
[9]
TAN Ming, LE Qing. Efficientnet: rethinking model scaling for convolutional neural networks[C]. International Conference on Machine Learning, 2019: 6105-6114.
[10]
唐小垚. 基于深度学习的公路隧道火灾检测研究[D]. 重庆: 重庆交通大学, 2023.
TANG Xiaoyao. Research on highway tunnel fire detection based on deep learning[D]. Chongqing: Chongqing Jiaotong University, 2023.
[11]
马庆禄, 孙枭, 唐小垚, 等. 基于改进YOLOv5s算法的隧道初期火灾检测模型[J]. 中国安全科学学报, 2023, 33(10): 214-223.
MA Qinglu, SUN Xiao, TANG Xiaoyao, et al. Optimization method for tunnel initial fire detection based on YOLOv5s algorithm[J]. China Safety Science Journal, 2023, 33(10): 214-223.
[12]
张晋瑞, 宋焕生, 孙士杰, 等. 改进YOLOv5的隧道火灾帧差检测网络与应用方法[J]. 计算机工程与应用, 2023, 59(2): 222-231.
ZHANG Jinrui, SONG Huansheng, SUN Shijie, et al. Tunnel fire frame difference detection network by improving YOLOv5 and application method[J]. Computer Engineering and Applications, 2023, 59(2): 222-231.
[13]
宋焕生, 文雅, 孙士杰, 等. 基于改进教师学生网络的隧道火灾检测[J]. 图学学报, 2023, 44(5): 978-987.
SONG Huansheng, WEN Ya, SUN Shijie, et al. Tunnel fire detection based on improved student-teacher network[J]. Journal of Graphics, 2023, 44(5): 978-987.
[14]
马庆禄, 马恋, 孔国英, 等. 基于红外热成像的隧道火焰检测技术研究[J]. 火灾科学, 2022, 31(4): 244-251.
MA Qinglu, MA Lian, KONG Guoying, et al. Research on tunnel flame detection technology based on infrared thermal imaging[J]. Fire Safety Science, 2022, 31(4): 244-251.
[15]
VARGHESE R, SAMBATH M. YOLOv8: a novel object detection algorithm with enhanced performance and robustness[C]. 2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS), 2024: 1-6.
[16]
WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C]. Proceedings of the European Conference on Computer Vision (ECCV), 2018: 3-19.
[17]
WANG Jinwang, XU Chang, YANG Wen, et al. A normalized gaussian wasserstein distance for tiny object detection[J]. arXiv Preprint, 2022: DOI: 10.48550/arXiv.2110.13389.
[18]
李耀, 徐红伟, 柯海森, 等. 基于改进YOLOv8n的织物疵点检测[J]. 棉纺织技术, 2024, 52(10): 11-18.
LI Yao, XU Hongwei, KE Haisen, et al. Fabric defect detection based on improved YOLOv8n[J]. Cotton Textile Technology, 2024, 52(10): 11-18.
2025年第35卷第3期
PDF下载
435
185
引用本文
BibTeX
文章信息
doi: 10.16265/j.cnki.issn1003-3033.2025.03.1181
  • 接收时间:2024-10-22
  • 首发时间:2025-07-05
  • 出版时间:2025-03-28
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-10-22
  • 修回日期:2024-12-24
基金
国家自然科学基金资助(52474238)
作者信息
    青岛理工大学 机械与汽车工程学院,山东 青岛 266520

通讯作者:

** 刘权捷(2000—),男,辽宁抚顺人,硕士研究生,研究方向为智能交通安全及目标检测跟踪算法。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN/10.16265/j.cnki.issn1003-3033.2025.03.1181
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