Article(id=1149733270813851754, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1149733267617788430, articleNumber=1003-3033(2024)12-0178-09, orderNo=null, doi=10.16265/j.cnki.issn1003-3033.2024.12.0465, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1723824000000, receivedDateStr=2024-08-17, revisedDate=1729008000000, revisedDateStr=2024-10-16, acceptedDate=null, acceptedDateStr=null, onlineDate=1752047372773, onlineDateStr=2025-07-09, pubDate=1735315200000, pubDateStr=2024-12-28, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1752047372773, onlineIssueDateStr=2025-07-09, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1752047372773, creator=13701087609, updateTime=1752047372773, updator=13701087609, issue=Issue{id=1149733267617788430, tenantId=1146029695717560320, journalId=1146031787341344770, year='2024', volume='34', issue='12', pageStart='1', pageEnd='228', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1752047372010, creator=13701087609, updateTime=1756361981736, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1167830052499628941, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1149733267617788430, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1167830052499628942, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1149733267617788430, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=178, endPage=186, ext={EN=ArticleExt(id=1149733271027761260, articleId=1149733270813851754, tenantId=1146029695717560320, journalId=1146031787341344770, language=EN, title=Urban spatially mixed traffic participants detection model based on improved YOLOv8n, columnId=1149733270084042840, journalTitle=China Safety Science Journal, columnName=Public safety, runingTitle=null, highlight=null, articleAbstract=

In order to improve the recognition accuracy and detection speed of traffic participants by intelligent networked vehicles and traffic monitoring systems so that they can timely respond to the safety hazards in the mixed traffic environment in urban space,a mixed traffic participant detection model in urban space based on the improved YOLOv8n algorithm was proposed. Firstly,geometric transformation and pixel transformation enhancement strategies were employed in the data input stage to prevent overfitting and improve robustness,and generalization. Secondly,the SPD-Conv module was used to replace all original convolution layers of the YOLOv8n algorithm,which enhances the feature extraction capability for low-resolution small targets. Meanwhile,the CA module was added to the fusion structure of the neck network of the YOLOv8n algorithm to improve the recognition accuracy of key information with almost no additional computational overhead. Then,the boundary box loss function EIoU was used to replace the original loss function,enabling the model to achieve superior convergence speed and recognition stability. Finally,the ablation and comparison experiments were carried out with the public and self-built integrated traffic participant dataset,and the real-time detection experiment was carried out with the automatic driving experiment platform. The experimental results show that compared to the YOLOv8n model,the improved SEC-YOLO model has increased mAP and FPS by 3.2% and 7.9% respectively. The SEC-YOLO model outperforms mainstream models in terms of mAP and FPS as well. The average accuracy of real-scene detection on the automatic driving experimental platform is around 95%. The SEC-YOLO algorithm model achieves higher detection accuracy for urban traffic participants,with stronger robustness and real-time performance.

, correspAuthors=Jianjie GAO, 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=Junchao ZHOU, Xin CHEN, Jianjie GAO, Jie ZHANG), CN=ArticleExt(id=1149733287234552601, articleId=1149733270813851754, tenantId=1146029695717560320, journalId=1146031787341344770, language=CN, title=基于改进YOLOv8n算法的城市空间混行交通参与者检测模型, columnId=1149733271510106222, journalTitle=中国安全科学学报, columnName=公共安全, runingTitle=null, highlight=null, articleAbstract=

为提升智能网联汽车和交通监测系统对交通参与者的识别精度和检测速度,及时应对城市空间混行交通环境中的安全隐患,提出一种基于改进YOLOv8n算法的城市空间混行交通参与者检测模型。首先,在数据输入阶段利用几何变换和像素变换策略来防止过度拟合,提高鲁棒性和泛化性;其次,使用空间到深度的非跨行卷积(SPD-Conv)模块代替所有原始卷积层,提高对低分辨率小目标的特征提取能力;同时,在颈部网络融合结构中加入轻量级坐标注意力机制(CA)模块,在几乎不添加任何计算开销的同时提高对关键信息的识别精度;然后,用边界框损失函数有效交并比(EIoU)替代原损失函数,使模型获得更卓越的收敛速度与识别稳定性;最后,利用平台公开和自建融合的交通参与者数据集(BNS)进行消融和对比试验,运用自动驾驶试验平台进行实景实时检测。结果表明:相比于基线模型YOLOv8n,改进模型SEC-YOLO的每秒传输帧数(FPS)和平均精度均值(mAP)分别提高了7.3%和3.2%;与主流模型对比,mAP和FPS性能值最佳;在自动驾驶试验平台上的实景检测平均准确率为95%。SEC-YOLO算法模型对城市交通参与者的检测准确率更高,鲁棒性和实时性更强。

, correspAuthors=高建杰, authorNote=null, correspAuthorsNote=
**高建杰(1985—),男,山东栖霞人,博士,副教授,主要从事城市交通安全管理、城市交通风险判别与检测等方面的研究。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=QnM/I8AqHjuhjth1y5b0qg==, magXml=oXMcFh9iHZEN0XhBu6HsCA==, pdfUrl=null, pdf=6ivtzLQyL9qToGznvLy3Hw==, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=NpF4odtFSMtw3M1y2+Pv+Q==, mapNumber=null, authorCompany=null, fund=null, authors=

周军超 (1987—),男,湖北襄阳人,博士,副教授,主要从事城市智能交通安全检测、智能网联汽车安全驾驶与风险识别、新型城市轨道车辆安全与控制等方面的研究。E-mail:

章杰,副教授。

, authorsList=周军超, 陈鑫, 高建杰, 章杰)}, authors=[Author(id=1167743189298458843, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=zhou1987g@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1167743189369762014, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, authorId=1167743189298458843, language=EN, stringName=Junchao ZHOU, firstName=Junchao, middleName=null, lastName=ZHOU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 School of Mechanical Engineering,Sichuan University of Science & Engineering,Zigong Sichuan 643000,China
2 Chengdu-Chongqing Economic Circle (Luzhou) Advanced Technology Research Institute,Luzhou Sichuan 646000,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1167743189428482271, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, authorId=1167743189298458843, language=CN, stringName=周军超, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 四川轻化工大学 机械工程学院,四川 自贡 643000
2 成渝地区双城经济圈(泸州)先进技术研究院,四川 泸州 646000, bio={"img":"pbDMIE6Gf35uZ2YWoSdi6w==","content":"

周军超 (1987—),男,湖北襄阳人,博士,副教授,主要从事城市智能交通安全检测、智能网联汽车安全驾驶与风险识别、新型城市轨道车辆安全与控制等方面的研究。E-mail:

"}, bioImg=pbDMIE6Gf35uZ2YWoSdi6w==, bioContent=

周军超 (1987—),男,湖北襄阳人,博士,副教授,主要从事城市智能交通安全检测、智能网联汽车安全驾驶与风险识别、新型城市轨道车辆安全与控制等方面的研究。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1167743188958720206, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, xref=1, ext=[AuthorCompanyExt(id=1167743188992274639, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743188958720206, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Mechanical Engineering,Sichuan University of Science & Engineering,Zigong Sichuan 643000,China), AuthorCompanyExt(id=1167743188996468944, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743188958720206, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 四川轻化工大学 机械工程学院,四川 自贡 643000)]), AuthorCompany(id=1167743189055189201, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, xref=2, ext=[AuthorCompanyExt(id=1167743189059383506, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743189055189201, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Chengdu-Chongqing Economic Circle (Luzhou) Advanced Technology Research Institute,Luzhou Sichuan 646000,China), AuthorCompanyExt(id=1167743189067772115, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743189055189201, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 成渝地区双城经济圈(泸州)先进技术研究院,四川 泸州 646000)])]), Author(id=1167743189491396833, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, 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=1167743189579477219, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, authorId=1167743189491396833, language=EN, stringName=Xin CHEN, firstName=Xin, middleName=null, lastName=CHEN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 School of Mechanical Engineering,Sichuan University of Science & Engineering,Zigong Sichuan 643000,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1167743189634003172, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, authorId=1167743189491396833, language=CN, stringName=陈鑫, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 四川轻化工大学 机械工程学院,四川 自贡 643000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1167743188958720206, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, xref=1, ext=[AuthorCompanyExt(id=1167743188992274639, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743188958720206, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Mechanical Engineering,Sichuan University of Science & Engineering,Zigong Sichuan 643000,China), AuthorCompanyExt(id=1167743188996468944, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743188958720206, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 四川轻化工大学 机械工程学院,四川 自贡 643000)])]), Author(id=1167743189797581030, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=jianjiecq@163.com, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1167743192679067880, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, authorId=1167743189797581030, language=EN, stringName=Jianjie GAO, firstName=Jianjie, middleName=null, lastName=GAO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, **, address=3 Intelligent Policing Key Laboratory of Sichuan Province,Sichuan Police College,Luzhou Sichuan 646000,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1167743192741982442, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, authorId=1167743189797581030, language=CN, stringName=高建杰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, **, address=3 四川警察学院 智能警务四川省重点实验室,四川 泸州 646000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1167743189130686676, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, xref=3, ext=[AuthorCompanyExt(id=1167743189139075285, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743189130686676, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 Intelligent Policing Key Laboratory of Sichuan Province,Sichuan Police College,Luzhou Sichuan 646000,China), AuthorCompanyExt(id=1167743189143269590, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743189130686676, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 四川警察学院 智能警务四川省重点实验室,四川 泸州 646000)])]), Author(id=1167743192804897004, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1167743192872005871, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, authorId=1167743192804897004, language=EN, stringName=Jie ZHANG, firstName=Jie, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, 4, address=3 Intelligent Policing Key Laboratory of Sichuan Province,Sichuan Police College,Luzhou Sichuan 646000,China
4 College of Automotive and Mechanical Engineering,Changsha University of Science & Technology,Changsha Hunan 410114,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1167743192926531824, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, authorId=1167743192804897004, language=CN, stringName=章杰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, 4, address=3 四川警察学院 智能警务四川省重点实验室,四川 泸州 646000
4 长沙理工大学 汽车与机械工程学院,湖南 长沙 410114, bio={"content":"

章杰,副教授。

"}, bioImg=null, bioContent=

章杰,副教授。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1167743189130686676, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, xref=3, ext=[AuthorCompanyExt(id=1167743189139075285, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743189130686676, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 Intelligent Policing Key Laboratory of Sichuan Province,Sichuan Police College,Luzhou Sichuan 646000,China), AuthorCompanyExt(id=1167743189143269590, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743189130686676, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 四川警察学院 智能警务四川省重点实验室,四川 泸州 646000)]), AuthorCompany(id=1167743189222961367, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, xref=4, ext=[AuthorCompanyExt(id=1167743189231349976, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743189222961367, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4 College of Automotive and Mechanical Engineering,Changsha University of Science & Technology,Changsha Hunan 410114,China), AuthorCompanyExt(id=1167743189235544281, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743189222961367, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4 长沙理工大学 汽车与机械工程学院,湖南 长沙 410114)])])], keywords=[Keyword(id=1167743193035583729, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, orderNo=1, keyword=YOLOv8n), Keyword(id=1167743193111081202, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, orderNo=2, keyword=spatial mixing), Keyword(id=1167743193173995763, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, orderNo=3, keyword=traffic participants), Keyword(id=1167743193236910324, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, orderNo=4, keyword=detection model), Keyword(id=1167743193295630581, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, orderNo=5, keyword=space-to-depth convolution (SPD-Conv)), Keyword(id=1167743193354350838, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, orderNo=6, keyword=coordinate attention (CA)), Keyword(id=1167743193413071095, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, orderNo=1, keyword=YOLOv8n), Keyword(id=1167743193471791352, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, orderNo=2, keyword=空间混行), Keyword(id=1167743193530511609, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, orderNo=3, keyword=交通参与者), Keyword(id=1167743193585037562, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, orderNo=4, keyword=检测模型), Keyword(id=1167743193643757819, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, orderNo=5, keyword=空间到深度的非跨行卷积(SPD-Conv)), Keyword(id=1167743193694089468, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, orderNo=6, keyword=坐标注意力机制(CA))], refs=[Reference(id=1167743197854839071, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2023, volume=33, issue=11, pageStart=133, pageEnd=141, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=刘晓芳, 吝涛, 于琪, journalName=中国安全科学学报, refType=null, unstructuredReference=刘晓芳, 吝涛, 于琪, 等. 城镇街巷空间人群通行疏散及碰撞风险定量模拟[J]. 中国安全科学学报, 2023, 33(11):133-141., articleTitle=城镇街巷空间人群通行疏散及碰撞风险定量模拟, refAbstract=null), Reference(id=1167743197905170720, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2023, volume=33, issue=11, pageStart=133, pageEnd=141, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=LIU Xiaofang, LIN Tao, YU Qi, journalName=China Safety Science Journal, refType=null, unstructuredReference=LIU Xiaofang, LIN Tao, YU Qi, et al. Quantitative simulation of pedestrian evacuation and collision risk in urban street space[J]. China Safety Science Journal, 2023, 33(11):133-141., articleTitle=Quantitative simulation of pedestrian evacuation and collision risk in urban street space, refAbstract=null), Reference(id=1167743197976473889, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2023, volume=33, issue=11, pageStart=75, pageEnd=81, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=张军, 尹柳, 巩欣飞, journalName=中国安全科学学报, refType=null, unstructuredReference=张军, 尹柳, 巩欣飞, 等. 基于改进YOLOv5的小目标烟雾检测算法[J]. 中国安全科学学报, 2023, 33(11):75-81., articleTitle=基于改进YOLOv5的小目标烟雾检测算法, refAbstract=null), Reference(id=1167743198035194146, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2023, volume=33, issue=11, pageStart=75, pageEnd=81, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=ZHANG Jun, YIN Liu, GONG Xinfei, journalName=China Safety Science Journal, refType=null, unstructuredReference=ZHANG Jun, YIN Liu, GONG Xinfei, et al. Small target smoke detection algorithm based on improved YOLOv5[J]. China Safety Science Journal, 2023, 33(11):75-81., articleTitle=Small target smoke detection algorithm based on improved YOLOv5, refAbstract=null), Reference(id=1167743198089720099, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=779, pageEnd=788, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=REDMON J, DIVVALA S, GIRSHICK R, journalName=Proceedings of the IEEE Conference on Computer Cision and Cattern Recognition, refType=null, unstructuredReference=REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]. Proceedings of the IEEE Conference on Computer Cision and Cattern Recognition, 2016: 779-788., articleTitle=You only look once: unified, real-time object detection, refAbstract=null), Reference(id=1167743198144246052, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=2980, pageEnd=2988, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=LIN Tianyi, GOYAL P, GIRSHICK R, journalName=Proceedings of the IEEE International Conference on Computer Vision, refType=null, unstructuredReference=LIN Tianyi, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[C]. Proceedings of the IEEE International Conference on Computer Vision, 2017: 2980-2988., articleTitle=Focal loss for dense object detection, refAbstract=null), Reference(id=1167743198207160613, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=21, pageEnd=37, url=null, language=null, rfNumber=[5], rfOrder=6, authorNames=LIU Wu, ANGUELOV D, ERHAN D, journalName=Proceedings of the 2016 European Conference on Computer Vision, refType=null, unstructuredReference=LIU Wu, ANGUELOV D, ERHAN D, et al. SSD: single shot multibox detector[C]. Proceedings of the 2016 European Conference on Computer Vision, 2016: 21-37., articleTitle=SSD: single shot multibox detector, refAbstract=null), Reference(id=1167743198265880870, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2023, volume=33, issue=2, pageStart=82, pageEnd=88, url=null, language=null, rfNumber=[6], rfOrder=7, authorNames=夏正洪, 魏汝祥, 李彦冬, journalName=中国安全科学学报, refType=null, unstructuredReference=夏正洪, 魏汝祥, 李彦冬. 改进 YOLOv3 算法下通航机场场面运动目标检测[J]. 中国安全科学学报, 2023, 33(2):82-88., articleTitle=改进 YOLOv3 算法下通航机场场面运动目标检测, refAbstract=null), Reference(id=1167743198337184039, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2023, volume=33, issue=2, pageStart=82, pageEnd=88, url=null, language=null, rfNumber=[6], rfOrder=8, authorNames=XIA Zhenghong, WEI Ruxiang, LI Yandong, journalName=China Safety Science Journal, refType=null, unstructuredReference=XIA Zhenghong, WEI Ruxiang, LI Yandong. Moving target detection of general aviation airport based on improved YOLOv3 algorithm[J]. China Safety Science Journal, 2023, 33(2):82-88., articleTitle=Moving target detection of general aviation airport based on improved YOLOv3 algorithm, refAbstract=null), Reference(id=1167743198387515688, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2023, volume=105, issue=null, pageStart=84, pageEnd=90, url=null, language=null, rfNumber=[7], rfOrder=9, authorNames=SHU Jijun, LING Qinghua, HAN Fei, journalName=Computers and Electrical Engineering, refType=null, unstructuredReference=SHU Jijun, LING Qinghua, HAN Fei. An improved algorithm for small object detection based on YOLO v4 and multi-scale contextual information[J]. Computers and Electrical Engineering, 2023, 105:84-90., articleTitle=An improved algorithm for small object detection based on YOLO v4 and multi-scale contextual information, refAbstract=null), Reference(id=1167743198450430249, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2023, volume=11, issue=null, pageStart=951, pageEnd=57, url=null, language=null, rfNumber=[8], rfOrder=10, authorNames=WANG Shouhe, WANG Yongda, CHANG Yujian, journalName=IEEE Access, refType=null, unstructuredReference=WANG Shouhe, WANG Yongda, CHANG Yujian, et al. EBSE-YOLO: high precision recognition algorithm for small target foreign object detection[J]. IEEE Access, 2023, 11:57951-57 964., articleTitle=EBSE-YOLO: high precision recognition algorithm for small target foreign object detection, refAbstract=null), Reference(id=1167743198500761898, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2024, volume=25, issue=1, pageStart=189, pageEnd=202, url=null, language=null, rfNumber=[9], rfOrder=11, authorNames=ZHOU Wei, WANG Chuan, XIA Jie, journalName=IEEE Transactions on Intelligent Transportation Systems, refType=null, unstructuredReference=ZHOU Wei, WANG Chuan, XIA Jie, et al. Monitoring-based traffic participant detection in urban mixed traffic: a novel dataset and a tailored detector[J]. IEEE Transactions on Intelligent Transportation Systems, 2024, 25(1):189-202., articleTitle=Monitoring-based traffic participant detection in urban mixed traffic: a novel dataset and a tailored detector, refAbstract=null), Reference(id=1167743198601425195, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2024, volume=46, issue=6, pageStart=1050, pageEnd=1062, url=null, language=null, rfNumber=[10], rfOrder=12, authorNames=王泽宇, 徐慧英, 朱信忠, journalName=计算机工程与科学, refType=null, unstructuredReference=王泽宇, 徐慧英, 朱信忠, 等. 基于YOLOv8改进的密集行人检测算法:MER-YOLO[J]. 计算机工程与科学, 2024, 46(6):1050-1062., articleTitle=基于YOLOv8改进的密集行人检测算法:MER-YOLO, refAbstract=null), Reference(id=1167743198668534060, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2024, volume=46, issue=6, pageStart=1050, pageEnd=1062, url=null, language=null, rfNumber=[10], rfOrder=13, authorNames=WANG Zeyu, XU Huiying, ZHU Xinzhong, journalName=Computer Engineering and Science, refType=null, unstructuredReference=WANG Zeyu, XU Huiying, ZHU Xinzhong, et al. An improved dense pedestrian detection algorithm based on YOLOv8: MER-YOLO[J]. Computer Engineering and Science, 2024, 46(6):1050-1062., articleTitle=An improved dense pedestrian detection algorithm based on YOLOv8: MER-YOLO, refAbstract=null), Reference(id=1167743198756614445, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=1, pageEnd=5, url=null, language=null, rfNumber=[11], rfOrder=14, authorNames=OUYANG Daliang, HE Su, ZHANG Guozhong, journalName=IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), refType=null, unstructuredReference=OUYANG Daliang, HE Su, ZHANG Guozhong, et al. Efficient multi-scale attention module with cross-spatial learning[C]. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023: 1-5., articleTitle=Efficient multi-scale attention module with cross-spatial learning, refAbstract=null), Reference(id=1167743198815334702, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2024, volume=17, issue=null, pageStart=1734, pageEnd=1747, url=null, language=null, rfNumber=[12], rfOrder=15, authorNames=YI Hao, LIU Bo, ZHAO Bin, journalName=IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, refType=null, unstructuredReference=YI Hao, LIU Bo, ZHAO Bin, et al. Small object detection algorithm based on improved YOLOv8 for remote sensing[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2024, 17:1734-1747., articleTitle=Small object detection algorithm based on improved YOLOv8 for remote sensing, refAbstract=null), Reference(id=1167743198861472047, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=2636, pageEnd=2645, url=null, language=null, rfNumber=[13], rfOrder=16, authorNames=YU Fu, CHEN Hui, WANG Xia, journalName=Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, refType=null, unstructuredReference=YU Fu, CHEN Hui, WANG Xia, et al. BDD100K: a diverse driving dataset for heterogeneous multitask learning[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2020: 2636-2645., articleTitle=BDD100K: a diverse driving dataset for heterogeneous multitask learning, refAbstract=null), Reference(id=1167743198915998000, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=93, pageEnd=100, url=null, language=null, rfNumber=[14], rfOrder=17, authorNames=CATAK O, TAN Yue, ALI S, journalName=IEEE International Conference on Artificial Intelligence Testing, refType=null, unstructuredReference=CATAK O, TAN Yue, ALI S. Prediction surface uncertainty quantification in object detection models for autonomous driving[C]. IEEE International Conference on Artificial Intelligence Testing, 2021: 93-100., articleTitle=Prediction surface uncertainty quantification in object detection models for autonomous driving, refAbstract=null), Reference(id=1167743198962135345, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=443, pageEnd=459, url=null, language=null, rfNumber=[15], rfOrder=18, authorNames=SUNKARA R, LUO Tao, journalName=Joint European Conference on Machine Learning and Knowledge Discovery in Databases, refType=null, unstructuredReference=SUNKARA R, LUO Tao. No more strided convolutions or pooling: a new CNN building block for low-resolution images and small objects[C]. Joint European Conference on Machine Learning and Knowledge Discovery in Databases, 2022: 443-459., articleTitle=No more strided convolutions or pooling: a new CNN building block for low-resolution images and small objects, refAbstract=null), Reference(id=1167743199012466994, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=713, pageEnd=13, url=null, language=null, rfNumber=[16], rfOrder=19, authorNames=HOU Qibin, ZHOU Daquan, FENG Jiashi, journalName=Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, refType=null, unstructuredReference=HOU Qibin, ZHOU Daquan, FENG Jiashi. Coordinate attention for efficient mobile network design[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2021: 13 713-13 722., articleTitle=Coordinate attention for efficient mobile network design, refAbstract=null), Reference(id=1167743199062798643, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, doi=null, pmid=null, pmcid=null, year=2021, volume=1924, issue=1, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=20, authorNames=YANG Zuomin, WANG Xialun, LI Jiang, journalName=Journal of Physics Conference Series, refType=null, unstructuredReference=YANG Zuomin, WANG Xialun, LI Jiang. EIoU: an improved vehicle detection algorithm based on vehiclenet neural network[J]. Journal of Physics Conference Series, 2021, 1924(1):DOI:10.1088/1742-6596/1924/1/012001., articleTitle=EIoU: an improved vehicle detection algorithm based on vehiclenet neural network, refAbstract=null)], funds=[Fund(id=1167743197628346652, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, awardId=ZNJW2023KFQN005, language=CN, fundingSource=智能警务四川省重点实验室开放课题资助(ZNJW2023KFQN005), fundOrder=null, country=null), Fund(id=1167743197687066909, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, awardId=ZNJW2022KFQN004, language=CN, fundingSource=智能警务四川省重点实验室开放课题资助(ZNJW2022KFQN004), fundOrder=null, country=null), Fund(id=1167743197741592862, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, awardId=2023JYJ066, language=CN, fundingSource=泸州市科技计划资助(2023JYJ066), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1167743188958720206, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, xref=1, ext=[AuthorCompanyExt(id=1167743188992274639, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743188958720206, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Mechanical Engineering,Sichuan University of Science & Engineering,Zigong Sichuan 643000,China), AuthorCompanyExt(id=1167743188996468944, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743188958720206, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 四川轻化工大学 机械工程学院,四川 自贡 643000)]), AuthorCompany(id=1167743189055189201, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, xref=2, ext=[AuthorCompanyExt(id=1167743189059383506, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743189055189201, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Chengdu-Chongqing Economic Circle (Luzhou) Advanced Technology Research Institute,Luzhou Sichuan 646000,China), AuthorCompanyExt(id=1167743189067772115, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743189055189201, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 成渝地区双城经济圈(泸州)先进技术研究院,四川 泸州 646000)]), AuthorCompany(id=1167743189130686676, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, xref=3, ext=[AuthorCompanyExt(id=1167743189139075285, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743189130686676, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 Intelligent Policing Key Laboratory of Sichuan Province,Sichuan Police College,Luzhou Sichuan 646000,China), AuthorCompanyExt(id=1167743189143269590, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743189130686676, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 四川警察学院 智能警务四川省重点实验室,四川 泸州 646000)]), AuthorCompany(id=1167743189222961367, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, xref=4, ext=[AuthorCompanyExt(id=1167743189231349976, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743189222961367, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4 College of Automotive and Mechanical Engineering,Changsha University of Science & Technology,Changsha Hunan 410114,China), AuthorCompanyExt(id=1167743189235544281, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, companyId=1167743189222961367, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4 长沙理工大学 汽车与机械工程学院,湖南 长沙 410114)])], figs=[ArticleFig(id=1167743193878638845, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, label=Fig.1, caption=Partially collected data set image, figureFileSmall=whqAjDtiYz+DwxapoCa9pg==, figureFileBig=vy61cF/nuocvwjSpie78nQ==, tableContent=null), ArticleFig(id=1167743193949942014, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, label=图1, caption=部分采集的数据集图像, figureFileSmall=whqAjDtiYz+DwxapoCa9pg==, figureFileBig=vy61cF/nuocvwjSpie78nQ==, tableContent=null), ArticleFig(id=1167743194008662271, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, label=Fig.2, caption=Part of the randomly introduced data enhancement effect comparison chart, figureFileSmall=zUXn20VWoLVdLd2jTbYuUg==, figureFileBig=rE4WAcWRhLv5CrK6X3g5Sw==, tableContent=null), ArticleFig(id=1167743194067382528, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, label=图2, caption=部分随机引入的数据增强效果对比, figureFileSmall=zUXn20VWoLVdLd2jTbYuUg==, figureFileBig=rE4WAcWRhLv5CrK6X3g5Sw==, tableContent=null), ArticleFig(id=1167743194126102785, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, label=Fig.3, caption=SEC-YOLO network structure, figureFileSmall=xaFVS0Psj97XXnTZKS/npw==, figureFileBig=mQPr6SwJjZM4HXlbVNb1dg==, tableContent=null), ArticleFig(id=1167743194184823042, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, label=图3, caption=SEC-YOLO网络结构

注:2个跨阶段卷积层(Cross-Stage Partial Layer_2Conv,CSPLayer_2Conv);快速空间金字塔池化(Spatial Pyramid Pooling Fast,SPPF)。

, figureFileSmall=xaFVS0Psj97XXnTZKS/npw==, figureFileBig=mQPr6SwJjZM4HXlbVNb1dg==, tableContent=null), ArticleFig(id=1167743194235154691, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, label=Fig.4, caption=Example of SPD-Conv down sampling, figureFileSmall=kuFHStu0TJlHG7qhr7mAgA==, figureFileBig=BNbzGcT7JxjQ0FTXiI1S/w==, tableContent=null), ArticleFig(id=1167743194285486340, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, label=图4, caption=SPD-Conv下采样示例, figureFileSmall=kuFHStu0TJlHG7qhr7mAgA==, figureFileBig=BNbzGcT7JxjQ0FTXiI1S/w==, tableContent=null), ArticleFig(id=1167743194344206597, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, label=Fig.5, caption=CA structure schematic diagram, figureFileSmall=gXd5W2BpIWVgdaYvbadgkQ==, figureFileBig=+0aVXhKM0W0pNNtKZ2mJnw==, tableContent=null), ArticleFig(id=1167743194415509766, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, label=图5, caption=CA结构原理, figureFileSmall=gXd5W2BpIWVgdaYvbadgkQ==, figureFileBig=+0aVXhKM0W0pNNtKZ2mJnw==, tableContent=null), ArticleFig(id=1167743194482618631, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, label=Fig.6, caption=CIoU loss function bounded box regression plot, figureFileSmall=wBGnV2CxrUiIBqH3edZ+nw==, figureFileBig=AjfRtGhnm1z15XmtKn7IRQ==, tableContent=null), ArticleFig(id=1167743194537144584, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, label=图6, caption=CIoU损失函数有界盒回归图, figureFileSmall=wBGnV2CxrUiIBqH3edZ+nw==, figureFileBig=AjfRtGhnm1z15XmtKn7IRQ==, tableContent=null), ArticleFig(id=1167743194591670537, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, label=Fig.7, caption=Model performance comparison chart, figureFileSmall=Rg2f+QvJ3n2HDN8przjDJQ==, figureFileBig=BCxAsPXKX42pkHigFYGTTw==, tableContent=null), ArticleFig(id=1167743194650390794, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, label=图7, caption=SEC-YOLO与YOLOv8n模型性能对比, figureFileSmall=Rg2f+QvJ3n2HDN8przjDJQ==, figureFileBig=BCxAsPXKX42pkHigFYGTTw==, tableContent=null), ArticleFig(id=1167743194709111051, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, label=Fig.8, caption=Comparison of model results, figureFileSmall=siD8sROiQtcn0Ah/td4gRg==, figureFileBig=nwUnz+a9TQFjR8oAL7YuZg==, tableContent=null), ArticleFig(id=1167743194767831309, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, label=图8, caption=模型结果对比, figureFileSmall=siD8sROiQtcn0Ah/td4gRg==, figureFileBig=nwUnz+a9TQFjR8oAL7YuZg==, tableContent=null), ArticleFig(id=1167743196776902926, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, label=Fig.9, caption=Comparison diagram of partial model checking, figureFileSmall=MDkGe4Gf44J0OdkfsjrEhw==, figureFileBig=wZCQ3nkFHkNu9bvJJwPjbw==, tableContent=null), ArticleFig(id=1167743196852400399, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, label=图9, caption=部分模型检测对比, figureFileSmall=MDkGe4Gf44J0OdkfsjrEhw==, figureFileBig=wZCQ3nkFHkNu9bvJJwPjbw==, tableContent=null), ArticleFig(id=1167743196906926352, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, label=Fig.10, caption=Part of the automatic driving experiment real-time detection result diagram, figureFileSmall=17k7QV4c14mCNnnI632DMg==, figureFileBig=sNVgUnLQUJGzdASh7IrHEw==, tableContent=null), ArticleFig(id=1167743196965646610, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, label=图10, caption=部分自动驾驶试验实时检测结果, figureFileSmall=17k7QV4c14mCNnnI632DMg==, figureFileBig=sNVgUnLQUJGzdASh7IrHEw==, tableContent=null), ArticleFig(id=1167743197032755476, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, label=Fig.11, caption=Comparison between the number of traffic participants tested and the actual number of traffic participants, figureFileSmall=c2feibdYfjeoZXKwz4jokA==, figureFileBig=QM7osY1+t3RFF0bqfYWx9g==, tableContent=null), ArticleFig(id=1167743197083087125, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, label=图11, caption=检测交通参与者与实际交通参与者数结果对比, figureFileSmall=c2feibdYfjeoZXKwz4jokA==, figureFileBig=QM7osY1+t3RFF0bqfYWx9g==, tableContent=null), ArticleFig(id=1167743197133418774, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, label=Table 1, caption=

Experimental environment configuration

, figureFileSmall=null, figureFileBig=null, tableContent=
软件 硬件
操作系统 Windows10 CPU 12代酷睿i5-12490F
模型构建 PyTorch 显卡 Nvidia GeForce RTX3060 Ti
CUDA版本 11.3 内存 32GB
编译语言 Python
开发平台 PyCharm
), ArticleFig(id=1167743197187944727, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, label=表1, caption=

试验环境配置

, figureFileSmall=null, figureFileBig=null, tableContent=
软件 硬件
操作系统 Windows10 CPU 12代酷睿i5-12490F
模型构建 PyTorch 显卡 Nvidia GeForce RTX3060 Ti
CUDA版本 11.3 内存 32GB
编译语言 Python
开发平台 PyCharm
), ArticleFig(id=1167743197271830808, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, label=Table 2, caption=

Ablation experiment results

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 SPD-
Conv
CA EIoU P/% R/% mAP@
0.5/%
mAP@0.5~
0.95/%
FLOPs/109
YOLOv8n 90.7 91.6 91.3 48.4 8.5
改进1 92.8 93.8 93.4 52.9 8.8
改进2 90.8 93.4 92.6 50.5 8.5
改进3 90.9 92.7 91.7 49.3 8.6
SEC-YOLO 93.1 95.9 94.5 54.2 8.9
), ArticleFig(id=1167743197389271321, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, label=表2, caption=

消融试验结果

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 SPD-
Conv
CA EIoU P/% R/% mAP@
0.5/%
mAP@0.5~
0.95/%
FLOPs/109
YOLOv8n 90.7 91.6 91.3 48.4 8.5
改进1 92.8 93.8 93.4 52.9 8.8
改进2 90.8 93.4 92.6 50.5 8.5
改进3 90.9 92.7 91.7 49.3 8.6
SEC-YOLO 93.1 95.9 94.5 54.2 8.9
), ArticleFig(id=1167743197456380186, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=EN, label=Table 3, caption=

Experimental results of the comprehensive index of the improved model and the mainstream model

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 R/% mAP@
0.5/%
mAP@
0.5~0.95/%
FPS
Faster R-CNN 85.2 84.3 42.1 14.7
YOLOv8n 91.6 91.3 48.4 87.4
SSD 83.1 84.4 37.7 44.8
YOLOv5n 89.8 88.5 45.3 63.5
YOLOv7 86.4 90.1 47.9 72.2
RetinaNet 79.8 83.2 39.5 24.9
SEC-YOLO 95.9 94.5 54.2 93.8
), ArticleFig(id=1167743197523489051, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149733270813851754, language=CN, label=表3, caption=

改进模型与主流模型综合指标试验结果

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 R/% mAP@
0.5/%
mAP@
0.5~0.95/%
FPS
Faster R-CNN 85.2 84.3 42.1 14.7
YOLOv8n 91.6 91.3 48.4 87.4
SSD 83.1 84.4 37.7 44.8
YOLOv5n 89.8 88.5 45.3 63.5
YOLOv7 86.4 90.1 47.9 72.2
RetinaNet 79.8 83.2 39.5 24.9
SEC-YOLO 95.9 94.5 54.2 93.8
)], 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.issn1003-3033.2024.12.0465, detailUrlEn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/EN/10.16265/j.cnki.issn1003-3033.2024.12.0465, pdfUrlCn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN/PDF/10.16265/j.cnki.issn1003-3033.2024.12.0465, pdfUrlEn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/EN/PDF/10.16265/j.cnki.issn1003-3033.2024.12.0465, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于改进YOLOv8n算法的城市空间混行交通参与者检测模型
收藏切换
PDF下载
周军超 1, 2 , 陈鑫 1 , 高建杰 3, ** , 章杰 3, 4
中国安全科学学报 | 公共安全 2024,34(12): 178-186
收起
收藏切换
中国安全科学学报 | 公共安全 2024, 34(12): 178-186
基于改进YOLOv8n算法的城市空间混行交通参与者检测模型
全屏
周军超1, 2 , 陈鑫1, 高建杰3, ** , 章杰3, 4
作者信息
  • 1 四川轻化工大学 机械工程学院,四川 自贡 643000
  • 2 成渝地区双城经济圈(泸州)先进技术研究院,四川 泸州 646000
  • 3 四川警察学院 智能警务四川省重点实验室,四川 泸州 646000
  • 4 长沙理工大学 汽车与机械工程学院,湖南 长沙 410114
  • 周军超 (1987—),男,湖北襄阳人,博士,副教授,主要从事城市智能交通安全检测、智能网联汽车安全驾驶与风险识别、新型城市轨道车辆安全与控制等方面的研究。E-mail:

    章杰,副教授。

通讯作者:

**高建杰(1985—),男,山东栖霞人,博士,副教授,主要从事城市交通安全管理、城市交通风险判别与检测等方面的研究。E-mail:
Urban spatially mixed traffic participants detection model based on improved YOLOv8n
Junchao ZHOU1, 2 , Xin CHEN1, Jianjie GAO3, ** , Jie ZHANG3, 4
Affiliations
  • 1 School of Mechanical Engineering,Sichuan University of Science & Engineering,Zigong Sichuan 643000,China
  • 2 Chengdu-Chongqing Economic Circle (Luzhou) Advanced Technology Research Institute,Luzhou Sichuan 646000,China
  • 3 Intelligent Policing Key Laboratory of Sichuan Province,Sichuan Police College,Luzhou Sichuan 646000,China
  • 4 College of Automotive and Mechanical Engineering,Changsha University of Science & Technology,Changsha Hunan 410114,China
出版时间: 2024-12-28 doi: 10.16265/j.cnki.issn1003-3033.2024.12.0465
文章导航
收藏切换

为提升智能网联汽车和交通监测系统对交通参与者的识别精度和检测速度,及时应对城市空间混行交通环境中的安全隐患,提出一种基于改进YOLOv8n算法的城市空间混行交通参与者检测模型。首先,在数据输入阶段利用几何变换和像素变换策略来防止过度拟合,提高鲁棒性和泛化性;其次,使用空间到深度的非跨行卷积(SPD-Conv)模块代替所有原始卷积层,提高对低分辨率小目标的特征提取能力;同时,在颈部网络融合结构中加入轻量级坐标注意力机制(CA)模块,在几乎不添加任何计算开销的同时提高对关键信息的识别精度;然后,用边界框损失函数有效交并比(EIoU)替代原损失函数,使模型获得更卓越的收敛速度与识别稳定性;最后,利用平台公开和自建融合的交通参与者数据集(BNS)进行消融和对比试验,运用自动驾驶试验平台进行实景实时检测。结果表明:相比于基线模型YOLOv8n,改进模型SEC-YOLO的每秒传输帧数(FPS)和平均精度均值(mAP)分别提高了7.3%和3.2%;与主流模型对比,mAP和FPS性能值最佳;在自动驾驶试验平台上的实景检测平均准确率为95%。SEC-YOLO算法模型对城市交通参与者的检测准确率更高,鲁棒性和实时性更强。

YOLOv8n  /  空间混行  /  交通参与者  /  检测模型  /  空间到深度的非跨行卷积(SPD-Conv)  /  坐标注意力机制(CA)

In order to improve the recognition accuracy and detection speed of traffic participants by intelligent networked vehicles and traffic monitoring systems so that they can timely respond to the safety hazards in the mixed traffic environment in urban space,a mixed traffic participant detection model in urban space based on the improved YOLOv8n algorithm was proposed. Firstly,geometric transformation and pixel transformation enhancement strategies were employed in the data input stage to prevent overfitting and improve robustness,and generalization. Secondly,the SPD-Conv module was used to replace all original convolution layers of the YOLOv8n algorithm,which enhances the feature extraction capability for low-resolution small targets. Meanwhile,the CA module was added to the fusion structure of the neck network of the YOLOv8n algorithm to improve the recognition accuracy of key information with almost no additional computational overhead. Then,the boundary box loss function EIoU was used to replace the original loss function,enabling the model to achieve superior convergence speed and recognition stability. Finally,the ablation and comparison experiments were carried out with the public and self-built integrated traffic participant dataset,and the real-time detection experiment was carried out with the automatic driving experiment platform. The experimental results show that compared to the YOLOv8n model,the improved SEC-YOLO model has increased mAP and FPS by 3.2% and 7.9% respectively. The SEC-YOLO model outperforms mainstream models in terms of mAP and FPS as well. The average accuracy of real-scene detection on the automatic driving experimental platform is around 95%. The SEC-YOLO algorithm model achieves higher detection accuracy for urban traffic participants,with stronger robustness and real-time performance.

YOLOv8n  /  spatial mixing  /  traffic participants  /  detection model  /  space-to-depth convolution (SPD-Conv)  /  coordinate attention (CA)
周军超, 陈鑫, 高建杰, 章杰. 基于改进YOLOv8n算法的城市空间混行交通参与者检测模型. 中国安全科学学报, 2024 , 34 (12) : 178 -186 . DOI: 10.16265/j.cnki.issn1003-3033.2024.12.0465
Junchao ZHOU, Xin CHEN, Jianjie GAO, Jie ZHANG. Urban spatially mixed traffic participants detection model based on improved YOLOv8n[J]. China Safety Science Journal, 2024 , 34 (12) : 178 -186 . DOI: 10.16265/j.cnki.issn1003-3033.2024.12.0465
随着城市化进程的不断加速和交通工具的多样化,交通参与者(行人、机动车、非机动车等)的种类和数量呈现指数级增长,且常常处于复杂的空间混行状态,使得现实城市交通环境变得越来越复杂,交通安全问题也愈发严峻[1]。智能目标检测技术的出现使得交通安全问题得到缓解,在空间混行状态下,能够在密集、遮挡、低分辨率小目标[2]和恶劣天气等复杂场景下快速且准确地识别和监测多类别交通参与者,这对于智能网联汽车的安全驾驶和城市智慧交通监测系统的实时安全管控具有重要意义。
近年来,由于深度学习能够从数据中学习有效特征,已逐渐替代了传统的机器学习算法,成为主流的交通参与者检测算法。单阶段检测算法如You Only Look Once(YOLO)系列[3]、RetinaNet[4]、和单次多盒检测(Single Shot multiBox Detector,SSD)[5]等,因从输入图像到获取边界框位置和类别的一步实现,在检测实时性和部署便利性上更优于两阶段算法,被广泛用于交通和行人的检测。夏正洪等[6]提出一种基于改进YOLOv3算法的运动目标检测模型,通过改进算法主干网络,实现了对车辆和人员的有效实时性检测。SHU Jijun等[7]基于改进的YOLOv4算法搭建了小目标行人检测模型,该模型获得了较好的鲁棒性和特征判别力。WANG Shouhe等[8]通过加强YOLOv5s算法的骨干和颈部网络层融合,对交通沿线侵入性目标异物的识别准确率达到93%左右。ZHOU Wei等[9]针对在恶劣环境下对交通车辆监测难、误检率高的问题,提出基于改进YOLOv6算法的交通车辆监测模型,大幅提升了模型的检测性能。王泽宇等[10]提出基于改进YOLOv8算法的密集行人检测模型,通过引入注意力机制策略[11],使模型的特征提取能力与检测精度得到较大提升。YOLOv8是YOLO系列目前最新的检测模型[12],在检测速度和精度上均有提升,同时,由于其便于轻量设备部署的缘由,改进后非常适用于空间混行场景下对交通参与者的检测。
上述研究在不同类别的交通目标检测上取得了良好成果,但对于多类别交通参与者的检测研究较少,尤其针对在城市空间混行复杂背景下的精准和实时检测研究比较鲜见。因此,笔者拟提出一种用于在城市空间混行状态下对交通参与者检测的模型,通过改进YOLOv8n算法模型的整体卷积层和颈部网络融合层,并优化数据的输入和输出阶段,以提升对交通参与者低分辨率小目标的特征提取能力,提高对多类别交通参与者检测的精确性、实时性和稳定性,为智能网联汽车和交通监测系统的智能安全检测与风险识别提供可靠的理论支持。
使用的数据集主要来自于平台公开数据集和实地采集数据集2处,以确保丰富的驾驶场景输入,提高模型的泛化能力和鲁棒性。由于交通参与者需处在空间混行下这一交通状态,选用伯克利深度驾驶100K(Berkeley Deep Drive 100K,BDD100K)[13]数据集和下一事件预测与轨迹(Next Event Prediction and Trajectory,NEXET)[14]数据集中符合场景的部分,并采集、标注自建数据集(Self-built),以补充数据集缺失场景,最后将3种数据集融合,且命名为BNS(BDD100K,NEXET,Self- built)数据集。
自建数据集采集地位于中国重庆市的多个街道和路口,获取设备为Canon r6二代,共采集时长为10min的视频42段。视频类型有人车密集混行的交叉路口、城市交通道上的斑马线路口、雨天非机动车道、夜晚人车混行街道等13个不同场景的视频,随后通过OpenCV进行关键帧提取,部分提取图像如图1所示。最后使用LabelImg工具标注交通参与者的类型和位置,格式为YOLO。
为提高改进模型对交通参与者特征的识别性能,采用几何变换和像素变换增强策略,在数据输入阶段随机对数据集引入几何变换、对比度调整、分辨率降低和噪声添加4种数据增强变换方式,如图2所示。以上方式丰富了数据训练地输入方式,可有效提高模型对交通参与者检测的鲁棒性和泛化能力。
基于YOLOv8n模型进行4个关键改进,详细结构如图3所示。首先,在数据集输入阶段采用几何变换和像素变换增强策略来随机引入4种数据增强方式;其次,用空间到深度的非跨行卷积(Space-to-Depth Convolution,SPD-Conv)[15]模块取代了原有的全部卷积结构;随后,在颈部融合网络结构中引入轻量级坐标注意力机制(Coordinate Attention,CA)[16]来改进Neck网络结构中的特征提取能力;最后,用有效交并比(Efficient IoU,EIoU)[17]替代原损失函数来优化边界框的计算。改进后算法模型即命名为SEC(根据SPD-Conv、EIoU和CA模块首字母命名)-YOLO。
交通监测系统和智能网联汽车在检测多类别城市交通参与者时,常出现检测目标为小物体或低分辨率,此时YOLOv8n的卷积神经网络(Convolution Neural Networks,CNN)会遭受细粒度信息丢失和特征学习不良的影响,检测准确率会大幅降低。针对此问题,引入SPD-Conv模块来取代CNN的每个跨行卷积层和池化层。SPD-Conv模块可在不丢失可学习信息的情况下对特征图进行下采样,极大地弥补了原模型对低分辨率和小目标的交通参与者的特征检测性能缺陷。SPD-Conv模块中的SPD组件通过图像转换技术对CNN整个的特征映射来进行下采样,原理如下。
令任意长宽高大小为 C 1 × S × S的中间特征映射为 X,则子特征映射序列切片为:
f ( 1,0 ) = X [ 1 : S : s c a l e 0 : S : s c a l e ] f ( s c a l e 0 ) = X [ s c a l e : S : s c a l e 0 : S : s c a l e ] ; f ( 1,1 ) f ( s c a l e 0 ) = X [ s c a l e : S : s c a l e 1 : S : s c a l e ] ; f ( 0 s c a l e ) = X [ 0 : S : s c a l e s c a l e : S : s c a l e ] f ( s c a l e - 1,0 ) f ( s c a l e s c a l e ) = X [ s c a l e : S : s c a l e 0 : S : s c a l e ]
式中: X为给定任何特征映射; f ( x y )子映射由所有特征映射组成特征图 X ( i j )。每个子图按一个比例因子向下采样X。这种映射使得卷积层得到不同的扩张率,从而在不同尺度上获得更广泛的视野。图4为当倍率 s c a l e = 2时的SPD-Conv下采样示例,可得4个子映射,它们的形状为(S/2,S/2,C1)。随后对 X进行2倍下采样,最终得到的通道维度就会增加一个比例因子2。通过引用SPD-Conv模块,使得YOLOv8n可以在不同尺度上获得更广泛的感受野。在城市空间混行状态下,极大增强了模型对多类别交通参与者小目标和低分辨率目标的特征提取能力,从而提高检测准确性。
在空间混行的城市交叉路口,采集到的图像往往包含多个种类的交通参与者,且数量密集,传统YOLO系列会因产生多目标信息卷积迭代而积累大量冗余的无效信息,从而目标信息被淹没。针对以上问题,引入轻量级CA模块。该模块通过将位置信息嵌入到通道注意力中,不仅捕获了跨通道信息,还包含了方向感知和位置感知的信息,使得YOLOv8n可以更准确地定位并识别到目标区域,同时也避免了产生大量的计算开销,其实现原理如图5所示。
图5中,N表示通道数,W表示宽度,H表示高度,1表示维度。输入图像首先通过X Avg Pool和 Y Avg Pool进行一维全局池化操作,分别将输入特征聚合为2个独立的方向感知特征图;随后通过BatchNorm和Non-linear将感知特征图编码为2个注意力图;最后通过split分离与方向加权,成功地让每个注意力图都捕获了输入特征图沿着一个空间方向的长程依赖。CA模块的融入,不仅能有效提高YOLOv8n对交通参与者关键信息的识别精度,而且由于其轻量化的结构特点,运行时几乎不会对设备增添任何计算开销。
损失函数常被用来衡量预测结果与真实结果之间的差距,YOLOv8系列均使用完全交并比(Complete IoU,CIoU)损失函数作为边界损失的计算。然而,根据CIoU损失函数有界盒回归图(图6),CIoU损失函数其纵横比却是相对值,在不规则目标帧之间容易失衡,造成模型的检测稳定性下降。针对以上问题,使用EIoU损失函数替代YOLOv8n模型中的CIoU损失函数。
EIoU损失函数通过分别计算宽高的差异值来取代纵横比,同时,解决了目标帧失衡的问题。EIoU损失函数定义如下:
L E I o U = L I o U + L d + L a
L I o U = 1 - I o U
L d = ρ 2 ( b g b p ) ( w c ) 2 + ( h c ) 2
L a = ρ 2 ( w w g ) ( w c ) 2 + ρ 2 ( h h g ) ( h c ) 2
其中,EIoU的损失函数包含3部分:LIoU损失,中心距离损失Ld和宽高损失Lawchc分别为覆盖2个盒子最小围框的宽度和高度,其余参数变量含义与图6相同。EIoU损失函数有效避免了在检测交通参与者时因纵横比不同而发生的失效。因此,EIoU损失函数的引用可更好地保证智能网联汽车和交通监测系统在对交通参与者检测时的稳定性和实时性。
所用的试验设备为计算机,其配置见表1。文中所涉及的所有试验皆在此环境下进行。在训练过程中,训练轮次设置为300轮,初始学习率为0.002,权重衰减为0.000 5,批处理大小设置为64。
选用衡量模型计算量的浮点计算数(Floating Point Operations,FLOPs)和衡量模型实时性的每秒传输帧数(Frames Per Second,FPS)作为模型性能的评价指标。使用精确率P、召回率R和平均精度均值(Mean Average Precision,mAP)作为模型精度的衡量指标。相应计算过程如下:
FPS=Q_n/T
P = T P T P + F P
R = T P T P + F N
m A P = 1 n i = 1 n 0 1 P R d R
式中:Qn为检测视图在为n时的总数量;T为所用总时间,s;TP为被正确识别的数量;FP为被错误识别的数量;FN为未被识别出的或识别错误的数量。
将改进后的SEC-YOLO模型与原模型进行平均精度均值和边界损失对比,训练轮次为0~300轮,结果如图7所示。由图7a可知:在约60轮迭代后,SEC-YOLO模型对交通参与者的识别精度有大幅提升;由7b可知:在约50轮迭代后,边界损失曲线已趋于稳定,SEC-YOLO模型收敛的速度更快且更稳定。基于上述分析,改进后的SEC-YOLO模型在对交通参与者检测任务上表现更好,证明了改进模型的可行性。
为验证每个模块对改进模型的有效性,设计
消融试验,其中符号“√”代表“模块使用”的意思。其结果见表2
表2结果可以看出,当仅使用SPD-Conv模块时,mAP@0.5提升2.1%,表明SPD-Conv模块能够提升网络模型的整体性能;仅使用CA模块时,mAP@0.5提升1.3%,而FLOPs却无任何提升,表明CA模块能够提高模型的检测精度,且几乎无任何计算开销;仅使用EIOU损失函数式时,mAP@0.5仍有小幅提升,表明EIOU损失函数能提升模型的检测准确度;当同时使用3种模型改进时,mAP@0.5和mAP@0.5~0.95分别提升3.2%和5.8%。由此可见:改进后的SEC-YOLO模型比原模型在交通参与者检测性能上有着显著的提升。
SEC-YOLO模型与另外6种主流目标检测模型的对比试验结果见表3图8。由表3可知:SEC-YOLO模型速率的FPS达到93.8帧/s,检测实时性能最强;由图8可知:SEC-YOLO模型mAP@0.5和R值最高,且收敛速度最快。
BNS数据集检测:选择基线模型YOLOv8n与本文改进模型SEC-YOLO进行交通参与者检测试验对比,结果如图9所示。图9中,交通参与者上方的预测框文字和数字分别表示类别和置信度,图片右上角为目标检测局部放大图。结果表明:基线模型YOLOv8n在对远处小目标、多类别密集聚集、低光亮环境等复杂场景检测中存在一定的漏检和误检,从第2列局部放大图可以看出,SEC-YOLO模型在上述3种环境中均能进行有效检测,且误检率更低,由第3列图可以看出。综上,SEC-YOLO模型能在空间混行场景中准确识别出部分遮挡或低分辨率的目标,在复杂交通场景中对交通参与者的检测更具优势。
自动驾驶车实景检测:采用自动驾驶试验平台,将SEC-YOLO模型导入平台检测交通参与者,检测地址为中国宜宾市某一街道,检测结果如图10所示。为准确测试出漏检率,平台在检测多类别交通参与者的同时统计了实时检测数量,实际数量则通
过录制的同检测场景画面进行后期人工观察数量统计,并对统计再次校核,频率为5m/次,得到的结果对比如图11所示。结果显示,当在光线小幅变化和直行街道空间混行场景的里程点时,不同类别的交通参与者检测数量和实际数量偏差极小,平台能稳定持续检测;当在交通参与者数量激增的交叉路口场景里程点时,检测数量和实际数量会出现小幅的偏差波动,但平台仍能稳定持续检测。将全里程的交通参与者正确检测总数与实际检测总数值相比,得到的平均准确识别率能达到95%以上。综上可见:所改进的算法模型在实景街道上对交通参与者有着良好的实时检测性和低误检性。
1) 将SPD-Conv模块和CA模块融入YOLOv8n模型中,可显著提升模型对低分辨率小目标的检测能力和特征提取能力;改进的数据输入增强策略结合边界框EIoU损失函数,能有效提高对交通参与者检测的鲁棒性和实时性。
2) 试验结果证明:SEC-YOLO算法对城市多类别交通参与者的检测有着较高的准确性和实时性,且检测精度优于YOLOv8n、YOLOv5n和Faster R-CNN等算法。就数据集检测而言,识别速度达到93.8帧/s,平均精度均值达到94.5%;就实景检测而言,准确率达到95%。
3) 下一步将重点对安全检测模型轻量化与设备嵌入部署性的优化展开研究,在保持检测性能不变的同时大幅降低对设备算力的要求,以求能在更多交通环境下进行便利部署。
  • 智能警务四川省重点实验室开放课题资助(ZNJW2023KFQN005)
  • 智能警务四川省重点实验室开放课题资助(ZNJW2022KFQN004)
  • 泸州市科技计划资助(2023JYJ066)
参考文献 引证文献
排序方式:
[1]
刘晓芳, 吝涛, 于琪, 等. 城镇街巷空间人群通行疏散及碰撞风险定量模拟[J]. 中国安全科学学报, 2023, 33(11):133-141.
LIU Xiaofang, LIN Tao, YU Qi, et al. Quantitative simulation of pedestrian evacuation and collision risk in urban street space[J]. China Safety Science Journal, 2023, 33(11):133-141.
[2]
张军, 尹柳, 巩欣飞, 等. 基于改进YOLOv5的小目标烟雾检测算法[J]. 中国安全科学学报, 2023, 33(11):75-81.
ZHANG Jun, YIN Liu, GONG Xinfei, et al. Small target smoke detection algorithm based on improved YOLOv5[J]. China Safety Science Journal, 2023, 33(11):75-81.
[3]
REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]. Proceedings of the IEEE Conference on Computer Cision and Cattern Recognition, 2016: 779-788.
[4]
LIN Tianyi, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[C]. Proceedings of the IEEE International Conference on Computer Vision, 2017: 2980-2988.
[5]
LIU Wu, ANGUELOV D, ERHAN D, et al. SSD: single shot multibox detector[C]. Proceedings of the 2016 European Conference on Computer Vision, 2016: 21-37.
[6]
夏正洪, 魏汝祥, 李彦冬. 改进 YOLOv3 算法下通航机场场面运动目标检测[J]. 中国安全科学学报, 2023, 33(2):82-88.
XIA Zhenghong, WEI Ruxiang, LI Yandong. Moving target detection of general aviation airport based on improved YOLOv3 algorithm[J]. China Safety Science Journal, 2023, 33(2):82-88.
[7]
SHU Jijun, LING Qinghua, HAN Fei. An improved algorithm for small object detection based on YOLO v4 and multi-scale contextual information[J]. Computers and Electrical Engineering, 2023, 105:84-90.
[8]
WANG Shouhe, WANG Yongda, CHANG Yujian, et al. EBSE-YOLO: high precision recognition algorithm for small target foreign object detection[J]. IEEE Access, 2023, 11:57951-57 964.
[9]
ZHOU Wei, WANG Chuan, XIA Jie, et al. Monitoring-based traffic participant detection in urban mixed traffic: a novel dataset and a tailored detector[J]. IEEE Transactions on Intelligent Transportation Systems, 2024, 25(1):189-202.
[10]
王泽宇, 徐慧英, 朱信忠, 等. 基于YOLOv8改进的密集行人检测算法:MER-YOLO[J]. 计算机工程与科学, 2024, 46(6):1050-1062.
WANG Zeyu, XU Huiying, ZHU Xinzhong, et al. An improved dense pedestrian detection algorithm based on YOLOv8: MER-YOLO[J]. Computer Engineering and Science, 2024, 46(6):1050-1062.
[11]
OUYANG Daliang, HE Su, ZHANG Guozhong, et al. Efficient multi-scale attention module with cross-spatial learning[C]. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023: 1-5.
[12]
YI Hao, LIU Bo, ZHAO Bin, et al. Small object detection algorithm based on improved YOLOv8 for remote sensing[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2024, 17:1734-1747.
[13]
YU Fu, CHEN Hui, WANG Xia, et al. BDD100K: a diverse driving dataset for heterogeneous multitask learning[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2020: 2636-2645.
[14]
CATAK O, TAN Yue, ALI S. Prediction surface uncertainty quantification in object detection models for autonomous driving[C]. IEEE International Conference on Artificial Intelligence Testing, 2021: 93-100.
[15]
SUNKARA R, LUO Tao. No more strided convolutions or pooling: a new CNN building block for low-resolution images and small objects[C]. Joint European Conference on Machine Learning and Knowledge Discovery in Databases, 2022: 443-459.
[16]
HOU Qibin, ZHOU Daquan, FENG Jiashi. Coordinate attention for efficient mobile network design[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2021: 13 713-13 722.
[17]
YANG Zuomin, WANG Xialun, LI Jiang. EIoU: an improved vehicle detection algorithm based on vehiclenet neural network[J]. Journal of Physics Conference Series, 2021, 1924(1):DOI:10.1088/1742-6596/1924/1/012001.
2024年第34卷第12期
PDF下载
446
191
引用本文
BibTeX
文章信息
doi: 10.16265/j.cnki.issn1003-3033.2024.12.0465
  • 接收时间:2024-08-17
  • 首发时间:2025-07-09
  • 出版时间:2024-12-28
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-08-17
  • 修回日期:2024-10-16
基金
智能警务四川省重点实验室开放课题资助(ZNJW2023KFQN005)
智能警务四川省重点实验室开放课题资助(ZNJW2022KFQN004)
泸州市科技计划资助(2023JYJ066)
作者信息
    1 四川轻化工大学 机械工程学院,四川 自贡 643000
    2 成渝地区双城经济圈(泸州)先进技术研究院,四川 泸州 646000
    3 四川警察学院 智能警务四川省重点实验室,四川 泸州 646000
    4 长沙理工大学 汽车与机械工程学院,湖南 长沙 410114

通讯作者:

**高建杰(1985—),男,山东栖霞人,博士,副教授,主要从事城市交通安全管理、城市交通风险判别与检测等方面的研究。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN/10.16265/j.cnki.issn1003-3033.2024.12.0465
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