Article(id=1203753463671927446, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1203753457208504777, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2402124, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1711296000000, receivedDateStr=2024-03-25, revisedDate=1730390400000, revisedDateStr=2024-11-01, acceptedDate=null, acceptedDateStr=null, onlineDate=1764926790396, onlineDateStr=2025-12-05, pubDate=1737129600000, pubDateStr=2025-01-18, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1764926790396, onlineIssueDateStr=2025-12-05, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1764926790396, creator=13701087609, updateTime=1764926790396, updator=13701087609, issue=Issue{id=1203753457208504777, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='2', pageStart='439', pageEnd='878', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1764926788856, creator=13701087609, updateTime=1764928745558, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1203761664261858014, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1203753457208504777, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1203761664261858015, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1203753457208504777, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=773, endPage=779, ext={EN=ArticleExt(id=1203753464154272457, articleId=1203753463671927446, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=High Speed Railway Station Car Linkage Control Technology Based on Machine Vision, columnId=1156262728772735295, journalTitle=Science Technology and Engineering, columnName=Papers·Traffics and Transportations, runingTitle=null, highlight=null, articleAbstract=

To efficiently identify the opening and closing status of train doors and control the synchronous opening and closing of platform doors, a lightweight MobileNet network and machine vision based image recognition method was proposed to achieve linkage control between high-speed railway platform doors and train doors. A large dataset of train door images was collected from Beijing South Station and preprocessed to serve as the training and testing dataset for the model. The constructed network was trained and optimized using a binary cross-entropy loss function and the Adam optimization algorithm to achieve efficient and accurate recognition of door status. Validation results demonstrate an accuracy rate of over 95% in recognizing train door actions, with recognition time kept within 400 milliseconds. These results meet the current industry application requirements and greatly enhance the automation and intelligence level of the platform door system.

, correspAuthors=Zhi-fei WANG, 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=Shuai LI, Zhi-fei WANG, Fan LI, Cheng-xin DU, Hao-dong WANG, Bo-xuan YANG), CN=ArticleExt(id=1203753465504838491, articleId=1203753463671927446, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=基于机器视觉的高速铁路站车联动控制技术, columnId=1156262730664366426, journalTitle=科学技术与工程, columnName=论文·交通运输, runingTitle=null, highlight=null, articleAbstract=

为高效识别列车车门的开关状态,并据此控制站台门的同步开关,提出了一种基于轻量级MobileNet网络和机器视觉的图像识别方法,实现高速铁路站台门与列车门的联动控制。在北京南站收集大量列车车门的图像资料,经过预处理后作为模型训练和测试的数据集,再利用二元交叉熵损失函数和Adam优化算法对构建的网络进行训练和优化,最终实现对车门状态的高效精准识别。验证结果表明:对列车开关门动作的识别准确率达到95%以上,识别时间控制在400 ms以内,均能满足当前行业应用需求,极大提高站台门系统的自动化和智能化水平。

, correspAuthors=王志飞, authorNote=null, correspAuthorsNote=
* 王志飞(1981—),男,汉族,内蒙古鄂尔多斯人,博士,研究员。研究方向:交通信息工程与控制。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=ayRXnL/StcVRZN/gWLMR/A==, magXml=kFr9t4Y+f75ODXa91WB9fg==, pdfUrl=null, pdf=bEF2Ns6uIM+fOHW+SdYxPQ==, pdfFileSize=5045164, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=Nu4F2rOSqaxK2A4F1czH4g==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=K/B4+AdN19giWtAghBukNg==, mapNumber=null, authorCompany=null, fund=null, authors=

李帅(1989—),男,汉族,河南郑州人,硕士,助理研究员。研究方向:控制工程。E-mail:

, authorsList=李帅, 王志飞, 李樊, 杜呈欣, 王浩东, 杨博璇)}, authors=[Author(id=1203787143161291370, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=1101986417@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1203787143249371761, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, authorId=1203787143161291370, language=EN, stringName=Shuai LI, firstName=Shuai, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Institute of Electronic Computing Technology, China Academy of Railway Sciences, Beijing 100081, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1203787143333257844, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, authorId=1203787143161291370, language=CN, stringName=李帅, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081, bio={"content":"

李帅(1989—),男,汉族,河南郑州人,硕士,助理研究员。研究方向:控制工程。E-mail:

"}, bioImg=null, bioContent=

李帅(1989—),男,汉族,河南郑州人,硕士,助理研究员。研究方向:控制工程。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1203787143043850852, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, xref=null, ext=[AuthorCompanyExt(id=1203787143048045158, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, companyId=1203787143043850852, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Institute of Electronic Computing Technology, China Academy of Railway Sciences, Beijing 100081, China), AuthorCompanyExt(id=1203787143056433767, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, companyId=1203787143043850852, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081)])]), Author(id=1203787143417143930, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=wfbuaa@126.com, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1203787143513612928, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, authorId=1203787143417143930, language=EN, stringName=Zhi-fei WANG, firstName=Zhi-fei, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=*, address=Institute of Electronic Computing Technology, China Academy of Railway Sciences, Beijing 100081, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1203787143605887619, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, authorId=1203787143417143930, language=CN, stringName=王志飞, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=*, address=中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1203787143043850852, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, xref=null, ext=[AuthorCompanyExt(id=1203787143048045158, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, companyId=1203787143043850852, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Institute of Electronic Computing Technology, China Academy of Railway Sciences, Beijing 100081, China), AuthorCompanyExt(id=1203787143056433767, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, companyId=1203787143043850852, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081)])]), Author(id=1203787144776098442, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1203787144906121872, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, authorId=1203787144776098442, language=EN, stringName=Fan LI, firstName=Fan, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Institute of Electronic Computing Technology, China Academy of Railway Sciences, Beijing 100081, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1203787144994202262, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, authorId=1203787144776098442, language=CN, stringName=李樊, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1203787143043850852, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, xref=null, ext=[AuthorCompanyExt(id=1203787143048045158, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, companyId=1203787143043850852, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Institute of Electronic Computing Technology, China Academy of Railway Sciences, Beijing 100081, China), AuthorCompanyExt(id=1203787143056433767, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, companyId=1203787143043850852, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081)])]), Author(id=1203787145078088347, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, 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=1203787145187140258, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, authorId=1203787145078088347, language=EN, stringName=Cheng-xin DU, firstName=Cheng-xin, middleName=null, lastName=DU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Institute of Electronic Computing Technology, China Academy of Railway Sciences, Beijing 100081, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1203787145291997866, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, authorId=1203787145078088347, language=CN, stringName=杜呈欣, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1203787143043850852, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, xref=null, ext=[AuthorCompanyExt(id=1203787143048045158, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, companyId=1203787143043850852, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Institute of Electronic Computing Technology, China Academy of Railway Sciences, Beijing 100081, China), AuthorCompanyExt(id=1203787143056433767, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, companyId=1203787143043850852, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081)])]), Author(id=1203787145396855472, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1203787145640125114, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, authorId=1203787145396855472, language=EN, stringName=Hao-dong WANG, firstName=Hao-dong, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Institute of Electronic Computing Technology, China Academy of Railway Sciences, Beijing 100081, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1203787145753371324, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, authorId=1203787145396855472, language=CN, stringName=王浩东, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1203787143043850852, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, xref=null, ext=[AuthorCompanyExt(id=1203787143048045158, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, companyId=1203787143043850852, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Institute of Electronic Computing Technology, China Academy of Railway Sciences, Beijing 100081, China), AuthorCompanyExt(id=1203787143056433767, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, companyId=1203787143043850852, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081)])]), Author(id=1203787145824674496, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, orderNo=5, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1203787145946309320, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, authorId=1203787145824674496, language=EN, stringName=Bo-xuan YANG, firstName=Bo-xuan, middleName=null, lastName=YANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Institute of Electronic Computing Technology, China Academy of Railway Sciences, Beijing 100081, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1203787146093109973, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, authorId=1203787145824674496, language=CN, stringName=杨博璇, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1203787143043850852, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, xref=null, ext=[AuthorCompanyExt(id=1203787143048045158, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, companyId=1203787143043850852, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Institute of Electronic Computing Technology, China Academy of Railway Sciences, Beijing 100081, China), AuthorCompanyExt(id=1203787143056433767, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, companyId=1203787143043850852, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081)])])], keywords=[Keyword(id=1203787146273465056, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=EN, orderNo=1, keyword=convolutional neural network), Keyword(id=1203787146369934055, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=EN, orderNo=2, keyword=machine vision), Keyword(id=1203787146453820140, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=EN, orderNo=3, keyword=platform door system), Keyword(id=1203787146558677745, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=EN, orderNo=4, keyword=linkage control), Keyword(id=1203787146638369525, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=CN, orderNo=1, keyword=卷积神经网络), Keyword(id=1203787146772587261, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=CN, orderNo=2, keyword=机器视觉), Keyword(id=1203787146919387908, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=CN, orderNo=3, keyword=站台门系统), Keyword(id=1203787147045217037, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=CN, orderNo=4, keyword=联动控制)], refs=[Reference(id=1203787150903975949, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=9, pageStart=56, pageEnd=63, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=方凯, 史天运, 陈瑞凤, journalName=铁道运输与经济, refType=null, unstructuredReference=方凯, 史天运, 陈瑞凤, 等. 面向北京冬奥会的京张高速铁路旅客服务智能化成套关键技术研究[J]. 铁道运输与经济, 2022(9): 56-63., articleTitle=面向北京冬奥会的京张高速铁路旅客服务智能化成套关键技术研究, refAbstract=null), Reference(id=1203787151134662684, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=9, pageStart=56, pageEnd=63, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=Fang Kai, Shi Tianyun, Chen Ruifeng, journalName=Railway Transportation and Economy, refType=null, unstructuredReference=Fang Kai, Shi Tianyun, Chen Ruifeng, et al. Research on key technologies for intelligent passenger service of Beijing Zhangjiakou high speed railway for the Beijing Winter Olympics[J]. Railway Transportation and Economy, 2022(9): 56-63., articleTitle=Research on key technologies for intelligent passenger service of Beijing Zhangjiakou high speed railway for the Beijing Winter Olympics, refAbstract=null), Reference(id=1203787151298240557, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=22, pageStart=114, pageEnd=120, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=王爱武, 黄海宾, 周梅, journalName=电子技术与软件工程, refType=null, unstructuredReference=王爱武, 黄海宾, 周梅, 等. CTCS2+ATO控制模式下的车地联动方案[J]. 电子技术与软件工程, 2018(22): 114-120., articleTitle=CTCS2+ATO控制模式下的车地联动方案, refAbstract=null), Reference(id=1203787151445041216, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=22, pageStart=114, pageEnd=120, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=Wang Aiwu, Huang Haibin, Zhou Mei, journalName=Electronic Technology and Software Engineering, refType=null, unstructuredReference=Wang Aiwu, Huang Haibin, Zhou Mei, et al. Train ground linkage scheme under CTCS2+ATO control mode[J]. Electronic Technology and Software Engineering, 2018 (22): 114-120., articleTitle=Train ground linkage scheme under CTCS2+ATO control mode, refAbstract=null), Reference(id=1203787151759614034, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=19, pageStart=90, pageEnd=93, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=刘磊, 周超, journalName=电子技术与软件工程, refType=null, unstructuredReference=刘磊, 周超. C3+ATO系统与站台门的接口设计[J]. 电子技术与软件工程, 2018(19): 90-93., articleTitle=C3+ATO系统与站台门的接口设计, refAbstract=null), Reference(id=1203787151969329258, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=19, pageStart=90, pageEnd=93, url=null, language=null, rfNumber=[3], rfOrder=5, authorNames=Liu Lei, journalName=Electronic Technology and Software Engineering, refType=null, unstructuredReference=Liu Lei, Zhou Chao. Interface design between C3+ATO system and platform doors[J]. Electronic Technology and Software Engineering, 2018 (19): 90-93., articleTitle=Interface design between C3+ATO system and platform doors, refAbstract=null), Reference(id=1203787152145490048, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2020, volume=23, issue=1, pageStart=310, pageEnd=313, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=范楷, 冯浩楠, 段宏伟, journalName=城市轨道交通研究, refType=null, unstructuredReference=范楷, 冯浩楠, 段宏伟, 等. 基于近距离通信的站台门控制新方案[J]. 城市轨道交通研究, 2020, 23(1): 310-313., articleTitle=基于近距离通信的站台门控制新方案, refAbstract=null), Reference(id=1203787152346816664, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2020, volume=23, issue=1, pageStart=310, pageEnd=313, url=null, language=null, rfNumber=[4], rfOrder=7, authorNames=Fan Kai, Feng Haonan, Duan Hongwei, journalName=Urban Rail Transit Research, refType=null, unstructuredReference=Fan Kai, Feng Haonan, Duan Hongwei, et al. A new scheme for platform door control based on close range communication[J]. Urban Rail Transit Research, 2020, 23 (1): 310-313., articleTitle=A new scheme for platform door control based on close range communication, refAbstract=null), Reference(id=1203787152497811627, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=郜洪民, 郑伟, 赵阳, journalName=null, refType=null, unstructuredReference=郜洪民, 郑伟, 赵阳, 等. 基于多传感器融合技术的车门站台门联动控制系统及方法: CN110469230A[P]. 2019-11-19., articleTitle=null, refAbstract=null), Reference(id=1203787153714159812, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[5], rfOrder=9, authorNames=Gao Hongmin, Zheng Wei, Zhao Yang, journalName=null, refType=null, unstructuredReference=Gao Hongmin, Zheng Wei, Zhao Yang, et al. A door platform door linkage control system and method based on multi-sensor fusion technology: CN110469230A[P]. 2019-11-19., articleTitle=null, refAbstract=null), Reference(id=1203787153844183253, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=9, pageStart=3596, pageEnd=3604, url=null, language=null, rfNumber=[6], rfOrder=10, authorNames=李玥辰, 赵晓, 王若男, journalName=科学技术与工程, refType=null, unstructuredReference=李玥辰, 赵晓, 王若男, 等. 基于改进卷积神经网络的中药饮片图像识别[J]. 科学技术与工程, 2024, 24(9): 3596-3604., articleTitle=基于改进卷积神经网络的中药饮片图像识别, refAbstract=null), Reference(id=1203787153970012388, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=9, pageStart=3596, pageEnd=3604, url=null, language=null, rfNumber=[6], rfOrder=11, authorNames=Li Yuechen, Zhao Xiao, Wang Ruonan, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Li Yuechen, Zhao Xiao, Wang Ruonan, et al. Image recognition of chinese herbal slices based on improved convolution neural network[J]. Science Technology and Engineering, 2024, 24(9): 3596-3604., articleTitle=Image recognition of chinese herbal slices based on improved convolution neural network, refAbstract=null), Reference(id=1203787154108424438, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=8, pageStart=204, pageEnd=207, url=null, language=null, rfNumber=[7], rfOrder=12, authorNames=牛嘉兴, 高利鹏, 张露, journalName=计算机仿真, refType=null, unstructuredReference=牛嘉兴, 高利鹏, 张露, 等. 基于SSD算法的戴口罩人脸识别方法研究[J]. 计算机仿真, 2022(8): 204-207., articleTitle=基于SSD算法的戴口罩人脸识别方法研究, refAbstract=null), Reference(id=1203787154238447878, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=8, pageStart=204, pageEnd=207, url=null, language=null, rfNumber=[7], rfOrder=13, authorNames=Niu Jiaxing, Gao Lipeng, Zhang Lu, journalName=Computer Simulation, refType=null, unstructuredReference=Niu Jiaxing, Gao Lipeng, Zhang Lu, et al. Research on mask wearing face recognition method based on SSD algorithm[J]. Computer Simulation, 2022(8): 204-207., articleTitle=Research on mask wearing face recognition method based on SSD algorithm, refAbstract=null), Reference(id=1203787154339111188, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2023, volume=43, issue=1, pageStart=67, pageEnd=73, url=null, language=null, rfNumber=[8], rfOrder=14, authorNames=刘小宇, 陈怀新, 刘壁源, journalName=计算机应用, refType=null, unstructuredReference=刘小宇, 陈怀新, 刘壁源, 等. 自适应置信度阈值的非限制场景车牌检测算法[J]. 计算机应用, 2023, 43(1): 67-73., articleTitle=自适应置信度阈值的非限制场景车牌检测算法, refAbstract=null), Reference(id=1203787154490106152, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2023, volume=43, issue=1, pageStart=67, pageEnd=73, url=null, language=null, rfNumber=[8], rfOrder=15, authorNames=Liu Xiaoyu, Chen Huaixin, Liu Biyuan, journalName=Computer Applications, refType=null, unstructuredReference=Liu Xiaoyu, Chen Huaixin, Liu Biyuan, et al. Adaptive confidence threshold non restricted scene license plate detection algorithm[J]. Computer Applications, 2023, 43 (1): 67-73., articleTitle=Adaptive confidence threshold non restricted scene license plate detection algorithm, refAbstract=null), Reference(id=1203787154641101111, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=30, pageStart=12796, pageEnd=12805, url=null, language=null, rfNumber=[9], rfOrder=16, authorNames=宋泽冈, 刘艳莉, 张长兴, journalName=科学技术与工程, refType=null, unstructuredReference=宋泽冈, 刘艳莉, 张长兴. 基于机器视觉的桥梁裂缝检测应用及发展综述[J]. 科学技术与工程, 2023, 23(30): 12796-12805., articleTitle=基于机器视觉的桥梁裂缝检测应用及发展综述, refAbstract=null), Reference(id=1203787154754347325, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=30, pageStart=12796, pageEnd=12805, url=null, language=null, rfNumber=[9], rfOrder=17, authorNames=Song Zegang, Liu Yanli, Zhang Changxing, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Song Zegang, Liu Yanli, Zhang Changxing. Review on application and development of bridge crack detection based on machine vision[J]. Science Technology and Engineering, 2023, 23(30): 12796-12805., articleTitle=Review on application and development of bridge crack detection based on machine vision, refAbstract=null), Reference(id=1203787154871787847, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2020, volume=32, issue=11, pageStart=1863, pageEnd=1870, url=null, language=null, rfNumber=[10], rfOrder=18, authorNames=艾鑫, 邹琪, 罗常津, journalName=计算机辅助设计与图形学学报, refType=null, unstructuredReference=艾鑫, 邹琪, 罗常津. 面向精确定位的列车车号文本定位与识别[J]. 计算机辅助设计与图形学学报, 2020, 32(11): 1863-1870., articleTitle=面向精确定位的列车车号文本定位与识别, refAbstract=null), Reference(id=1203787154972451158, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2020, volume=32, issue=11, pageStart=1863, pageEnd=1870, url=null, language=null, rfNumber=[10], rfOrder=19, authorNames=Ai Xin, Zou Qi, journalName=Journal of Computer Aided Design and Graphics, refType=null, unstructuredReference=Ai Xin, Zou Qi, Luo Changjin. Train number text positioning and recognition for precise positioning[J]. Journal of Computer Aided Design and Graphics, 2020, 32(11): 1863-1870., articleTitle=Train number text positioning and recognition for precise positioning, refAbstract=null), Reference(id=1203787155089891691, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=1, pageStart=180, pageEnd=184, url=null, language=null, rfNumber=[11], rfOrder=20, authorNames=董咚, 王鹏月, journalName=现代科学仪器, refType=null, unstructuredReference=董咚, 王鹏月. 铁路货车车号图像识别系统的算法研究[J]. 现代科学仪器, 2022(1): 180-184., articleTitle=铁路货车车号图像识别系统的算法研究, refAbstract=null), Reference(id=1203787155165389174, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=1, pageStart=180, pageEnd=184, url=null, language=null, rfNumber=[11], rfOrder=21, authorNames=Dong Dong, Wang Pengyue, journalName=Modern Scienti-fic Instruments, refType=null, unstructuredReference=Dong Dong, Wang Pengyue. Research on algorithm for image recognition system of railway freight car number[J]. Modern Scienti-fic Instruments, 2022(1): 180-184., articleTitle=Research on algorithm for image recognition system of railway freight car number, refAbstract=null), Reference(id=1203787155274441096, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2022, volume=31, issue=9, pageStart=20, pageEnd=24, url=null, language=null, rfNumber=[12], rfOrder=22, authorNames=朱俊霖, 段钰, 滕凯, journalName=铁路计算机应用, refType=null, unstructuredReference=朱俊霖, 段钰, 滕凯, 等. 基于图像处理的城轨列车车号识别系统[J]. 铁路计算机应用, 2022, 31(9): 20-24., articleTitle=基于图像处理的城轨列车车号识别系统, refAbstract=null), Reference(id=1203787155400270233, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2022, volume=31, issue=9, pageStart=20, pageEnd=24, url=null, language=null, rfNumber=[12], rfOrder=23, authorNames=Zhu Junlin, Duan Yu, Teng Kai, journalName=Railway Computer Applications, refType=null, unstructuredReference=Zhu Junlin, Duan Yu, Teng Kai, et al. Urban rail train number recognition system based on image processing[J]. Railway Computer Applications, 2022, 31(9): 20-24., articleTitle=Urban rail train number recognition system based on image processing, refAbstract=null), Reference(id=1203787155538682281, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2021, volume=57, issue=11, pageStart=57, pageEnd=69, url=null, language=null, rfNumber=[13], rfOrder=24, authorNames=张驰, 郭媛, 黎明, journalName=计算机工程与应用, refType=null, unstructuredReference=张驰, 郭媛, 黎明. 人工神经网络模型发展及应用综述[J]. 计算机工程与应用, 2021, 57(11): 57-69., articleTitle=人工神经网络模型发展及应用综述, refAbstract=null), Reference(id=1203787155668705720, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2021, volume=57, issue=11, pageStart=57, pageEnd=69, url=null, language=null, rfNumber=[13], rfOrder=25, authorNames=Zhang Chi, journalName=Computer Engineering and Applications, refType=null, unstructuredReference=Zhang Chi, Guo Yuan, Li Ming. Dawn overview of the development and application of artificial neural network models[J]. Computer Engineering and Applications, 2021, 57(11): 57-69., articleTitle=Dawn overview of the development and application of artificial neural network models, refAbstract=null), Reference(id=1203787155786146249, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=12, pageStart=7, pageEnd=11, url=null, language=null, rfNumber=[14], rfOrder=26, authorNames=赵正阳, 吴艳华, 程智博, journalName=铁路计算机应用, refType=null, unstructuredReference=赵正阳, 吴艳华, 程智博, 等. 基于3D卷积神经网络的高铁轨道质量指数预测方法[J]. 铁路计算机应用, 2020(12): 7-11., articleTitle=基于3D卷积神经网络的高铁轨道质量指数预测方法, refAbstract=null), Reference(id=1203787155916169688, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=12, pageStart=7, pageEnd=11, url=null, language=null, rfNumber=[14], rfOrder=27, authorNames=Zhao Zhengyang, Wu Yanhua, Cheng Zhibo, journalName=Railway Computer Application, refType=null, unstructuredReference=Zhao Zhengyang, Wu Yanhua, Cheng Zhibo, et al. Prediction method of high-speed rail track quality index based on 3D convolutional neural network[J]. Railway Computer Application, 2020(12): 7-11., articleTitle=Prediction method of high-speed rail track quality index based on 3D convolutional neural network, refAbstract=null), Reference(id=1203787156029415911, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2023, volume=42, issue=9, pageStart=137, pageEnd=140, url=null, language=null, rfNumber=[15], rfOrder=28, authorNames=郭松, 尹明臣, journalName=传感器与微系统, refType=null, unstructuredReference=郭松, 尹明臣. 基于MobileNetV2的自然场景口罩检测算法[J]. 传感器与微系统, 2023, 42(9): 137-140., articleTitle=基于MobileNetV2的自然场景口罩检测算法, refAbstract=null), Reference(id=1203787156184605176, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2023, volume=42, issue=9, pageStart=137, pageEnd=140, url=null, language=null, rfNumber=[15], rfOrder=29, authorNames=Guo Song, Yin Mingchen, journalName=Sensors and Microsystems, refType=null, unstructuredReference=Guo Song, Yin Mingchen. Natural scene mask detection algorithm based on MobileNetV2[J]. Sensors and Microsystems, 2023, 42 (9): 137-140., articleTitle=Natural scene mask detection algorithm based on MobileNetV2, refAbstract=null), Reference(id=1203787156318822919, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=5, pageStart=118, pageEnd=122, url=null, language=null, rfNumber=[16], rfOrder=30, authorNames=Padmavathi P, Harikiran J, journalName=International Journal of Image and Graphics, refType=null, unstructuredReference=Padmavathi P, Harikiran J. Wireless capsule endoscopy infected images detection and classification using MobileNetV2-BiLSTM model[J]. International Journal of Image and Graphics, 2023, 23(5): 118-122., articleTitle=Wireless capsule endoscopy infected images detection and classification using MobileNetV2-BiLSTM model, refAbstract=null), Reference(id=1203787156494983703, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2022, volume=53, issue=2, pageStart=136, pageEnd=141, url=null, language=null, rfNumber=[17], rfOrder=31, authorNames=冯文彬, 厉舒南, 田昊, journalName=煤矿安全, refType=null, unstructuredReference=冯文彬, 厉舒南, 田昊, 等. 基于融合边缘优化的煤矿图像语义分割方法[J]. 煤矿安全, 2022, 53(2): 136-141., articleTitle=基于融合边缘优化的煤矿图像语义分割方法, refAbstract=null), Reference(id=1203787156616618527, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2022, volume=53, issue=2, pageStart=136, pageEnd=141, url=null, language=null, rfNumber=[17], rfOrder=32, authorNames=Feng Wenbin, Li Shunan, Tian Hao, journalName=Coal Mine Safety, refType=null, unstructuredReference=Feng Wenbin, Li Shunan, Tian Hao, et al. A semantic segmentation method for coal mine images based on fusion edge optimization[J]. Coal Mine Safety, 2022, 53 (2): 136-141., articleTitle=A semantic segmentation method for coal mine images based on fusion edge optimization, refAbstract=null), Reference(id=1203787156746641959, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2023, volume=44, issue=6, pageStart=1796, pageEnd=1803, url=null, language=null, rfNumber=[18], rfOrder=33, authorNames=王程远, 赵俊莉, 计晓斐, journalName=计算机工程与设计, refType=null, unstructuredReference=王程远, 赵俊莉, 计晓斐, 等. 基于改进的VGG网络的烟柜状态识别模型[J]. 计算机工程与设计, 2023, 44(6): 1796-1803., articleTitle=基于改进的VGG网络的烟柜状态识别模型, refAbstract=null), Reference(id=1203787156893442613, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2023, volume=44, issue=6, pageStart=1796, pageEnd=1803, url=null, language=null, rfNumber=[18], rfOrder=34, authorNames=Wang Chengyuan, Zhao Junli, Ji Xiaofei, journalName=Computer Engineering and Design, refType=null, unstructuredReference=Wang Chengyuan, Zhao Junli, Ji Xiaofei, et al. A smoke cabinet state recognition model based on an improved VGG network[J]. Computer Engineering and Design, 2023, 44(6): 1796-1803., articleTitle=A smoke cabinet state recognition model based on an improved VGG network, refAbstract=null), Reference(id=1203787157023466047, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=35, pageStart=15688, pageEnd=15696, url=null, language=null, rfNumber=[19], rfOrder=35, authorNames=熊明福, 李家辉, 熊捷繁, journalName=科学技术与工程, refType=null, unstructuredReference=熊明福, 李家辉, 熊捷繁, 等. 基于光流重构和深度姿态特征的行人跌倒检测方法[J]. 科学技术与工程, 2022, 22(35): 15688-15696., articleTitle=基于光流重构和深度姿态特征的行人跌倒检测方法, refAbstract=null), Reference(id=1203787157782635083, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=35, pageStart=15688, pageEnd=15696, url=null, language=null, rfNumber=[19], rfOrder=36, authorNames=Xiong Mingfu, Li Jiahui, Xiong Jiefan, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Xiong Mingfu, Li Jiahui, Xiong Jiefan, et al. Pedestrian fall detection method based on optical flow and posture network[J]. Science Technology and Engineering, 2022, 22(35): 15688-15696., articleTitle=Pedestrian fall detection method based on optical flow and posture network, refAbstract=null)], funds=[Fund(id=1203787150421631972, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, awardId=U21A20516, language=CN, fundingSource=国家自然科学基金(U21A20516), fundOrder=null, country=null), Fund(id=1203787150564238318, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, awardId=2023YJ129, language=CN, fundingSource=中国铁道科学研究院重点基金(2023YJ129), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1203787143043850852, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, xref=null, ext=[AuthorCompanyExt(id=1203787143048045158, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, companyId=1203787143043850852, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Institute of Electronic Computing Technology, China Academy of Railway Sciences, Beijing 100081, China), AuthorCompanyExt(id=1203787143056433767, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, companyId=1203787143043850852, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081)])], figs=[ArticleFig(id=1203787147322041125, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=EN, label=Fig.1, caption=System architecture diagram, figureFileSmall=TFAdq12UrQthLfbK3Gd1Pw==, figureFileBig=KvQkSWOSQs0uXZLmt3uGQA==, tableContent=null), ArticleFig(id=1203787147410121518, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=CN, label=图1, caption=系统架构图, figureFileSmall=TFAdq12UrQthLfbK3Gd1Pw==, figureFileBig=KvQkSWOSQs0uXZLmt3uGQA==, tableContent=null), ArticleFig(id=1203787147573699385, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=EN, label=Fig.2, caption=MobileNet model network architecture, figureFileSmall=XMlcxqGVdOYYS9GT6n8PVg==, figureFileBig=9ITB3lViM8SGH+hVI0GXQw==, tableContent=null), ArticleFig(id=1203787147733082949, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=CN, label=图2, caption=MobileNet模型网络结构, figureFileSmall=XMlcxqGVdOYYS9GT6n8PVg==, figureFileBig=9ITB3lViM8SGH+hVI0GXQw==, tableContent=null), ArticleFig(id=1203787147816969039, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=EN, label=Fig.3, caption=Train door opening and closing status, figureFileSmall=44aJkiVJ9Lm2IvWn7CNtLQ==, figureFileBig=D/b/IJ7O6g5XW6MJkJkV8A==, tableContent=null), ArticleFig(id=1203787147917632342, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=CN, label=图3, caption=列车开门和关门状态, figureFileSmall=44aJkiVJ9Lm2IvWn7CNtLQ==, figureFileBig=D/b/IJ7O6g5XW6MJkJkV8A==, tableContent=null), ArticleFig(id=1203787148022489949, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=EN, label=Table 1, caption=

Dataset information summary

, figureFileSmall=null, figureFileBig=null, tableContent=
列车车型 开、关门视频资料数量
CR400AF 11
CR400AF-B 15
CR400BF 22
CR400BF-A 11
CR400BF-B 21
CRH380A 17
CRH380B 19
CRH380BG 18
CRH380BL 25
CRH380CL 12
合计 171
), ArticleFig(id=1203787148106376035, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=CN, label=表1, caption=

数据集信息汇总

, figureFileSmall=null, figureFileBig=null, tableContent=
列车车型 开、关门视频资料数量
CR400AF 11
CR400AF-B 15
CR400BF 22
CR400BF-A 11
CR400BF-B 21
CRH380A 17
CRH380B 19
CRH380BG 18
CRH380BL 25
CRH380CL 12
合计 171
), ArticleFig(id=1203787149310141299, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=EN, label=Table 2, caption=

Confusion matrix of prediction results

, figureFileSmall=null, figureFileBig=null, tableContent=
真实类别\预测类别 开门 关门
开门 TP FN
关门 FP TN
), ArticleFig(id=1203787149431776124, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=CN, label=表2, caption=

预测结果的混淆矩阵

, figureFileSmall=null, figureFileBig=null, tableContent=
真实类别\预测类别 开门 关门
开门 TP FN
关门 FP TN
), ArticleFig(id=1203787149565993867, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=EN, label=Table 3, caption=

The Influence of learning rate on the model

, figureFileSmall=null, figureFileBig=null, tableContent=
学习率 识别准确率/%
开门 关门
0.01 95.3 96.4
0.001 96.4 97.1
0.000 1 97.8 98.2
), ArticleFig(id=1203787149696017306, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=CN, label=表3, caption=

学习率对模型的影响

, figureFileSmall=null, figureFileBig=null, tableContent=
学习率 识别准确率/%
开门 关门
0.01 95.3 96.4
0.001 96.4 97.1
0.000 1 97.8 98.2
), ArticleFig(id=1203787149842817960, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=EN, label=Table 4, caption=

The influence of width multipliers onthe model

, figureFileSmall=null, figureFileBig=null, tableContent=
宽度乘子 识别准确率/% 平均识别
时间/ms
开门 关门
1 97.8 98.5 362
0.75 96.7 97.4 319
0.5 96.0 96.7 283
), ArticleFig(id=1203787149985424308, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=CN, label=表4, caption=

宽度乘子对模型的影响

, figureFileSmall=null, figureFileBig=null, tableContent=
宽度乘子 识别准确率/% 平均识别
时间/ms
开门 关门
1 97.8 98.5 362
0.75 96.7 97.4 319
0.5 96.0 96.7 283
), ArticleFig(id=1203787150098670529, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=EN, label=Table 5, caption=

The influence of resolution multipliers on the model

, figureFileSmall=null, figureFileBig=null, tableContent=
分辨率乘子 识别准确率/% 平均识别
时间/ms
开门 关门
1 97.4 97.8 362
0.75 96.0 96.4 326
0.5 94.9 95.3 294
), ArticleFig(id=1203787150199333835, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1203753463671927446, language=CN, label=表5, caption=

分辨率乘子对模型的影响

, figureFileSmall=null, figureFileBig=null, tableContent=
分辨率乘子 识别准确率/% 平均识别
时间/ms
开门 关门
1 97.4 97.8 362
0.75 96.0 96.4 326
0.5 94.9 95.3 294
)], attaches=null, journal=Journal(id=1146119176004939786, delFlag=0, nameCn=科学技术与工程, nameEn=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, issn=1671-1815, eissn=, cn=11-4688/T, coden=null, periodic=4, language=CN, oaType=是, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=UKU/O7GSka5polgCTkbIIw==, journalPrice=null, startedYear=null, abbrevIsoEn=Sci Technol Eng, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754445529766, createdBy=null, updatedBy=13701087609, firstLetterCn=S, firstLetterEn=S, subjectCode=Natural Sciences, subjectName=自然科学, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=UKU/O7GSka5polgCTkbIIw==, picEn=5hwlULoNwcbj3xUmVi9MAQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1159791870395564357, language=CN, name=科学技术与工程, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529793, updatedTime=1754445529793, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.stae.com.cn/jsygc/site/menus/20090429150146001, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1159791870441701702, language=EN, name=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529804, updatedTime=1754445529804, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1146123166801305609, websiteList=[Website(id=1148243202391400884, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kxjsygc/CN, language=CN, createTime=1751692112777, createBy=18614031015, updateTime=1753520965431, updateBy=18614031015, name=科学技术与工程-中文站点, tplId=1146099689490845704, title=科学技术与工程, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148622798802673703, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=articleTextType, value=kx, createTime=1751782615614, updateTime=1751782615614, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798781702180, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=banner, value=null, createTime=1751782615609, updateTime=1751782615609, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798769119267, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1751782615606, updateTime=1751782615606, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798794285094, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751782615612, updateTime=1751782615612, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798790090789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751782615611, updateTime=1751782615611, creator=18614031015, updator=18614031015)]), Website(id=1155914124811976731, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kxjsygc/EN, language=EN, createTime=1753521003206, createBy=18614031015, updateTime=1753521003206, updateBy=18614031015, name=科学技术与工程-英文站点, tplId=1146101810881728533, title=Science Technology and Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155914371227308235, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=articleTextType, value=kx, createTime=1753521061952, updateTime=1753521061952, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371210531016, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=banner, value=null, createTime=1753521061947, updateTime=1753521061947, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371202142407, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1753521061945, updateTime=1753521061945, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371223113930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753521061950, updateTime=1753521061950, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371218919625, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753521061949, updateTime=1753521061949, creator=18614031015, updator=18614031015)])], journalTitle=科学技术与工程, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=Science Technology and Engineering, journalPhotoCn=UKU/O7GSka5polgCTkbIIw==, journalPhotoEn=5hwlULoNwcbj3xUmVi9MAQ==, journalFirstLetter=S, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2402124, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2402124, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2402124, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2402124, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于机器视觉的高速铁路站车联动控制技术
收藏切换
PDF下载
李帅 , 王志飞 * , 李樊 , 杜呈欣 , 王浩东 , 杨博璇
科学技术与工程 | 论文·交通运输 2025,25(2): 773-779
收起
收藏切换
科学技术与工程 | 论文·交通运输 2025, 25(2): 773-779
基于机器视觉的高速铁路站车联动控制技术
全屏
李帅 , 王志飞* , 李樊, 杜呈欣, 王浩东, 杨博璇
作者信息
  • 中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081
  • 李帅(1989—),男,汉族,河南郑州人,硕士,助理研究员。研究方向:控制工程。E-mail:

通讯作者:

* 王志飞(1981—),男,汉族,内蒙古鄂尔多斯人,博士,研究员。研究方向:交通信息工程与控制。E-mail:
High Speed Railway Station Car Linkage Control Technology Based on Machine Vision
Shuai LI , Zhi-fei WANG* , Fan LI, Cheng-xin DU, Hao-dong WANG, Bo-xuan YANG
Affiliations
  • Institute of Electronic Computing Technology, China Academy of Railway Sciences, Beijing 100081, China
出版时间: 2025-01-18 doi: 10.12404/j.issn.1671-1815.2402124
文章导航
收藏切换

为高效识别列车车门的开关状态,并据此控制站台门的同步开关,提出了一种基于轻量级MobileNet网络和机器视觉的图像识别方法,实现高速铁路站台门与列车门的联动控制。在北京南站收集大量列车车门的图像资料,经过预处理后作为模型训练和测试的数据集,再利用二元交叉熵损失函数和Adam优化算法对构建的网络进行训练和优化,最终实现对车门状态的高效精准识别。验证结果表明:对列车开关门动作的识别准确率达到95%以上,识别时间控制在400 ms以内,均能满足当前行业应用需求,极大提高站台门系统的自动化和智能化水平。

卷积神经网络  /  机器视觉  /  站台门系统  /  联动控制

To efficiently identify the opening and closing status of train doors and control the synchronous opening and closing of platform doors, a lightweight MobileNet network and machine vision based image recognition method was proposed to achieve linkage control between high-speed railway platform doors and train doors. A large dataset of train door images was collected from Beijing South Station and preprocessed to serve as the training and testing dataset for the model. The constructed network was trained and optimized using a binary cross-entropy loss function and the Adam optimization algorithm to achieve efficient and accurate recognition of door status. Validation results demonstrate an accuracy rate of over 95% in recognizing train door actions, with recognition time kept within 400 milliseconds. These results meet the current industry application requirements and greatly enhance the automation and intelligence level of the platform door system.

convolutional neural network  /  machine vision  /  platform door system  /  linkage control
李帅, 王志飞, 李樊, 杜呈欣, 王浩东, 杨博璇. 基于机器视觉的高速铁路站车联动控制技术. 科学技术与工程, 2025 , 25 (2) : 773 -779 . DOI: 10.12404/j.issn.1671-1815.2402124
Shuai LI, Zhi-fei WANG, Fan LI, Cheng-xin DU, Hao-dong WANG, Bo-xuan YANG. High Speed Railway Station Car Linkage Control Technology Based on Machine Vision[J]. Science Technology and Engineering, 2025 , 25 (2) : 773 -779 . DOI: 10.12404/j.issn.1671-1815.2402124
目前,站台门系统已在高速铁路中广泛应用,但受限于行业发展水平和运营组织模式,大多数高铁车站的站台门系统仍依赖站务人员手动操作,仅有少部分线路或车站通过列车控制系统(China train control system,CTCS)和列车自动驾驶(automatic train operation,ATO)系统实现站台门和列车车门的联动控制[1]。王爱武等[2]提出了一种基于C2+ATO的站台门与列车门联动控制方案;刘磊等[3]提出了一种站台门与C3+ATO系统控制接口的方案,实现站台门与列车门的联动控制。然而并非每条线路和每组列车都配备ATO设备,若要通过ATO系统实现全行业内站台门与列车车门的联动控制,则需在车站和列车均配备复杂昂贵的ATO设备,技术难度较大、成本极高,短期内可行性较低[4]。因此,为实现站台门与列车车门的联动,必须利用其他技术手段来判断列车车门的开关状态,并根据这一信息控制站台门的同步开关。郜洪民等[5]基于多传感器融合技术探测列车进站和车门的开、关状态,并以此实现站台门和列车门的联动控制,但其探测信号来源多样,系统架构较为复杂。
随着机器视觉技术不断的演进和成熟,其潜在的应用领域也日益扩大,特别是在人脸识别、物体检测、车牌识别以及缺陷检测等领域[6-9],机器视觉技术凭借其简洁的网络架构、卓越的识别精度及显著的成本效益比,赢得了广泛的认可和成功应用,这为如何构建高效、稳定且经济的模型,来解决高速铁路站台门与列车间的车地联动控制问题,提供了宝贵的经验与启发。艾鑫等[10]通过VGG16网络提取特征并融合多尺度的特征图,实现铁路货车车号的定位和识别。董咚等[11]利用YOLO网络模型结构和货车车号字符特点,实现对车号位置的准确检测。朱俊霖等[12]利用工业相机拍摄城轨列车侧面车号,并基于VGG16网络模型对车号字符进行识别,实现了城轨列车车号的获取。
在轨道交通领域,机器视觉的相关研究多集中在货车作业自动化,对客运列车的识别多停留在为列车检测和管理提供数据支持的层面,基于机器视觉实现站台门与列车门的联动控制尚无成功应用的先例。因此,现提出一种利用轻量级MobileNet架构为基础的卷积神经网络(convolutional neural network,CNN)解决方案。得益于MobileNet架构的轻量化设计,该方案在不牺牲精度的同时大幅减少模型的复杂度和计算量,利用CNN强大的特征提取能力,采用先进的机器视觉技术,准确识别列车的门开启和关闭动作,以实现站台门与列车车门的智能联动控制。
图1所示,本文设计的站台门控制系统由图像识别单元、主控系统、辅助管理系统和门控单元4个主要部分构成。
图像识别单元包含图像采集设备和前端智能识别终端,两者共同完成图像的采集和列车开关门动作的识别。前端智能识别终端与图像采集设备的集成,能够减少数据传输需要的时间,提升系统响应速度。
主控系统由两个核心控制器、I/O接口和人机交互界面组成,负责整个系统的流程控制和逻辑运算。两个核心控制器互相冗余,共同组成“二取二”架构,降低系统误动作可能性,提升系统可靠性和安全性。
辅助管理系统由视频监控与存储服务器、系统管理服务器组成,负责系统配置管理、深度学习训练、系统运行信息、故障信息和图像信息的存储与备份。
门控单元接收主控系统的控制指令,负责控制对应站台门的开关门动作,同时采集站台门运行信息并上传到主控系统。
CNN的权值共享网络结构使其更类似于生物神经网络,能够减少网络的训练参数,使网络结构更简单、适应性更强。此外,CNN省略了特征提取和数据重建的过程,能够直接接收输入的图像,从而极大地提高了图像识别的效率[13-14]。相比于其他复杂的CNN架构,MobileNet结构更加轻量化,专为移动和嵌入式视觉应用而设计,能够在保持较高准确率的同时提供快速的处理速度和较低的延迟,非常适合本文所述的应用场景。
MobileNet模型基于深度可分离的卷积构建,深度可分离卷积将标准的卷积层分解为深度卷积层和1×1点卷积层。深度卷积层中的卷积核个数和输入通道数目相同,每个卷积核只与输入的一个通道进行卷积。点卷积则对深度卷积的输出进行线性组合[15]。通过这种分解,可以大幅降低模型的大小和所需的计算量。MobileNet的网络结构如图2所示。
本文研究采用MobileNetV2模型,它是MobileNetV1的改进版本,引入了线性瓶颈和逆残差结构。逆残差结构是MobileNetV2的核心,用于实现高效的特征提取并减少计算量和模型大小[16];线性瓶颈层能够保持特征的丰富性并避免梯度消失或爆炸。通过两种结构相互结合,MobileNetV2能够有效地学习特征,并在保持网络紧凑性和计算效率的同时,提供令人满意的性能。这使得MobileNetV2成为本文研究的理想选择。
(1)激活函数。逆残差模块的扩展卷积层后采用ReLU6激活函数,其计算公式为
ReLU(x)=min[6,max(0,x)]
ReLU6提供的输出值上限可以在一定程度上减少神经网络训练过程中的梯度爆炸问题,并在8位或更低位宽的量化中保持更好的数值稳定性,非常适用于在移动和嵌入式设备上的深度学习模型。
(2)损失函数。在预测任务中,预测值不可能与实际值完全一致,两者之间的误差即为损失函数。列车车门只有开门和关门两种状态,神经网络只会输出2种互斥的结果预测,属于典型的二分类问题,故本文研究选用二元交叉熵损失函数[17],它在处理二分类问题时非常常用,与其他损失函数相比,其在参数更新过程中的效率不受激活函数导数的影响,因此可以持续地实现较快的收敛速度,其计算公式为
L(y, y ˙)=- 1 N i = 1 N[yiln y ˙ i+(1-yi)ln(1- y ˙ i)]
式(2)中:L为损失函数;N为样本的数量;yi为第i个样本的真实标签(0或1); y ˙ i为第i个样本的预测概率。
(3)SIGMOD函数。在本文所描述的应用场景中,MobileNetV2模型用于识别列车的开门和关门动作。针对这个二分类问题,Sigmoid激活函数是最适合作为最后输出层的激活函数。它可以将实数输入映射到(0,1)的范围内,对应着概率值的输出,其计算公式为
f(x)= 1 1 + e - x
式(3)中:x为输入值;f(x)为对应的激活函数输出。
通过使用Sigmoid激活函数,可以将网络的输出转化为列车开门和关门状态的概率值,对列车车门状态进行识别。
本文研究选择了TensorFlow深度学习框架来进行MobileNetV2模型的构建、训练、优化和测试。TensorFlow以其良好的灵活性、可扩展性和易用性等优点,已成为机器学习领域的首选工具[18]
为了使构建的MobileNetV2模型能够达到理想的效果,首先对现场采集的列车图像信息进行预处理,将其作为模型的输入,随后使用TensorFlow对构建好的网络模型进行训练,训练过程中,采用了Adam优化算法,它是一种自适应学习率的优化算法,能够对网络参数进行细致的调整和优化。通过反向传播和梯度下降,模型可以不断更新和调整参数,使其逐渐收敛到一个较优的状态。
通过选择TensorFlow框架、预处理图像数据、使用Adam优化算法,并进行持续的训练和优化,本文研究致力于构建一个高效且精确的MobileNetV2模型,以满足实际应用中的识别需求。
在行业内,利用机器视觉实现列车开关门动作识别的研究尚属首次,目前缺乏现成的网络数据集支持。因此,在北京南站现场采集列车开、关门图像资料,其详细信息如表1所示。
表1可知,所采集的数据集涵盖了CRH400系列和CRH380系列的10种不同车型,共计171份。由于图像采集设备在实际应用场景下的安装位置无法正对列车车门,为了尽可能地反映真实工况下的情况,并验证该方法的有效性和系统的识别效率,在数据采集过程中模拟实际应用场景,从不同的倾斜角度拍摄了图像资料。
由于数据集数量较少,在处理数据时,研究团队采取了数据扩充的方法,将采集到的图像数据进行了0°~5°的随机旋转处理[18]。这样可以增加数据集的多样性,提高模型的泛化能力和鲁棒性。其计算公式为
x = x 0 c o s θ + y 0 s i n θ y = - x 0 s i n θ + y 0 c o s θ
式(4)中:(x0,y0)为旋转前的点;(x,y)为旋转后的点;θ为旋转的角度。
通过旋转进行数据扩充,每份数据旋转3次,即可得到513份扩充数据;将原有数据和扩充后的数据(共计684份)合并后随机分为2份,按照6∶4的比例划分为训练集和测试集,构成模型训练和测试的数据集。
为了保证模型效果和计算效率,本文研究选择224×224作为MobileNetV2网络的输入图像尺寸。为了避免图像的变形,在保持图像的纵横比例的前提下对图像进行缩放,必要时对图像进行裁剪,以适应网络的输入要求。
对训练数据集进行批处理后,将训练数据集转换为TesnsorFlow平台可读取的二进制存储格式,即TFRecord文件,该格式的文件可以更方便地进行数据的读取、处理和输入模型中进行训练,并与数据导入和预处理功能无缝集成。
合理的初始化策略可以帮助模型快速收敛并提高模型训练的最终性能。由于本文研究采用的数据集较小,通过预训练模型参数对MobileNetV2网络进行初始化,并通过Adam优化器对网络进行优化。
(1)学习率。考虑到使用的数据集较小,本文研究使用较小的学习率来避免过拟合,初始学习率设置在0.001,并使用Adam算法在训练过程中调整学习率。
(2)宽度乘子。宽度因子允许模型根据可用的计算资源调整其宽度。参数越小,模型训练越快,但精度损失越大,本文研究使用1作为初始训练参数来保障模型识别的准确度。
(3)分辨率乘子。分辨率乘子用来调整输入图像的分辨率,本文设为1,即使用标准分辨率224×224。
(4)批量大小。考虑到硬件性能和数据集大小,选择初始批量大小为64。这个大小既兼顾了训练效率,又不会造成过多的性能开销。
完成初始化后,选择Adam优化器对MobileNetV2网络进行优化,Adam优化器结合了RMSprop和Momentum两种优化算法的优点,受益于自适应学习率的特点,能够实现快速收敛、减少调参负担、具备更强的适应性,而且结构简单,在TensorFlow框架下非常容易实现,非常适合本文所述机器视觉识别任务。
图3所示,受限于图像识别单元的安装位置,采集的图像通常会出现一定程度的倾斜,导致目标区域的位置、角度发生变化。因此,本文研究提出了可旋转的候选框、多边形候选框[19]等建模方法,以提高候选框对真实目标的覆盖率并减少背景干扰。
为了精确捕捉列车在站台停靠后的开关门动作,在站台与列车车头、车尾以及车身中部的对应位置,分别架设一组图像识别单元。3组图像识别单元可以从不同角度和位置获取图像信息,并将识别结果同时上传到主控系统,主控系统会根据3组识别结果,对列车的开关门状态进行智能判断。
列车同侧车门只会同时开门或关门(不考虑列车门故障场景),故识别到其中一组列车门打开即可认为列车门全部打开。
在识别到列车开门动作的一瞬间,图像识别单元能够迅速地捕捉到开门动作,并及时将此信号传达给主控系统;同理,在乘客上下车结束,列车门准备关闭,图像识别单元检测列车关门动作后,将关门信息发送给主控系统,主控系统控制站台门随之打开或关闭,从而实现站台门与列车门的联动控制。
由于本文研究是针对列车的开门和关门两种状态进行预测,预测结果如表2所示的混淆矩阵。
识别准确率和识别速度是该应用场景下最重要的关键指标,高准确率可以确保系统能够提供准确的结果,从而提高用户体验和应用场景的可靠性;更短的识别速度可以提高应用的实时性和响应能力。在本文所示应用场景中直接关系到系统的安全性和可用性。
为了在提升识别效率的同时保持较高的识别准确率,可以通过调整学习率、宽度乘子和分辨率乘子的值来实现。
在实验设置中,将模型训练的迭代次数固定为100次,表3详尽地列出了在不同学习率下模型训练的结果,通过对比分析可知,学习率设置为0.000 1时,模型展现出了最优的性能,实现了98.2%的高准确率。因此在后续的实验分析中,将学习率设定为0.000 1,进一步验证模型的性能表现。
保持分辨率乘子为1不变,依次降低宽度乘子到0.75和0.5来测试网络的识别效果,以验证网络在不同参数下的识别效果。其具体识别准确率和平均识别时间如表4所示。
保持宽度因子为1不变,依次降低分辨率乘子到0.75和0.5来测试网络的识别效果,以验证网络在不同参数下的识别效果。分辨率乘子调整为0.75和0.5时,输入图像分辨率分别为160×160和128×128。对其识别准确率和识别时间进行测试,数据如表5所示。
通过表4表5可以看出,宽度乘子和分辨率乘子的降低能够影响模型的识别准确率,同时减少平均识别时间。网络模型能以较低的精度损失为代价大大减少网络的识别时间,并且保持着良好的特征提取能力。模型的识别准确率均超过95%,识别速度在400 ms以内。这样的准确度和响应速度完全符合与列车车门联动系统的实际需求,展现了本文方法的实用性和高效能。
考虑到站台门开关是涉及安全运行的重要动作,而其与列车门在时间上同步则更侧重于功能性而非安全性,故虽然较低的宽度乘子和分辨率乘子并未对识别准确率造成过高的影响,出于安全考虑,本系统仍选用较大的宽度乘子和分辨率乘子,尽可能提升系统的识别准确率。在这样的参数设置下,系统的平均时间仍保持在400 ms之内,并不会对站台门与列车门的联动造成过大影响。
为了确保乘客的安全乘坐,本系统在监测到列车到站停稳,且在预定时间内未检测到列车开门动作时,将自动发出警报,提示站务人员进行手动操作以打开站台门。此外,系统会在数据库中创建一条故障记录,以协助维护工作人员迅速定位系统的问题并进行优化。这一过程不仅增强了乘客的安全保障,同时提升了运营效率,确保了列车服务的连续性和可靠性。
通过神经网络和机器视觉实现列车开关门动作的自动识别,能够在不依赖信号系统前提下,实现站台门与列车门联动控制,兼顾方案的可行性和经济性。在实现功能的同时,保留站台门系统的降级处理功能,允许站务人员介入站台门的控制,以确保在出现异常情况时,站务人员可以直接介入对站台门的控制。为了进一步提高系统的性能,研究将继续探索如何优化神经网络模型,以提高系统的识别精度、处理速度,以及其在面对复杂环境干扰时的鲁棒性,这些改进将进一步增强系统的实用性,满足不断增长的安全和效率需求,为公共交通安全和效率提供坚实的技术支持。
  • 国家自然科学基金(U21A20516)
  • 中国铁道科学研究院重点基金(2023YJ129)
参考文献 引证文献
排序方式:
[1]
方凯, 史天运, 陈瑞凤, 等. 面向北京冬奥会的京张高速铁路旅客服务智能化成套关键技术研究[J]. 铁道运输与经济, 2022(9): 56-63.
Fang Kai, Shi Tianyun, Chen Ruifeng, et al. Research on key technologies for intelligent passenger service of Beijing Zhangjiakou high speed railway for the Beijing Winter Olympics[J]. Railway Transportation and Economy, 2022(9): 56-63.
[2]
王爱武, 黄海宾, 周梅, 等. CTCS2+ATO控制模式下的车地联动方案[J]. 电子技术与软件工程, 2018(22): 114-120.
Wang Aiwu, Huang Haibin, Zhou Mei, et al. Train ground linkage scheme under CTCS2+ATO control mode[J]. Electronic Technology and Software Engineering, 2018 (22): 114-120.
[3]
刘磊, 周超. C3+ATO系统与站台门的接口设计[J]. 电子技术与软件工程, 2018(19): 90-93.
Liu Lei, Zhou Chao. Interface design between C3+ATO system and platform doors[J]. Electronic Technology and Software Engineering, 2018 (19): 90-93.
[4]
范楷, 冯浩楠, 段宏伟, 等. 基于近距离通信的站台门控制新方案[J]. 城市轨道交通研究, 2020, 23(1): 310-313.
Fan Kai, Feng Haonan, Duan Hongwei, et al. A new scheme for platform door control based on close range communication[J]. Urban Rail Transit Research, 2020, 23 (1): 310-313.
[5]
郜洪民, 郑伟, 赵阳, 等. 基于多传感器融合技术的车门站台门联动控制系统及方法: CN110469230A[P]. 2019-11-19.
Gao Hongmin, Zheng Wei, Zhao Yang, et al. A door platform door linkage control system and method based on multi-sensor fusion technology: CN110469230A[P]. 2019-11-19.
[6]
李玥辰, 赵晓, 王若男, 等. 基于改进卷积神经网络的中药饮片图像识别[J]. 科学技术与工程, 2024, 24(9): 3596-3604.
Li Yuechen, Zhao Xiao, Wang Ruonan, et al. Image recognition of chinese herbal slices based on improved convolution neural network[J]. Science Technology and Engineering, 2024, 24(9): 3596-3604.
[7]
牛嘉兴, 高利鹏, 张露, 等. 基于SSD算法的戴口罩人脸识别方法研究[J]. 计算机仿真, 2022(8): 204-207.
Niu Jiaxing, Gao Lipeng, Zhang Lu, et al. Research on mask wearing face recognition method based on SSD algorithm[J]. Computer Simulation, 2022(8): 204-207.
[8]
刘小宇, 陈怀新, 刘壁源, 等. 自适应置信度阈值的非限制场景车牌检测算法[J]. 计算机应用, 2023, 43(1): 67-73.
Liu Xiaoyu, Chen Huaixin, Liu Biyuan, et al. Adaptive confidence threshold non restricted scene license plate detection algorithm[J]. Computer Applications, 2023, 43 (1): 67-73.
[9]
宋泽冈, 刘艳莉, 张长兴. 基于机器视觉的桥梁裂缝检测应用及发展综述[J]. 科学技术与工程, 2023, 23(30): 12796-12805.
Song Zegang, Liu Yanli, Zhang Changxing. Review on application and development of bridge crack detection based on machine vision[J]. Science Technology and Engineering, 2023, 23(30): 12796-12805.
[10]
艾鑫, 邹琪, 罗常津. 面向精确定位的列车车号文本定位与识别[J]. 计算机辅助设计与图形学学报, 2020, 32(11): 1863-1870.
Ai Xin, Zou Qi, Luo Changjin. Train number text positioning and recognition for precise positioning[J]. Journal of Computer Aided Design and Graphics, 2020, 32(11): 1863-1870.
[11]
董咚, 王鹏月. 铁路货车车号图像识别系统的算法研究[J]. 现代科学仪器, 2022(1): 180-184.
Dong Dong, Wang Pengyue. Research on algorithm for image recognition system of railway freight car number[J]. Modern Scienti-fic Instruments, 2022(1): 180-184.
[12]
朱俊霖, 段钰, 滕凯, 等. 基于图像处理的城轨列车车号识别系统[J]. 铁路计算机应用, 2022, 31(9): 20-24.
Zhu Junlin, Duan Yu, Teng Kai, et al. Urban rail train number recognition system based on image processing[J]. Railway Computer Applications, 2022, 31(9): 20-24.
[13]
张驰, 郭媛, 黎明. 人工神经网络模型发展及应用综述[J]. 计算机工程与应用, 2021, 57(11): 57-69.
Zhang Chi, Guo Yuan, Li Ming. Dawn overview of the development and application of artificial neural network models[J]. Computer Engineering and Applications, 2021, 57(11): 57-69.
[14]
赵正阳, 吴艳华, 程智博, 等. 基于3D卷积神经网络的高铁轨道质量指数预测方法[J]. 铁路计算机应用, 2020(12): 7-11.
Zhao Zhengyang, Wu Yanhua, Cheng Zhibo, et al. Prediction method of high-speed rail track quality index based on 3D convolutional neural network[J]. Railway Computer Application, 2020(12): 7-11.
[15]
郭松, 尹明臣. 基于MobileNetV2的自然场景口罩检测算法[J]. 传感器与微系统, 2023, 42(9): 137-140.
Guo Song, Yin Mingchen. Natural scene mask detection algorithm based on MobileNetV2[J]. Sensors and Microsystems, 2023, 42 (9): 137-140.
[16]
Padmavathi P, Harikiran J. Wireless capsule endoscopy infected images detection and classification using MobileNetV2-BiLSTM model[J]. International Journal of Image and Graphics, 2023, 23(5): 118-122.
[17]
冯文彬, 厉舒南, 田昊, 等. 基于融合边缘优化的煤矿图像语义分割方法[J]. 煤矿安全, 2022, 53(2): 136-141.
Feng Wenbin, Li Shunan, Tian Hao, et al. A semantic segmentation method for coal mine images based on fusion edge optimization[J]. Coal Mine Safety, 2022, 53 (2): 136-141.
[18]
王程远, 赵俊莉, 计晓斐, 等. 基于改进的VGG网络的烟柜状态识别模型[J]. 计算机工程与设计, 2023, 44(6): 1796-1803.
Wang Chengyuan, Zhao Junli, Ji Xiaofei, et al. A smoke cabinet state recognition model based on an improved VGG network[J]. Computer Engineering and Design, 2023, 44(6): 1796-1803.
[19]
熊明福, 李家辉, 熊捷繁, 等. 基于光流重构和深度姿态特征的行人跌倒检测方法[J]. 科学技术与工程, 2022, 22(35): 15688-15696.
Xiong Mingfu, Li Jiahui, Xiong Jiefan, et al. Pedestrian fall detection method based on optical flow and posture network[J]. Science Technology and Engineering, 2022, 22(35): 15688-15696.
2025年第25卷第2期
PDF下载
305
130
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2402124
  • 接收时间:2024-03-25
  • 首发时间:2025-12-05
  • 出版时间:2025-01-18
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-03-25
  • 修回日期:2024-11-01
基金
国家自然科学基金(U21A20516)
中国铁道科学研究院重点基金(2023YJ129)
作者信息
    中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081

通讯作者:

* 王志飞(1981—),男,汉族,内蒙古鄂尔多斯人,博士,研究员。研究方向:交通信息工程与控制。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2402124
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