Article(id=1200070649684263641, tenantId=1146029695717560320, journalId=1189918454225211397, issueId=1200070646895051378, articleNumber=null, orderNo=null, doi=10.20104/j.cnki.1674-6546.20240167, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=null, receivedDateStr=null, revisedDate=1717948800000, revisedDateStr=2024-06-10, acceptedDate=null, acceptedDateStr=null, onlineDate=1764048739119, onlineDateStr=2025-11-25, pubDate=1723651200000, pubDateStr=2024-08-15, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1764048739119, onlineIssueDateStr=2025-11-25, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1764048739119, creator=13701087609, updateTime=1764048739119, updator=13701087609, issue=Issue{id=1200070646895051378, tenantId=1146029695717560320, journalId=1189918454225211397, year='2024', volume='', issue='8', pageStart='1', pageEnd='48', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1764048738454, creator=13701087609, updateTime=1764049350066, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1200073212257203051, tenantId=1146029695717560320, journalId=1189918454225211397, issueId=1200070646895051378, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1200073212257203052, tenantId=1146029695717560320, journalId=1189918454225211397, issueId=1200070646895051378, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=36, endPage=41, ext={EN=ArticleExt(id=1200070650699285231, articleId=1200070649684263641, tenantId=1146029695717560320, journalId=1189918454225211397, language=EN, title=Study on the Dynamic Detection Algorithm for Risk Targets in the Blind Spot of Right-Turning Dump Trucks, columnId=1200070647679386243, journalTitle=Automotive Engineer, columnName=Special Issue on Intelligent Vehicle Environmental Perception and Target Detection Technology, runingTitle=null, highlight=null, articleAbstract=

To address the issue of extensive blind spots during right turns due to the oversized nature of dump trucks, this paper proposes a dynamic detection algorithm for risk targets in the right-turn blind spots of dump trucks. The algorithm improves the YOLOv8 model by enhancing the C2f module and lossing calculation module to refine the model’s detection accuracy. Additionally, four position threshold lines are preset in the blind spots, the risk warning module of the blind spots of the dump truck is added, and the auxiliary driving system of the blind spots of the dump truck is established. The results indicate that the proposed dynamic detection algorithm can recognize various types of targets, including cars, trucks, buses, pedestrians and electric bicycles, with a mean Average Precision (mAP50) of 0.87 at a 50% intersection over union threshold for all categories of targets. The right-turn blind spots assisted driving system of the dump truck can make different degrees of early warning according to the position of the risk target box in the image.

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, authorCompany=null, fund=null, authors=null, authorsList=Penglin He, Zhifang Chen, Chang Wang), CN=ArticleExt(id=1200070652624470896, articleId=1200070649684263641, tenantId=1146029695717560320, journalId=1189918454225211397, language=CN, title=自卸车右转盲区风险目标动态检测算法研究, columnId=1200070647847158411, journalTitle=汽车工程师, columnName=智能车辆环境感知与目标检测技术专刊, runingTitle=null, highlight=null, articleAbstract=

针对自卸车外形尺寸大导致车辆右转弯时存在较大视野盲区的问题,提出一种自卸车右转盲区风险目标动态检测算法,该算法利用YOLOv8模型的C2f模块和损失计算模块,提高了模型的检测精确率。同时,在盲区中预设4条位置阈值线,增加盲区风险预警模块,建立了自卸车右转盲区辅助驾驶系统。结果表明:所提出的风险目标动态检测算法能够识别小型乘用车、载货汽车、公交车、行人和电动自行车等多种类型的目标,且所有类别目标的50%交并比阈值下的平均精度均值(mAP50)为0.87;自卸车右转盲区辅助驾驶系统能够根据图像中风险目标框的位置进行不同程度的预警。

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=+VQphwNsSeuIqrzNjSl/jQ==, magXml=TpUb7oYhxbJYs+xmNup9YQ==, pdfUrl=null, pdf=us/lD2tUrqZ7ZI37aSHwaw==, pdfFileSize=22391853, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=EY/knJxgKQP93H9MZfH/pA==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=w+ZQTM7xnD2rrm+LqPxDlw==, mapNumber=null, authorCompany=null, fund=null, authors=null, authorsList=贺鹏麟, 陈志芳, 王畅)}, authors=[Author(id=1200070653266199463, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1200070653354279859, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, authorId=1200070653266199463, language=EN, stringName=Penglin He, firstName=Penglin, middleName=null, lastName=He, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 Shenzhen Smart Chelian Technology Co., Ltd., Shenzhen 518100, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1200070653471720378, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, authorId=1200070653266199463, language=CN, stringName=贺鹏麟, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 深圳智慧车联科技有限公司, 深圳 518100, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1200070652867740548, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, xref=1, ext=[AuthorCompanyExt(id=1200070652876129156, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, companyId=1200070652867740548, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Shenzhen Smart Chelian Technology Co., Ltd., Shenzhen 518100), AuthorCompanyExt(id=1200070652884517765, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, companyId=1200070652867740548, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 深圳智慧车联科技有限公司, 深圳 518100)])]), Author(id=1200070653576577985, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, 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=1200070653702407116, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, authorId=1200070653576577985, language=EN, stringName=Zhifang Chen, firstName=Zhifang, middleName=null, lastName=Chen, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 Zhejiang Haikang Technology Co., Ltd., Hangzhou 310000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1200070653840819162, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, authorId=1200070653576577985, language=CN, stringName=陈志芳, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 浙江海康科技有限公司, 杭州 310000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1200070652972598158, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, xref=2, ext=[AuthorCompanyExt(id=1200070653006152593, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, companyId=1200070652972598158, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Zhejiang Haikang Technology Co., Ltd., Hangzhou 310000), AuthorCompanyExt(id=1200070653014541202, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, companyId=1200070652972598158, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 浙江海康科技有限公司, 杭州 310000)])]), Author(id=1200070655015224290, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, 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=1200070655182996461, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, authorId=1200070655015224290, language=EN, stringName=Chang Wang, firstName=Chang, middleName=null, lastName=Wang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3 Chang’an University, Xi’an 710064, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1200070655321408504, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, authorId=1200070655015224290, language=CN, stringName=王畅, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3 长安大学, 西安 710064, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1200070653119398809, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, xref=3, ext=[AuthorCompanyExt(id=1200070653127787419, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, companyId=1200070653119398809, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 Chang’an University, Xi’an 710064), AuthorCompanyExt(id=1200070653144564636, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, companyId=1200070653119398809, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 长安大学, 西安 710064)])])], keywords=[Keyword(id=1200070655543705602, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=EN, orderNo=1, keyword=Dump truck blind area), Keyword(id=1200070655631785990, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=EN, orderNo=2, keyword=Object detection), Keyword(id=1200070655820529682, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=EN, orderNo=3, keyword=YOLOv8), Keyword(id=1200070655937970200, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=EN, orderNo=4, keyword=Warning system), Keyword(id=1200070656063799332, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=CN, orderNo=1, keyword=自卸车盲区), Keyword(id=1200070656160268331, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=CN, orderNo=2, keyword=目标检测), Keyword(id=1200070656319651892, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=CN, orderNo=3, keyword=YOLOv8), Keyword(id=1200070656466452543, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=CN, orderNo=4, keyword=预警系统)], refs=[Reference(id=1200070659893199109, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=中华人民共和国工业和信息化部, journalName=机动车辆间接视野装置性能和安装要求: GB 15084—2022, refType=null, unstructuredReference=中华人民共和国工业和信息化部. 机动车辆间接视野装置性能和安装要求: GB 15084—2022[S]. 北京: 中国标准出版社, 2023., articleTitle=null, refAbstract=null), Reference(id=1200070660031611148, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=Ministry of Industry and Information Technology of the People’s Republic of China, journalName=Motor Vehicles—Device for Indirect Vision-Requirement of Performance and Installation: GB 15084—2022, refType=null, unstructuredReference=Ministry of Industry and Information Technology of the People’s Republic of China. Motor Vehicles—Device for Indirect Vision-Requirement of Performance and Installation: GB 15084—2022[S]. Beijing: Standards Press of China, 2023., articleTitle=null, refAbstract=null), Reference(id=1200070660140663059, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, doi=null, pmid=null, pmcid=null, year=2017, volume=135, issue=null, pageStart=353, pageEnd=365, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=LIU G R, ZHOU M Z, WANG L L, journalName=Optik, refType=null, unstructuredReference=LIU G R, ZHOU M Z, WANG L L, et al. A Blind Spot Detection and Warning System Based on Millimeter Wave Radar for Driver Assistance[J]. Optik, 2017, 135: 353-365., articleTitle=A Blind Spot Detection and Warning System Based on Millimeter Wave Radar for Driver Assistance, refAbstract=null), Reference(id=1200070660262297882, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, doi=null, pmid=null, pmcid=null, year=2018, volume=101, issue=null, pageStart=116, pageEnd=128, url=null, language=null, rfNumber=[3], rfOrder=3, authorNames=RA M, JUNG H G, SUHR J K, journalName=Expert Systems with Applications, refType=null, unstructuredReference=RA M, JUNG H G, SUHR J K, et al. Part-Based Vehicle Detection in Side-Rectilinear Images for Blind-Spot Detection[J]. Expert Systems with Applications, 2018, 101: 116-128., articleTitle=Part-Based Vehicle Detection in Side-Rectilinear Images for Blind-Spot Detection, refAbstract=null), Reference(id=1200070660350378275, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, doi=null, pmid=null, pmcid=null, year=2019, volume=8, issue=2, pageStart=233, pageEnd=null, url=null, language=null, rfNumber=[4], rfOrder=4, authorNames=ZHAO Y M, BAI L, LÜ Y C, journalName=Electronics, refType=null, unstructuredReference=ZHAO Y M, BAI L, Y C, et al. Camera-Based Blind Spot Detection with a General Purpose Lightweight Neural Network[J]. Electronics, 2019, 8(2): 233., articleTitle=Camera-Based Blind Spot Detection with a General Purpose Lightweight Neural Network, refAbstract=null), Reference(id=1200070660455235884, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, doi=null, pmid=null, pmcid=null, year=2019, volume=14, issue=7, pageStart=737, pageEnd=742, url=null, language=null, rfNumber=[5], rfOrder=5, authorNames=杨炜, 张志威, 周凯霞, journalName=中国科技论文, refType=null, unstructuredReference=杨炜, 张志威, 周凯霞, 等. 大型车辆右转弯盲区预警系统设计[J]. 中国科技论文, 2019, 14(7): 737-742., articleTitle=大型车辆右转弯盲区预警系统设计, refAbstract=null), Reference(id=1200070660560093492, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, doi=null, pmid=null, pmcid=null, year=2019, volume=14, issue=7, pageStart=737, pageEnd=742, url=null, language=null, rfNumber=[5], rfOrder=6, authorNames=YANG W, ZHANG Z W, ZHOU K X, journalName=China Sciencepaper, refType=null, unstructuredReference=YANG W, ZHANG Z W, ZHOU K X, et al. Design of Right Turn Blind Area Warning System for Large Vehicles[J]. China Sciencepaper, 2019, 14(7): 737-742., articleTitle=Design of Right Turn Blind Area Warning System for Large Vehicles, refAbstract=null), Reference(id=1200070660669145402, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=4, pageStart=65, pageEnd=69, url=null, language=null, rfNumber=[6], rfOrder=7, authorNames=高鲜萍, 卞学良, 成英, journalName=机械设计与制造, refType=null, unstructuredReference=高鲜萍, 卞学良, 成英, 等. 汽车视觉盲区动态目标智能检测算法[J]. 机械设计与制造, 2023(4): 65-69., articleTitle=汽车视觉盲区动态目标智能检测算法, refAbstract=null), Reference(id=1200070660811751747, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=4, pageStart=65, pageEnd=69, url=null, language=null, rfNumber=[6], rfOrder=8, authorNames=GAO X P, BIAN X L, CHENG Y, journalName=Machinery Design & Manufacture, refType=null, unstructuredReference=GAO X P, BIAN X L, CHENG Y, et al. Intelligent Detection Algorithm of Dynamic Target in Blind Spot[J]. Machinery Design & Manufacture, 2023(4): 65-69., articleTitle=Intelligent Detection Algorithm of Dynamic Target in Blind Spot, refAbstract=null), Reference(id=1200070660971135310, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, doi=null, pmid=null, pmcid=null, year=2023, volume=7, issue=6, pageStart=168, pageEnd=171, url=null, language=null, rfNumber=[7], rfOrder=9, authorNames=孟银阔, 刘江涛, 龚佑祥, journalName=现代信息科技, refType=null, unstructuredReference=孟银阔, 刘江涛, 龚佑祥, 等. 基于超声波的汽车盲区检测[J]. 现代信息科技, 2023, 7(6): 168-171., articleTitle=基于超声波的汽车盲区检测, refAbstract=null), Reference(id=1200070661109547346, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, doi=null, pmid=null, pmcid=null, year=2023, volume=7, issue=6, pageStart=168, pageEnd=171, url=null, language=null, rfNumber=[7], rfOrder=10, authorNames=MENG Y K, LIU J T, GONG Y X, journalName=Modern Information Technology, refType=null, unstructuredReference=MENG Y K, LIU J T, GONG Y X, et al. Automobile Blind Spot Detection Based on Ultrasound Wave[J]. Modern Information Technology, 2023, 7(6): 168-171., articleTitle=Automobile Blind Spot Detection Based on Ultrasound Wave, refAbstract=null), Reference(id=1200070661218599253, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, doi=null, pmid=null, pmcid=null, year=2023, volume=11, issue=7, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[8], rfOrder=11, authorNames=HUSSAIN M, journalName=Machines, refType=null, unstructuredReference=HUSSAIN M. YOLO-v1 to YOLO-v8, the Rise of YOLO and Its Complementary Nature Toward Digital Manufacturing and Industrial Defect Detection[J]. Machines, 2023, 11(7)., articleTitle=YOLO-v1 to YOLO-v8, the Rise of YOLO and Its Complementary Nature Toward Digital Manufacturing and Industrial Defect Detection, refAbstract=null), Reference(id=1200070661373788513, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, doi=null, pmid=null, pmcid=null, year=2022, volume=45, issue=3, pageStart=3139, pageEnd=3153, url=null, language=null, rfNumber=[9], rfOrder=12, authorNames=LI X, LÜ C Q, WANG W H, journalName=IEEE Transactions on Pattern Analysis and Machine Intelligence, refType=null, unstructuredReference=LI X, C Q, WANG W H, et al. Generalized Focal Loss: Towards Efficient Representation Learning for Dense Object Detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 45(3): 3139-3153., articleTitle=Generalized Focal Loss: Towards Efficient Representation Learning for Dense Object Detection, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1200070652867740548, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, xref=1, ext=[AuthorCompanyExt(id=1200070652876129156, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, companyId=1200070652867740548, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Shenzhen Smart Chelian Technology Co., Ltd., Shenzhen 518100), AuthorCompanyExt(id=1200070652884517765, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, companyId=1200070652867740548, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 深圳智慧车联科技有限公司, 深圳 518100)]), AuthorCompany(id=1200070652972598158, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, xref=2, ext=[AuthorCompanyExt(id=1200070653006152593, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, companyId=1200070652972598158, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Zhejiang Haikang Technology Co., Ltd., Hangzhou 310000), AuthorCompanyExt(id=1200070653014541202, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, companyId=1200070652972598158, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 浙江海康科技有限公司, 杭州 310000)]), AuthorCompany(id=1200070653119398809, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, xref=3, ext=[AuthorCompanyExt(id=1200070653127787419, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, companyId=1200070653119398809, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 Chang’an University, Xi’an 710064), AuthorCompanyExt(id=1200070653144564636, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, companyId=1200070653119398809, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 长安大学, 西安 710064)])], figs=[ArticleFig(id=1200070656625836111, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=EN, label=null, caption=null, figureFileSmall=YlfKkebVPn6RHTjyDVFJ+A==, figureFileBig=EY/knJxgKQP93H9MZfH/pA==, tableContent=null), ArticleFig(id=1200070656835551320, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=CN, label=图1, caption=检测算法及辅助驾驶系统, figureFileSmall=YlfKkebVPn6RHTjyDVFJ+A==, figureFileBig=EY/knJxgKQP93H9MZfH/pA==, tableContent=null), ArticleFig(id=1200070657112375405, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=EN, label=null, caption=null, figureFileSmall=tIXEPJqcvv7Cb/r5SZgbig==, figureFileBig=iI5NSHNy6WV//ABA71usNA==, tableContent=null), ArticleFig(id=1200070657213038710, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=CN, label=图2, caption=C2f模块, figureFileSmall=tIXEPJqcvv7Cb/r5SZgbig==, figureFileBig=iI5NSHNy6WV//ABA71usNA==, tableContent=null), ArticleFig(id=1200070657309507712, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=EN, label=null, caption=null, figureFileSmall=KY7SkhlxBbYcPOGGKioyVw==, figureFileBig=URTgPT7oFlU/PhzVQBWBPw==, tableContent=null), ArticleFig(id=1200070657397588107, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=CN, label=图3, caption=无锚点解耦头, figureFileSmall=KY7SkhlxBbYcPOGGKioyVw==, figureFileBig=URTgPT7oFlU/PhzVQBWBPw==, tableContent=null), ArticleFig(id=1200070657473085588, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=EN, label=null, caption=null, figureFileSmall=6paoNORV+0UUCLT/pQhfDQ==, figureFileBig=EnrTMr8M35soh58UXqNZIw==, tableContent=null), ArticleFig(id=1200070657569554591, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=CN, label=图4, caption=不同摄像头盲区图像位置阈值示意, figureFileSmall=6paoNORV+0UUCLT/pQhfDQ==, figureFileBig=EnrTMr8M35soh58UXqNZIw==, tableContent=null), ArticleFig(id=1200070657674412199, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=EN, label=null, caption=null, figureFileSmall=CzNUOuj/ul8eESDZ8ptwQw==, figureFileBig=qItaerkq3kiSy2fSDV1LPw==, tableContent=null), ArticleFig(id=1200070657808629936, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=CN, label=图5, caption=右后方盲区数据集, figureFileSmall=CzNUOuj/ul8eESDZ8ptwQw==, figureFileBig=qItaerkq3kiSy2fSDV1LPw==, tableContent=null), ArticleFig(id=1200070657934459067, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=EN, label=null, caption=null, figureFileSmall=gSF9FZraowQvURhR8WzTzg==, figureFileBig=0ImCx5GiiKatB6WQ4iIhQQ==, tableContent=null), ArticleFig(id=1200070658102231239, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=CN, label=图6, caption=YOLOv8网络模型在训练集和评估集上的学习过程, figureFileSmall=gSF9FZraowQvURhR8WzTzg==, figureFileBig=0ImCx5GiiKatB6WQ4iIhQQ==, tableContent=null), ArticleFig(id=1200070658177728718, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=EN, label=null, caption=null, figureFileSmall=HTrpW1TsQbeGlZvknlP2Cg==, figureFileBig=RYMASTpQPeB70+H15LiuUA==, tableContent=null), ArticleFig(id=1200070658274197716, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=CN, label=图7, caption=模型P-R曲线及mAP50指标变化情况, figureFileSmall=HTrpW1TsQbeGlZvknlP2Cg==, figureFileBig=RYMASTpQPeB70+H15LiuUA==, tableContent=null), ArticleFig(id=1200070658370666716, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=EN, label=null, caption=null, figureFileSmall=0ZXIYsX/EceEwHoiYnm72g==, figureFileBig=ujtAcanYS7nfrqx5ZflYwA==, tableContent=null), ArticleFig(id=1200070659461185763, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=CN, label=图8, caption=模型识别目标结果, figureFileSmall=0ZXIYsX/EceEwHoiYnm72g==, figureFileBig=ujtAcanYS7nfrqx5ZflYwA==, tableContent=null), ArticleFig(id=1200070659595403503, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
类别 P R mAP50AP
整体 0.877 0.833 0.870
小型乘用车 0.816 0.958 0.900
电动自行车 0.818 0.793 0.867
载货汽车 0.788 0.968 0.937
行人 0.885 0.800 0.831
公交车 0.981 0.942 0.991
自行车 0.976 0.538 0.691
), ArticleFig(id=1200070659704455418, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1200070649684263641, language=CN, label=表1, caption=

各类别检测精度

, figureFileSmall=null, figureFileBig=null, tableContent=
类别 P R mAP50AP
整体 0.877 0.833 0.870
小型乘用车 0.816 0.958 0.900
电动自行车 0.818 0.793 0.867
载货汽车 0.788 0.968 0.937
行人 0.885 0.800 0.831
公交车 0.981 0.942 0.991
自行车 0.976 0.538 0.691
)], attaches=null, journal=Journal(id=1189918244568731652, delFlag=0, nameCn=汽车工程师, nameEn=Automotive Engineer, nameHistory1=null, nameHistory2=null, issn=1674-6546, eissn=null, cn=22-1432/U, coden=null, periodic=0, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=+bJsKkKt/pjz9u6EwhnksQ==, journalPrice=null, startedYear=null, abbrevIsoEn=null, journalRemark=null, publicationField=null, createdTime=1761628217121, updatedTime=1761735708780, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=A, firstLetterEn=A, subjectCode=Engineering, subjectName=Engineering, subjectCodeEn=Engineering, subjectNameEn=null, picCn=+bJsKkKt/pjz9u6EwhnksQ==, picEn=O3Sn3tnYYrh/jm6emnnMWA==, jcr=null, cjcr=null, exts=[JournalExt(id=1190369097415233706, language=CN, name=汽车工程师, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1761735708812, updatedTime=1761735708812, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=1, submissionEditorUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=3, submissionReviewUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=2, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1190369097553645739, language=EN, name=Automotive Engineer, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1761735708845, updatedTime=1761735708845, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=1, submissionEditorUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=3, submissionReviewUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=2, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1189918454225211397, websiteList=[Website(id=1189918982430847716, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1189918454225211397, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/qcgcs/CN, language=CN, createTime=1761628393037, createBy=18614031015, updateTime=1761628422913, updateBy=18614031015, name=汽车工程师-中文, tplId=1146099689490845704, title=汽车工程师, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1189919800185917791, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=articleTextType, value=kx, createTime=1761628588005, updateTime=1761628588005, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800164946268, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=banner, value=null, createTime=1761628588000, updateTime=1761628588000, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800211083618, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=grayFlag, value=0, createTime=1761628588011, updateTime=1761628588011, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800156557659, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=logo, value=https://castjournals.cast.org.cn/joweb/qcgcs/CN/file/pic?fileId=yiZ96RYoYcnGnRMuWdmkWA==, createTime=1761628587998, updateTime=1761628587998, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800223666532, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=minRunFlag, value=0, createTime=1761628588014, updateTime=1761628588014, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800181723486, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/qcgcs/CN/file/pic, createTime=1761628588004, updateTime=1761628588004, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800215277923, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=silenceFlag, value=0, createTime=1761628588012, updateTime=1761628588012, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800173334877, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1761628588002, updateTime=1761628588002, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800194306400, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=themeColor, value=null, createTime=1761628588007, updateTime=1761628588007, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800202695009, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=themeStyle, value=null, createTime=1761628588009, updateTime=1761628588009, creator=18614031015, updator=18614031015)]), Website(id=1189918982527316711, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1189918454225211397, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/qcgcs/EN, language=EN, createTime=1761628393061, createBy=18614031015, updateTime=1761628543075, updateBy=18614031015, name=汽车工程师-英文, tplId=1146101810881728533, title=Automotive Engineer, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1189919837561352952, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=articleTextType, value=kx, createTime=1761628596916, updateTime=1761628596916, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837540381429, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=banner, value=null, createTime=1761628596911, updateTime=1761628596911, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837582324475, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=grayFlag, value=0, createTime=1761628596921, updateTime=1761628596921, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837527798516, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=logo, value=https://castjournals.cast.org.cn/joweb/qcgcs/EN/file/pic?fileId=yiZ96RYoYcnGnRMuWdmkWA==, createTime=1761628596908, updateTime=1761628596908, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837594907389, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=minRunFlag, value=0, createTime=1761628596924, updateTime=1761628596924, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837557158647, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/qcgcs/EN/file/pic, createTime=1761628596915, updateTime=1761628596915, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837586518780, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=silenceFlag, value=0, createTime=1761628596922, updateTime=1761628596922, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837548770038, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1761628596913, updateTime=1761628596913, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837569741561, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=themeColor, value=null, createTime=1761628596918, updateTime=1761628596918, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837573935866, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=themeStyle, value=null, createTime=1761628596919, updateTime=1761628596919, creator=18614031015, updator=18614031015)])], journalTitle=汽车工程师, weixinUrl=null, journalUrl=https://tjqc.cbpt.cnki.net/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Automotive Engineer, journalPhotoCn=+bJsKkKt/pjz9u6EwhnksQ==, journalPhotoEn=O3Sn3tnYYrh/jm6emnnMWA==, journalFirstLetter=A, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/qcgcs/CN/10.20104/j.cnki.1674-6546.20240167, detailUrlEn=https://castjournals.cast.org.cn/joweb/qcgcs/EN/10.20104/j.cnki.1674-6546.20240167, pdfUrlCn=https://castjournals.cast.org.cn/joweb/qcgcs/CN/PDF/10.20104/j.cnki.1674-6546.20240167, pdfUrlEn=https://castjournals.cast.org.cn/joweb/qcgcs/EN/PDF/10.20104/j.cnki.1674-6546.20240167, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
自卸车右转盲区风险目标动态检测算法研究
收藏切换
PDF下载
贺鹏麟 1 , 陈志芳 2 , 王畅 3
汽车工程师 | 智能车辆环境感知与目标检测技术专刊 2024,(8): 36-41
收起
收藏切换
汽车工程师 | 智能车辆环境感知与目标检测技术专刊 2024, (8): 36-41
自卸车右转盲区风险目标动态检测算法研究
全屏
贺鹏麟1, 陈志芳2, 王畅3
作者信息
  • 1 深圳智慧车联科技有限公司, 深圳 518100
  • 2 浙江海康科技有限公司, 杭州 310000
  • 3 长安大学, 西安 710064
Study on the Dynamic Detection Algorithm for Risk Targets in the Blind Spot of Right-Turning Dump Trucks
Penglin He1, Zhifang Chen2, Chang Wang3
Affiliations
  • 1 Shenzhen Smart Chelian Technology Co., Ltd., Shenzhen 518100
  • 2 Zhejiang Haikang Technology Co., Ltd., Hangzhou 310000
  • 3 Chang’an University, Xi’an 710064
出版时间: 2024-08-15 doi: 10.20104/j.cnki.1674-6546.20240167
文章导航
收藏切换

针对自卸车外形尺寸大导致车辆右转弯时存在较大视野盲区的问题,提出一种自卸车右转盲区风险目标动态检测算法,该算法利用YOLOv8模型的C2f模块和损失计算模块,提高了模型的检测精确率。同时,在盲区中预设4条位置阈值线,增加盲区风险预警模块,建立了自卸车右转盲区辅助驾驶系统。结果表明:所提出的风险目标动态检测算法能够识别小型乘用车、载货汽车、公交车、行人和电动自行车等多种类型的目标,且所有类别目标的50%交并比阈值下的平均精度均值(mAP50)为0.87;自卸车右转盲区辅助驾驶系统能够根据图像中风险目标框的位置进行不同程度的预警。

自卸车盲区  /  目标检测  /  YOLOv8  /  预警系统

To address the issue of extensive blind spots during right turns due to the oversized nature of dump trucks, this paper proposes a dynamic detection algorithm for risk targets in the right-turn blind spots of dump trucks. The algorithm improves the YOLOv8 model by enhancing the C2f module and lossing calculation module to refine the model’s detection accuracy. Additionally, four position threshold lines are preset in the blind spots, the risk warning module of the blind spots of the dump truck is added, and the auxiliary driving system of the blind spots of the dump truck is established. The results indicate that the proposed dynamic detection algorithm can recognize various types of targets, including cars, trucks, buses, pedestrians and electric bicycles, with a mean Average Precision (mAP50) of 0.87 at a 50% intersection over union threshold for all categories of targets. The right-turn blind spots assisted driving system of the dump truck can make different degrees of early warning according to the position of the risk target box in the image.

Dump truck blind area  /  Object detection  /  YOLOv8  /  Warning system
贺鹏麟, 陈志芳, 王畅. 自卸车右转盲区风险目标动态检测算法研究. 汽车工程师, 2024 , (8) : 36 -41 . DOI: 10.20104/j.cnki.1674-6546.20240167
Penglin He, Zhifang Chen, Chang Wang. Study on the Dynamic Detection Algorithm for Risk Targets in the Blind Spot of Right-Turning Dump Trucks[J]. Automotive Engineer, 2024 , (8) : 36 -41 . DOI: 10.20104/j.cnki.1674-6546.20240167
受整体尺寸影响,大型商用车右转弯时存在内轮差盲区,给盲区内突然出现的行人或小型机动车等小型目标造成一定的安全隐患。如何避免大型商用车右转时视觉盲区引起的交通事故,已成为当下交通运输安全领域急需解决的问题之一。
车辆盲区目标检测系统是解决上述问题的有效方法[1]。Liu等基于毫米波雷达检测车辆后方盲区中的运动目标,提出单元最大、最小和平均-恒定误报率(Cell Greatest, Smallest and Averaging Constant False-Alarm Rate,CGSA-CFAR)目标检测算法,通过调整功率检测阈值保持较高的检测精度[2]。Ra等引入侧直线图像生成模型实现视野盲区中汽车和摩托车检测,建立的盲区检测系统解决了目标在不同位置外观不一致的问题,该系统的精确率与召回率均较基于雷达的检测系统更高[3]。Zhao等基于车辆侧边盲区摄像机,提出结合深度可分离卷积(Depthwise Separable Convolution)、残差学习(Residual Learning)及挤压和激励(Squeeze and Excitation)的Sep-Res-SE模型,可以用更少的参数和操作实现相近的检测精度,相较于计算机视觉组(Visual Geometry Group,VGG)模型,该模型计算复杂度更低,目标检测精度近似一致[4]。杨炜等基于车辆右后侧的红外摄像头,提出一种应用于车辆右转工况的盲区检测系统,采用Inception-V3模型检测行人和自行车,验证结果表明,预警系统的平均准确率可达97%[5]。高鲜萍等考虑视觉盲区的动态特性,基于YOLOv3对算法的骨干网络进行了改进,通过引入密集连接增强了网络层间特征复用的能力,利用空间金字塔加强多尺度特征的融合,从而提升检测精度,使模型对15 m范围内的近距离目标的检测精度比基础的YOLOv3和单次多框检测器(Single Shot MultiBox Detector,SSD)等高出3%以上,在CPU上的推理时间缩短了约10%[6]。孟银阔等基于超声波模块,根据接收反射波的时间判断车辆侧后方有无车辆及障碍物,该检测系统具有简单、可靠和成本低的优点[7]
目前,国内外车辆盲区目标检测系统的目标物识别种类较为单一,且雷达等检测设备成本较高。同时,盲区目标检测作为一种实时嵌入式应用功能,不仅需要较高的识别精确率,还需要高推理速度和低计算复杂度。因此,本文提出一种自卸车右转盲区风险目标动态检测算法,基于YOLOv8模型,采用改进的C2f模块和损失(Loss)计算模块提高模型的检测精确率,并通过预设位置阈值实现右转盲区预警,最后在自卸车右转盲区数据集上进行训练与评估。
本文提出的自卸车右转盲区风险目标动态检测算法如图1所示,利用YOLOv8模型引入“C2f结构+解耦头+任务对齐学习(Task Alignment Learning,TAL)+分布焦点损失(Distribution Focal Loss,DFL)”等结构的特点,增强检测算法的识别精度,最后,在盲区中预设位置阈值线,实现分级预警。
YOLOv8支持图像分类、目标检测和图像分割等任务[8],YOLOv8目标检测模型通常采用“骨干(Backbone)-颈部(Neck)-头部(Head)”的网络结构。与YOLOv5相比,其对骨干网络和颈部进行了改进,将C3结构替换为具有更丰富梯度流的C2f结构,并根据不同规模的模型调整了通道数量,从而显著提高了性能。在Head部分,YOLOv8采用了解耦头的设计,将分类头和检测头分离,并从基于锚点的策略转变为无锚点策略。在损失函数计算方面,使用TAL正样本分配策略,并引入了DFL[9]
YOLOv8中的骨干与颈部网络结构中均使用了C2f模块,如图2所示,包含卷积(Conv)模块、分割(Split)模块、瓶颈(Bottleneck)模块、直连拼接(Concat)模块,其中,wh分别为输入图片的宽度、高度,cin为输入通道数,k为核尺寸(Kernel Size),s为步长(Stride),p为池化层(Pooling Layer)参数,cout为输出通道数,c=cout/2,用于后续分通道处理,n为瓶颈层的数量。由图2可知,C2f模块中每一级瓶颈都直连拼接模块,同时分割模块可将单个张量(Tensor)拆分为多个,相当于拼接的逆过程。C2f模块不仅具有轻量化的优点,也提供了更多的梯度流信息。
图3所示,相比于之前的YOLO版本,YOLOv8将耦合头替换为解耦头(Decoupled Head),损失计算仅有边界框损失(Bounding Box Loss,Bbox Loss)和分类损失(Cla Loss),取消了之前的目标(Obj)二元交叉熵损失(Binary Cross-Entropy Loss,BCE Loss),并且增加了分布焦点损失。
基于锚框(Anchor based)的目标检测算法会通过显式或隐式的方法生成一组具有不同尺寸和长宽比的检测框(锚框),然后对这些锚框进行分类或回归。算法通常根据训练数据中物体的尺寸和长宽比的分布来确定锚框的大小、比例、数量,以及交并比阈值等超参数。这些超参数的设置对模型的最终准确性具有显著影响,所以此方法不仅依赖于先验知识,而且泛化能力也较为有限。无锚点(Anchor-Free)目标检测算法在训练时直接学习真实的目标框,在推理过程中不依靠聚类来生成锚框,而是通过学习到的边框距离或关键点位置估算物体的尺寸,先确定目标中心点,然后预测目标中心点到边框的距离。
通过TAL实现动态分配策略,从而计算目标真实框的对齐程度:
T=sα×uβ
式中:s为预测的分类得分,u为预测框与实际框的完全交并比,αβ为可调参数。
根据与T的匹配程度进行排序,选择排名靠前的K个样本为正样本,且当一个预测框与多个真实目标框匹配时,只保留完全交并比(Complete Intersection over Union,CIoU)最高的个例。
分类损失为二元交叉熵损失,与YOLOv5中的分类损失一致,计算公式为:
${L}_{Cla}=-\frac{1}{N}\sum _{i=1}^{N}\left[{y}_{i}log\left(\stackrel{\wedge }{{y}_{i}}\right)+\right.\left.\left(1-{y}_{i}\right)log\left(1-\stackrel{\wedge }{{y}_{i}}\right)\right]$
式中:N为样本实例总数量,yi${\widehat{y}}_{i}$分别为第i个样本的实际标签值和预测标签值。
回归损失包含DFL与完全交并比损失(Complete Intersection over Union Loss,CIoU Loss)2个部分。DFL解决了目标框边界较为灵活或者边界不够明确的情况(如目标框之间存在遮掩),将边界框的位置坐标分布假设为一种通用分布函数,使用交叉熵改进最接近边界框的2个位置的概率,促使网络迅速聚焦于靠近目标位置的区域。该策略通过调整网络对目标位置附近区域的关注度,提高这些区域的概率,以便更有效地捕捉目标的位置信息。
利用完全交并比损失计算目标框损失,公式为:
${L}_{CIoU}=1-{I}_{OU}+\frac{{D}^{2}(b,{b}_{gt})}{{c}^{2}}+\alpha v$
其中:
$v=\frac{4}{{\pi }^{2}}{\left(arctan\frac{{w}_{gt}}{{h}_{gt}}-arctan\frac{{w}_{ab}}{{h}_{ab}}\right)}^{2}$
$\alpha =\frac{v}{\left(1-{I}_{OU}\right)+v}$
式中:IOU为交并比;D为中心点间欧式距离计算函数;bbgt分别为锚框和目标框的中心点;c为目标矩形框的最小对角线距离;α为权衡参数,用于平衡比例;v为修正因子,用于衡量锚框和目标框的比例一致性;wabhab分别为锚框的宽度、高度;wgthgt分别为目标框的宽度、高度。
以自车边缘线为基准线(蓝色),其与图像上边缘的真实距离记为h0,以真实距离0.5 m为间隔,依次标定出红色、黄色、绿色框线,作为预设位置阈值。因摄像头安装位置与角度不同,在不同图像中位置阈值线间的距离存在差异。根据摄像头的标定结果,获取4条阈值线在图像中的位置,与图像上边缘的距离分别记作HBHRHYHG,如图4所示。
算法输出识别结果中目标框的中心点(x,y),以及目标框的高度hgt和宽度wgt,则可计算出识别目标框在图像中位置阈值线的坐标(以视频中图像的左上角点为坐标原点)为y+hgt/2,将其与系统设定的阈值进行比较,可进行不同级别的预警:
a. y+hgt/2≤HG时,目标处于安全区域,无需预警;
b. HG<y+hgt/2≤HY时,车辆可能对盲区中的目标造成安全威胁,需预警提示驾驶人;
c. HY<y+hgt/2≤HR时,车辆对盲区中目标造成安全威胁的可能性迅速增大,需持续预警提示驾驶人;
d. HR<y+hgt/2≤HB时,持续预警提示驾驶人,并辅助制动;
e. y+hgt/2>HB时,紧急制动。
自卸车右转时的右后方盲区视频数据由某商用车智能监控平台提供,道路场景包括城市普通道路和快速干道,天气场景包括白天、夜间及雨天等。
基于Python语言脚本,将视频数据逐帧输出为图片,经过人工剔除相似度较高及无关的图片,得到包含小型乘用车(Car)、载货汽车(Truck)、公交车(Bus)、行人(Person)、电动自行车(E-Bike)、自行车(Bicycle)等类别的样本4 000张,采用LabelImg矩形框标注脚本对图片进行人工标注。最终,将总数据集以6∶1的比例划分为训练集和评估集,如图5所示。
试验设备配置英特尔酷睿i7-13700KF处理器、英伟达GeForce GTX 4070显卡,采用Windows 10操作系统。基于PyCharm编程软件搭建YOLOv8网络模型。
使用YOLOv8网络架构,并采用迁移学习的方法减少模型训练配置资源,加载预训练的权重“yolov8s.pt”。在训练过程中,选择合适的超参数来调整训练过程,寻找最优模型。
模型训练过程中,使用精确度-召回率(Precision-Recall,P-R)曲线以及50%交并比阈值下的平均精度均值(mean Average Precision at 50% intersection over union,mAP50)作为评估指标。由P-R曲线可获得不同阈值下训练得到的模型精确度与召回率之间的联系。曲线下面积越大,表明模型在精确度和召回率之间的平衡效果越好。平均精度(Average Precision,AP)和mAP50的计算公式分别为:
${A}_{P}={\int }_{0}^{1}P\left(R\right)dR$
${m}_{AP50}=\frac{{\sum }_{i=1}^{m}{A}_{Pi}}{m}$
式中:PR分别为精确度、召回率,APi为第i个类别的平均精度,m为类别数量。
模型每次迭代训练集和评估集的损失结果如图6所示。
图6可知,随迭代次数增加,模型训练损失下降,评估集损失出现上升趋势,即模型出现了过拟合。模型训练后的P-R曲线与mAP50指标结果如图7所示,各检测类别的平均精度如表1所示。
图6图7可知,模型评估指标mAP50在迭代后仍处于波动上升状态,但考虑评估集损失下降趋势,最优的模型权重参数应处于总迭代次数(迭代过程中mAP最大值对应的迭代次数)前,即169轮。
模型检测单张图像需要的时间分为3个部分:预处理时长、推理时长、后处理时长。部署最优模型后,在本文试验设备的计算下,检测500张图片的总时间为1 100 ms,单张图片平均处理时长为2.2 ms,帧速率超过25 帧/s,满足试验车辆的使用需求。
采用上述最优模型检测右后盲区中的动态目标,检测结果如图8所示。
图8可知,检测目标不完整以及边界不够明确时,算法仍能识别出目标类型,且目标识别置信度较高。
本文提出一种自卸车右转盲区风险目标动态检测算法,通过YOLOv8模型改进的C2f模块和损失计算模块提高了风险目标动态检测算法的精确率,最终在自卸车右转盲区数据集上进行训练与评估,结果表明,算法在识别小型乘用车、载货汽车、公交车、行人和电动自行车等方面具有较高的检测精度,各类别的整体mAP50为0.87,模型检测帧速率超过25 帧/s,满足实时性要求。同时,通过预先设定目标框位置阈值线,构成右转盲区辅助驾驶系统,可根据图像中风险目标框的位置,实现不同程度的预警。
参考文献 引证文献
排序方式:
[1]
中华人民共和国工业和信息化部. 机动车辆间接视野装置性能和安装要求: GB 15084—2022[S]. 北京: 中国标准出版社, 2023.
Ministry of Industry and Information Technology of the People’s Republic of China. Motor Vehicles—Device for Indirect Vision-Requirement of Performance and Installation: GB 15084—2022[S]. Beijing: Standards Press of China, 2023.
[2]
LIU G R, ZHOU M Z, WANG L L, et al. A Blind Spot Detection and Warning System Based on Millimeter Wave Radar for Driver Assistance[J]. Optik, 2017, 135: 353-365.
[3]
RA M, JUNG H G, SUHR J K, et al. Part-Based Vehicle Detection in Side-Rectilinear Images for Blind-Spot Detection[J]. Expert Systems with Applications, 2018, 101: 116-128.
[4]
ZHAO Y M, BAI L, Y C, et al. Camera-Based Blind Spot Detection with a General Purpose Lightweight Neural Network[J]. Electronics, 2019, 8(2): 233.
[5]
杨炜, 张志威, 周凯霞, 等. 大型车辆右转弯盲区预警系统设计[J]. 中国科技论文, 2019, 14(7): 737-742.
YANG W, ZHANG Z W, ZHOU K X, et al. Design of Right Turn Blind Area Warning System for Large Vehicles[J]. China Sciencepaper, 2019, 14(7): 737-742.
[6]
高鲜萍, 卞学良, 成英, 等. 汽车视觉盲区动态目标智能检测算法[J]. 机械设计与制造, 2023(4): 65-69.
GAO X P, BIAN X L, CHENG Y, et al. Intelligent Detection Algorithm of Dynamic Target in Blind Spot[J]. Machinery Design & Manufacture, 2023(4): 65-69.
[7]
孟银阔, 刘江涛, 龚佑祥, 等. 基于超声波的汽车盲区检测[J]. 现代信息科技, 2023, 7(6): 168-171.
MENG Y K, LIU J T, GONG Y X, et al. Automobile Blind Spot Detection Based on Ultrasound Wave[J]. Modern Information Technology, 2023, 7(6): 168-171.
[8]
HUSSAIN M. YOLO-v1 to YOLO-v8, the Rise of YOLO and Its Complementary Nature Toward Digital Manufacturing and Industrial Defect Detection[J]. Machines, 2023, 11(7).
[9]
LI X, C Q, WANG W H, et al. Generalized Focal Loss: Towards Efficient Representation Learning for Dense Object Detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 45(3): 3139-3153.
2024年第卷第8期
PDF下载
199
95
引用本文
BibTeX
文章信息
doi: 10.20104/j.cnki.1674-6546.20240167
  • 首发时间:2025-11-25
  • 出版时间:2024-08-15
补充材料
相关文章
文章信息
作者
出版历史
  • 修回日期:2024-06-10
基金
作者信息
    1 深圳智慧车联科技有限公司, 深圳 518100
    2 浙江海康科技有限公司, 杭州 310000
    3 长安大学, 西安 710064
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/qcgcs/CN/10.20104/j.cnki.1674-6546.20240167
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