Article(id=1149743089322996103, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1149743083069288795, articleNumber=1003-3033(2024)06-0090-09, orderNo=null, doi=10.16265/j.cnki.issn1003-3033.2024.06.1565, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1708444800000, receivedDateStr=2024-02-21, revisedDate=1712764800000, revisedDateStr=2024-04-11, acceptedDate=null, acceptedDateStr=null, onlineDate=1752049713687, onlineDateStr=2025-07-09, pubDate=1719504000000, pubDateStr=2024-06-28, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1752049713687, onlineIssueDateStr=2025-07-09, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1752049713687, creator=13701087609, updateTime=1752049713687, updator=13701087609, issue=Issue{id=1149743083069288795, tenantId=1146029695717560320, journalId=1146031787341344770, year='2024', volume='34', issue='6', pageStart='1', pageEnd='252', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1752049712197, creator=13701087609, updateTime=1756468919644, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1168278582599098697, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1149743083069288795, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1168278582599098698, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1149743083069288795, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=90, endPage=98, ext={EN=ArticleExt(id=1149743089562071437, articleId=1149743089322996103, tenantId=1146029695717560320, journalId=1146031787341344770, language=EN, title=Multi-scale attention feature-enhanced fusion of a new network for infrared small object detection, columnId=1149733269173878863, journalTitle=China Safety Science Journal, columnName=Safety engineering technology, runingTitle=null, highlight=null, articleAbstract=

In order to improve the performance of small target detection in infrared imaging and the ability of low altitude airspace supervision,an infrared small target detection network based on multi-scale attention feature enhancement fusion was proposed. Firstly,Resnet34 was used to extract the multi-scale features of infrared images. Secondly,the multi-scale spatial attention feature enhancement module(MFEM) was used to improve the ability of feature extraction. Then,in the step-by-step up sampling process,the dual channel attention feature fusion module(DFFM) was used to fuse the semantic information and detail information to better protect the characteristics of infrared small targets. Finally,taking the video sequence detection of ground/air infrared dim small aircraft target as an example,the real scene test was carried out by comparing with other methods. The results show that compared with existing methods,the proposed method improves the scores of intersection over union(IoU),F-measure and false negative rate(FNR),and can accurately locate the target and generate good segmentation results. The DFFM can simultaneously use multi-scale context information and spatial attention mechanism to highlight infiared small targets. The DFFM assigns weights to sets of different channel features,thereby obtaining the most appropriate feature map for feature fusion and improving the detection performance.

, 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=Guimin JIA, Yu CHENG, Mengfei QI), CN=ArticleExt(id=1149743107199119483, articleId=1149743089322996103, tenantId=1146029695717560320, journalId=1146031787341344770, language=CN, title=多尺度注意力特征增强融合的红外小目标检测新网络, columnId=1149733269727526997, journalTitle=中国安全科学学报, columnName=安全工程技术, runingTitle=null, highlight=null, articleAbstract=

为提高红外成像中小目标检测的性能,提高低空空域监管能力,提出一种基于多尺度注意力特征增强融合的红外小目标检测新网络。首先,使用Resnet34提取红外图像的多尺度特征;其次,使用多尺度空间注意力特征增强模块(MFEM)来提高特征提取能力;然后,在逐级上采样过程中使用双通道注意力特征融合模块(DFFM),融合语义信息和细节信息,以更好地保护红外小目标的特征;最后,与其他方法对比,并以地/空红外弱小飞机目标视频序列检测为例测试真实场景。结果表明:新方法与现有方法相比,交互比(IoU)、F值和漏检率(FNR)的评分均获得改进;通过多尺度注意力特征增强融合可准确地定位到目标并生成精细的分割结果;MFEM能够同时利用多尺度上下文信息和空间注意力机制来突出红外小目标;DFFM通过给不同通道特征的集合赋予权重,得到最合适的特征图进行特征融合,从而提高检测性能。

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=HsofFgTff1SVksMKazWPbw==, magXml=NoJkwt/ylYXTj0yz0FyEMg==, pdfUrl=null, pdf=Y2f4Rz++uBR0obLGRHIPug==, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=VoLWj+PC3hYB7xp9NZOHzQ==, mapNumber=null, authorCompany=null, fund=null, authors=

贾桂敏 (1982—),女,河北沧州人,博士,副教授,主要从事光电探测与成像,模式识别等方面的研究。E-mail:

, authorsList=贾桂敏, 程羽, 齐孟飞)}, authors=[Author(id=1168181749784785881, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=gmjia@cauc.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1168181749847700444, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, authorId=1168181749784785881, language=EN, stringName=Guimin JIA, firstName=Guimin, middleName=null, lastName=JIA, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 Tianjin Key Lab for Advanced Signal Processing,Civil Aviation University of China,Tianjin 300300,China
2 College of Electronic Information and Automation,Civil Aviation University of China,Tianjin 300300,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1168181749898032093, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, authorId=1168181749784785881, language=CN, stringName=贾桂敏, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 中国民航大学 天津市智能信号与图像处理重点实验室,天津 300300
2 中国民航大学 电子信息与自动化学院,天津 300300, bio={"img":"QuDxLtaGvkCeKCnXkH5nfg==","content":"

贾桂敏 (1982—),女,河北沧州人,博士,副教授,主要从事光电探测与成像,模式识别等方面的研究。E-mail:

"}, bioImg=QuDxLtaGvkCeKCnXkH5nfg==, bioContent=

贾桂敏 (1982—),女,河北沧州人,博士,副教授,主要从事光电探测与成像,模式识别等方面的研究。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1168181749642179538, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, xref=1, ext=[AuthorCompanyExt(id=1168181749646373843, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, companyId=1168181749642179538, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Tianjin Key Lab for Advanced Signal Processing,Civil Aviation University of China,Tianjin 300300,China), AuthorCompanyExt(id=1168181749650568148, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, companyId=1168181749642179538, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国民航大学 天津市智能信号与图像处理重点实验室,天津 300300)]), AuthorCompany(id=1168181749696705493, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, xref=2, ext=[AuthorCompanyExt(id=1168181749709288406, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, companyId=1168181749696705493, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 College of Electronic Information and Automation,Civil Aviation University of China,Tianjin 300300,China), AuthorCompanyExt(id=1168181749713482711, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, companyId=1168181749696705493, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 中国民航大学 电子信息与自动化学院,天津 300300)])]), Author(id=1168181749939975135, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, 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=1168181749994501090, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, authorId=1168181749939975135, language=EN, stringName=Yu CHENG, firstName=Yu, middleName=null, lastName=CHENG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 Tianjin Key Lab for Advanced Signal Processing,Civil Aviation University of China,Tianjin 300300,China
2 College of Electronic Information and Automation,Civil Aviation University of China,Tianjin 300300,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1168181750044832739, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, authorId=1168181749939975135, language=CN, stringName=程羽, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 中国民航大学 天津市智能信号与图像处理重点实验室,天津 300300
2 中国民航大学 电子信息与自动化学院,天津 300300, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1168181749642179538, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, xref=1, ext=[AuthorCompanyExt(id=1168181749646373843, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, companyId=1168181749642179538, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Tianjin Key Lab for Advanced Signal Processing,Civil Aviation University of China,Tianjin 300300,China), AuthorCompanyExt(id=1168181749650568148, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, companyId=1168181749642179538, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国民航大学 天津市智能信号与图像处理重点实验室,天津 300300)]), AuthorCompany(id=1168181749696705493, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, xref=2, ext=[AuthorCompanyExt(id=1168181749709288406, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, companyId=1168181749696705493, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 College of Electronic Information and Automation,Civil Aviation University of China,Tianjin 300300,China), AuthorCompanyExt(id=1168181749713482711, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, companyId=1168181749696705493, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 中国民航大学 电子信息与自动化学院,天津 300300)])]), Author(id=1168181750090970085, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, 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=1168181750158078952, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, authorId=1168181750090970085, language=EN, stringName=Mengfei QI, firstName=Mengfei, middleName=null, lastName=QI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 Tianjin Key Lab for Advanced Signal Processing,Civil Aviation University of China,Tianjin 300300,China
2 College of Electronic Information and Automation,Civil Aviation University of China,Tianjin 300300,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1168181750212604905, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, authorId=1168181750090970085, language=CN, stringName=齐孟飞, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 中国民航大学 天津市智能信号与图像处理重点实验室,天津 300300
2 中国民航大学 电子信息与自动化学院,天津 300300, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1168181749642179538, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, xref=1, ext=[AuthorCompanyExt(id=1168181749646373843, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, companyId=1168181749642179538, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Tianjin Key Lab for Advanced Signal Processing,Civil Aviation University of China,Tianjin 300300,China), AuthorCompanyExt(id=1168181749650568148, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, companyId=1168181749642179538, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国民航大学 天津市智能信号与图像处理重点实验室,天津 300300)]), AuthorCompany(id=1168181749696705493, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, xref=2, ext=[AuthorCompanyExt(id=1168181749709288406, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, companyId=1168181749696705493, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 College of Electronic Information and Automation,Civil Aviation University of China,Tianjin 300300,China), AuthorCompanyExt(id=1168181749713482711, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, companyId=1168181749696705493, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 中国民航大学 电子信息与自动化学院,天津 300300)])])], keywords=[Keyword(id=1168181750321656810, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, orderNo=1, keyword=infrared image), Keyword(id=1168181750367794155, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, orderNo=2, keyword=small target detection), Keyword(id=1168181750451680236, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, orderNo=3, keyword=feature enhancement), Keyword(id=1168181750502011885, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, orderNo=4, keyword=feature fusion), Keyword(id=1168181750569120750, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, orderNo=5, keyword=attention mechanism), Keyword(id=1168181750623646703, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, orderNo=1, keyword=红外图像), Keyword(id=1168181750669784048, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, orderNo=2, keyword=小目标检测), Keyword(id=1168181750715921393, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, orderNo=3, keyword=特征增强), Keyword(id=1168181750770447346, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, orderNo=4, keyword=特征融合), Keyword(id=1168181750841750515, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, orderNo=5, keyword=注意力机制)], refs=[Reference(id=1168181754071363601, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2023, volume=33, issue=7, pageStart=82, pageEnd=89, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=张磊, 李熙蔚, 燕倩如, journalName=中国安全科学学报, refType=null, unstructuredReference=张磊, 李熙蔚, 燕倩如, 等. 基于改进YOLOv5s的综采工作面人员检测算法[J]. 中国安全科学学报, 2023, 33(7):82-89., articleTitle=基于改进YOLOv5s的综采工作面人员检测算法, refAbstract=null), Reference(id=1168181754142666770, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2023, volume=33, issue=7, pageStart=82, pageEnd=89, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=ZHANG Lei, LI Xiwei, YAN Qianru, journalName=China Safety Science Journal, refType=null, unstructuredReference=ZHANG Lei, LI Xiwei, YAN Qianru, et al. Personnel detection algorithm in fully mechanized coal face based on improved YOLOv5s[J]. China Safety Science Journal, 2023, 33(7):82-89., articleTitle=Personnel detection algorithm in fully mechanized coal face based on improved YOLOv5s, refAbstract=null), Reference(id=1168181754239135763, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2023, volume=32, issue=5, pageStart=140, pageEnd=146, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=张萌, 韩豫, 刘泽峰, journalName=中国安全科学学报, refType=null, unstructuredReference=张萌, 韩豫, 刘泽峰. 深度学习下建筑工人高空安全防护装备检测方法[J]. 中国安全科学学报, 2023, 32(5):140-146., articleTitle=深度学习下建筑工人高空安全防护装备检测方法, refAbstract=null), Reference(id=1168181754323021844, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2023, volume=32, issue=5, pageStart=140, pageEnd=146, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=ZHANG Meng, HAN Yu, LIU Zefeng, journalName=China Safety Science Journal, refType=null, unstructuredReference=ZHANG Meng, HAN Yu, LIU Zefeng. Detection method of high-altitude safety protective equipment for construction workers based on deep learning[J]. China Safety Science Journal, 2023, 32(5):140-146., articleTitle=Detection method of high-altitude safety protective equipment for construction workers based on deep learning, refAbstract=null), Reference(id=1168181754398519317, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2010, volume=43, issue=6, pageStart=2145, pageEnd=2156, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=BAI Xiangzhi, ZHOU Fugen, journalName=Pattern Recognition, refType=null, unstructuredReference=BAI Xiangzhi, ZHOU Fugen. Analysis of new top-hat transformation and the application for infrared dim small target detection[J]. Pattern Recognition, 2010, 43(6):2145-2156., articleTitle=Analysis of new top-hat transformation and the application for infrared dim small target detection, refAbstract=null), Reference(id=1168181754461433878, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2014, volume=52, issue=1, pageStart=574, pageEnd=581, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=CHEN C L P, LI Hong, WEI Yantao, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=CHEN C L P, LI Hong, WEI Yantao, et al. A local contrast method for small infrared target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(1):574-581., articleTitle=A local contrast method for small infrared target detection, refAbstract=null), Reference(id=1168181754515959831, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2020, volume=17, issue=3, pageStart=514, pageEnd=518, url=null, language=null, rfNumber=[5], rfOrder=6, authorNames=DU Peng, HAMDULLA A, journalName=IEEE Geoscience and Remote Sensing Letters, refType=null, unstructuredReference=DU Peng, HAMDULLA A. Infrared small target detection using homogeneity-weighted local contrast measure[J]. IEEE Geoscience and Remote Sensing Letters, 2020, 17(3):514-518., articleTitle=Infrared small target detection using homogeneity-weighted local contrast measure, refAbstract=null), Reference(id=1168181754574680088, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2016, volume=58, issue=null, pageStart=216, pageEnd=226, url=null, language=null, rfNumber=[6], rfOrder=7, authorNames=WEI Yantao, YOU Xinge, LI Hong, journalName=Pattern Recognition, refType=null, unstructuredReference=WEI Yantao, YOU Xinge, LI Hong. Multiscale patch-based contrast measure for small infrared target detection[J]. Pattern Recognition, 2016, 58:216-226., articleTitle=Multiscale patch-based contrast measure for small infrared target detection, refAbstract=null), Reference(id=1168181754625011737, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2021, volume=18, issue=9, pageStart=1670, pageEnd=1674, url=null, language=null, rfNumber=[7], rfOrder=8, authorNames=HAN Jinhui, MORADI S, FARAMARZI I, journalName=IEEE Geoscience and Remote Sensing Letters, refType=null, unstructuredReference=HAN Jinhui, MORADI S, FARAMARZI I, et al. Infrared small target detection based on the weighted strengthened local contrast measure[J]. IEEE Geoscience and Remote Sensing Letters, 2021, 18(9):1670-1674., articleTitle=Infrared small target detection based on the weighted strengthened local contrast measure, refAbstract=null), Reference(id=1168181754700509210, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2013, volume=22, issue=12, pageStart=4996, pageEnd=5009, url=null, language=null, rfNumber=[8], rfOrder=9, authorNames=GAO Chenqiang, MENG Deyu, YANG Yi, journalName=IEEE Transactions on Image Processing, refType=null, unstructuredReference=GAO Chenqiang, MENG Deyu, YANG Yi, et al. Infrared patch-image model for small target detection in a single image[J]. IEEE Transactions on Image Processing, 2013, 22(12):4996-5009., articleTitle=Infrared patch-image model for small target detection in a single image, refAbstract=null), Reference(id=1168181754750840859, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2019, volume=11, issue=4, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=10, authorNames=ZHANG Landan, PENG Zhenming, journalName=Remote Sensing, refType=null, unstructuredReference=ZHANG Landan, PENG Zhenming. Infrared small target detection based on partial sum of the tensor nuclear norm[J]. Remote Sensing, 2019, 11(4): DOI: 10.3390/rs11040382., articleTitle=Infrared small target detection based on partial sum of the tensor nuclear norm, refAbstract=null), Reference(id=1168181754826338332, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2022, volume=19, issue=null, pageStart=1, pageEnd=5, url=null, language=null, rfNumber=[10], rfOrder=11, authorNames=HOU Qingyu, ZHANG Liuwei, TAN Fanjiao, journalName=IEEE Geoscience and Remote Sensing Letters, refType=null, unstructuredReference=HOU Qingyu, ZHANG Liuwei, TAN Fanjiao, et al. ISTDU-Net: infrared small-target detection U-Net[J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19:1-5., articleTitle=ISTDU-Net: infrared small-target detection U-Net, refAbstract=null), Reference(id=1168181754880864285, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2022, volume=19, issue=null, pageStart=1, pageEnd=5, url=null, language=null, rfNumber=[11], rfOrder=12, authorNames=MA Tianlei, YANG Zhen, WANG Jiaqi, journalName=IEEE Geoscience and Remote Sensing Letters, refType=null, unstructuredReference=MA Tianlei, YANG Zhen, WANG Jiaqi, et al. Infrared small target detection network with generate label and feature mapping[J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19:1-5., articleTitle=Infrared small target detection network with generate label and feature mapping, refAbstract=null), Reference(id=1168181754947973150, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2021, volume=59, issue=5, pageStart=4481, pageEnd=4492, url=null, language=null, rfNumber=[12], rfOrder=13, authorNames=ZHAO Bin, WANG Chunping, FU Qiang, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=ZHAO Bin, WANG Chunping, FU Qiang, et al. A novel pattern for infrared small target detection with generative adversarial network[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(5):4481-4492., articleTitle=A novel pattern for infrared small target detection with generative adversarial network, refAbstract=null), Reference(id=1168181755002499103, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=950, pageEnd=959, url=null, language=null, rfNumber=[13], rfOrder=14, authorNames=DAI Yimian, WU Yiquan, ZHOU Fei, journalName=Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, refType=null, unstructuredReference=DAI Yimian, WU Yiquan, ZHOU Fei, et al. Asymmetric contextual modulation for infrared small target detection[C]. Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2021:950-959., articleTitle=Asymmetric contextual modulation for infrared small target detection, refAbstract=null), Reference(id=1168181755065413664, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2023, volume=59, issue=4, pageStart=4250, pageEnd=4261, url=null, language=null, rfNumber=[14], rfOrder=15, authorNames=ZHANG Tianfang, LEI Li, CAO Siying, journalName=IEEE Transactions on Aerospace and Electronic Systems, refType=null, unstructuredReference=ZHANG Tianfang, LEI Li, CAO Siying, et al. Attention-guided pyramid context networks for detecting infrared small target under complex background[J]. IEEE Transactions on Aerospace and Electronic Systems, 2023, 59(4):4250-4261., articleTitle=Attention-guided pyramid context networks for detecting infrared small target under complex background, refAbstract=null), Reference(id=1168181755128328225, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=770, pageEnd=778, url=null, language=null, rfNumber=[15], rfOrder=16, authorNames=HE Kaiming, ZHANG Xiangyu, REN Shaoqing, journalName=Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), refType=null, unstructuredReference=HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016:770-778., articleTitle=Deep residual learning for image recognition, refAbstract=null), Reference(id=1168181755224797218, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2021, volume=59, issue=11, pageStart=9813, pageEnd=9824, url=null, language=null, rfNumber=[16], rfOrder=17, authorNames=DAI Yimian, WU Yiquan, ZHOU Fei, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=DAI Yimian, WU Yiquan, ZHOU Fei, et al. Attentional local contrast networks for infrared small target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(11):9813-9824., articleTitle=Attentional local contrast networks for infrared small target detection, refAbstract=null), Reference(id=1168181755329654819, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2020, volume=5, issue=3, pageStart=291, pageEnd=302, url=null, language=null, rfNumber=[17], rfOrder=18, authorNames=回丙伟, 宋志勇, 范红旗, journalName=中国科学数据, refType=null, unstructuredReference=回丙伟, 宋志勇, 范红旗, 等. 地/空背景下红外图像弱小飞机目标检测跟踪数据集[J]. 中国科学数据, 2020, 5(3):291-302., articleTitle=地/空背景下红外图像弱小飞机目标检测跟踪数据集, refAbstract=null), Reference(id=1168181755468066852, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2020, volume=5, issue=3, pageStart=291, pageEnd=302, url=null, language=null, rfNumber=[17], rfOrder=19, authorNames=HUI Bingwei, SONG Zhiyong, FAN Hongqi, journalName=Science Data Bank, refType=null, unstructuredReference=HUI Bingwei, SONG Zhiyong, FAN Hongqi, et al. A dataset for infrared image dim-small aircraft target detection and tracking under ground/air background[J]. Science Data Bank, 2020, 5(3):291-302., articleTitle=A dataset for infrared image dim-small aircraft target detection and tracking under ground/air background, refAbstract=null), Reference(id=1168181755526787109, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=234, pageEnd=244, url=null, language=null, rfNumber=[18], rfOrder=20, authorNames=RAHMAN M A, WANG Yang, journalName=Proceedings of 12th International Symposium of Advances in Visual Computing (ISVC), refType=null, unstructuredReference=RAHMAN M A, WANG Yang. Optimizing intersection-over-union in deep neural networks for image segmentation[C]. Proceedings of 12th International Symposium of Advances in Visual Computing (ISVC), 2016:234-244., articleTitle=Optimizing intersection-over-union in deep neural networks for image segmentation, refAbstract=null), Reference(id=1168181755652616230, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=833, pageEnd=851, url=null, language=null, rfNumber=[19], rfOrder=21, authorNames=CHEN L C, ZHU Yukun, PAPANDREOU G, journalName=Proceedings of the European Conference on Computer Vision (ECCV), refType=null, unstructuredReference=CHEN L C, ZHU Yukun, PAPANDREOU G, et al. Encoder-decoder with atrous separable convolution for semantic image segmentation[C]. Proceedings of the European Conference on Computer Vision (ECCV), 2018:833-851., articleTitle=Encoder-decoder with atrous separable convolution for semantic image segmentation, refAbstract=null)], funds=[Fund(id=1168181753630961677, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, awardId=61502498, language=CN, fundingSource=国家自然科学基金资助(61502498), fundOrder=null, country=null), Fund(id=1168181753677099022, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, awardId=2022KJ062, language=CN, fundingSource=天津市科研计划项目(2022KJ062), fundOrder=null, country=null), Fund(id=1168181753735819279, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, awardId=21JCYBJC00680, language=CN, fundingSource=天津市自然基金资助(21JCYBJC00680), fundOrder=null, country=null), Fund(id=1168181753857454096, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, awardId=3122019058, language=CN, fundingSource=中央高校基本科研项目(3122019058), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1168181749642179538, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, xref=1, ext=[AuthorCompanyExt(id=1168181749646373843, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, companyId=1168181749642179538, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Tianjin Key Lab for Advanced Signal Processing,Civil Aviation University of China,Tianjin 300300,China), AuthorCompanyExt(id=1168181749650568148, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, companyId=1168181749642179538, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国民航大学 天津市智能信号与图像处理重点实验室,天津 300300)]), AuthorCompany(id=1168181749696705493, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, xref=2, ext=[AuthorCompanyExt(id=1168181749709288406, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, companyId=1168181749696705493, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 College of Electronic Information and Automation,Civil Aviation University of China,Tianjin 300300,China), AuthorCompanyExt(id=1168181749713482711, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, companyId=1168181749696705493, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 中国民航大学 电子信息与自动化学院,天津 300300)])], figs=[ArticleFig(id=1168181750984356852, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, label=Fig.1, caption=Block diagram of infrared small targe detection principle based on feature enhancement and fusion with multiscale attention, figureFileSmall=ujI/RiwkChPOFOub6lEY7g==, figureFileBig=5tr6pl/o7Xj8Lb5NAzpyiQ==, tableContent=null), ArticleFig(id=1168181751043077109, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, label=图1, caption=基于多尺度注意力特征增强融合的红外小目标检测原理, figureFileSmall=ujI/RiwkChPOFOub6lEY7g==, figureFileBig=5tr6pl/o7Xj8Lb5NAzpyiQ==, tableContent=null), ArticleFig(id=1168181751122768886, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, label=Fig.2, caption=Block diagram of principle of MFEM, figureFileSmall=7XyND4Z4ynpPRgx3PEtOgQ==, figureFileBig=ENFjn8lMHomgdHICcdEnmA==, tableContent=null), ArticleFig(id=1168181751269569527, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, label=图2, caption=MFEM原理, figureFileSmall=7XyND4Z4ynpPRgx3PEtOgQ==, figureFileBig=ENFjn8lMHomgdHICcdEnmA==, tableContent=null), ArticleFig(id=1168181751311512568, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, label=Fig.3, caption=Block of diagram of principle of DFFM, figureFileSmall=i6uqZ6OpL3MRkgZ3kc12qg==, figureFileBig=8c/u/3qYYkhxNeK4t+PbYg==, tableContent=null), ArticleFig(id=1168181751349261305, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, label=图3, caption=DFFM原理, figureFileSmall=i6uqZ6OpL3MRkgZ3kc12qg==, figureFileBig=8c/u/3qYYkhxNeK4t+PbYg==, tableContent=null), ArticleFig(id=1168181751462507514, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, label=Fig.4, caption=Comparison of detection results using different methods, figureFileSmall=5R1ICNoFlE38Xb+AvvvHSw==, figureFileBig=ffjKt84pfTT3wbtW5Xf7cA==, tableContent=null), ArticleFig(id=1168181751592530939, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, label=图4, caption=不同方法实际检测结果对比, figureFileSmall=5R1ICNoFlE38Xb+AvvvHSw==, figureFileBig=ffjKt84pfTT3wbtW5Xf7cA==, tableContent=null), ArticleFig(id=1168181751709971452, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, label=Fig.5, caption=ROC curve on SIRST dataset, figureFileSmall=iUN9ETnFkeC+2Gbbgy+Img==, figureFileBig=6l3+tjq3UBpsiX2UbbwCbg==, tableContent=null), ArticleFig(id=1168181751839994877, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, label=图5, caption=SIRST数据集的ROC曲线, figureFileSmall=iUN9ETnFkeC+2Gbbgy+Img==, figureFileBig=6l3+tjq3UBpsiX2UbbwCbg==, tableContent=null), ArticleFig(id=1168181751936463870, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, label=Fig.6, caption=ROC curve on self-constructed dataset, figureFileSmall=q7zb7DnSancNROVHznY/aQ==, figureFileBig=ZGsdhTm7c0AahbOcB2178Q==, tableContent=null), ArticleFig(id=1168181752003572735, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, label=图6, caption=自建数据集的ROC曲线, figureFileSmall=q7zb7DnSancNROVHznY/aQ==, figureFileBig=ZGsdhTm7c0AahbOcB2178Q==, tableContent=null), ArticleFig(id=1168181752062292992, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, label=Fig.7, caption=3D visualization results of different methods, figureFileSmall=grOjkXCRLm9CiuroZBQtOw==, figureFileBig=730DIg1BK6UiMAcVHBvEmg==, tableContent=null), ArticleFig(id=1168181752116817920, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, label=图7, caption=不同方法的三维可视化结果展示, figureFileSmall=grOjkXCRLm9CiuroZBQtOw==, figureFileBig=730DIg1BK6UiMAcVHBvEmg==, tableContent=null), ArticleFig(id=1168181752183926785, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, label=Fig.8, caption=Visualization results of this method on video sequences, figureFileSmall=1qOuQeImHHUKdNuyoVFRHQ==, figureFileBig=AbExnWbc2vQXjBfPDo9fkQ==, tableContent=null), ArticleFig(id=1168181752272007170, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, label=图8, caption=文中方法在视频序列上的可视化结果, figureFileSmall=1qOuQeImHHUKdNuyoVFRHQ==, figureFileBig=AbExnWbc2vQXjBfPDo9fkQ==, tableContent=null), ArticleFig(id=1168181752389447683, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, label=Table 1, caption=

MFEM and DFFM ablation experiment on SIRST dataset

, figureFileSmall=null, figureFileBig=null, tableContent=
不同方法 IoU F FPR/10-2 FNR
Baseline 0.746 6 0.854 8 0.012 4 0.151 3
Baseline+MFEM 0.748 9 0.857 6 0.010 8 0.148 7
Baseline+DFFM 0.757 4 0.860 2 0.009 5 0.136 6
Baseline+MFEM+DFFM 0.762 0 0.864 9 0.008 6 0.128 5
), ArticleFig(id=1168181752473333764, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, label=表1, caption=

SIRST数据集上MFEM、DFFM消融试验

, figureFileSmall=null, figureFileBig=null, tableContent=
不同方法 IoU F FPR/10-2 FNR
Baseline 0.746 6 0.854 8 0.012 4 0.151 3
Baseline+MFEM 0.748 9 0.857 6 0.010 8 0.148 7
Baseline+DFFM 0.757 4 0.860 2 0.009 5 0.136 6
Baseline+MFEM+DFFM 0.762 0 0.864 9 0.008 6 0.128 5
), ArticleFig(id=1168181752687243269, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, label=Table 2, caption=

MFEM and DFFM ablation experiment on self-constructed dataset

, figureFileSmall=null, figureFileBig=null, tableContent=
不同方法 IoU F FPR/10-2 FNR
Baseline 0.753 8 0.859 6 0.013 6 0.147 7
Baseline+MFEM 0.756 2 0.861 1 0.012 8 0.142 3
Baseline+DFFM 0.766 6 0.867 9 0.011 3 0.135 8
Baseline+MFEM+DFFM 0.771 2 0.870 8 0.010 2 0.129 1
), ArticleFig(id=1168181752863404038, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, label=表2, caption=

自行构建的数据集上MFEM、DFFM消融试验

, figureFileSmall=null, figureFileBig=null, tableContent=
不同方法 IoU F FPR/10-2 FNR
Baseline 0.753 8 0.859 6 0.013 6 0.147 7
Baseline+MFEM 0.756 2 0.861 1 0.012 8 0.142 3
Baseline+DFFM 0.766 6 0.867 9 0.011 3 0.135 8
Baseline+MFEM+DFFM 0.771 2 0.870 8 0.010 2 0.129 1
), ArticleFig(id=1168181752926318599, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, label=Table 3, caption=

Comparative experiments on SIRST dataset

, figureFileSmall=null, figureFileBig=null, tableContent=
不同方法 IoU F值 FPR/10-2 FNR
WSLCM 0.049 8 0.094 8 0.798 4 0.406 1
PSTNN 0.515 3 0.680 2 0.003 9 0.390 5
ACM 0.743 0.831 2 0.011 7 0.266 6
DeepLabv3+ 0.684 7 0.812 8 0.032 6 0.191 7
AGPCNet 0.736 1 0.848 0 0.008 9 0.153 9
文中方法 0.762 0 0.864 9 0.008 6 0.128 5
), ArticleFig(id=1168181752993427464, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, label=表3, caption=

SIRST数据集上的对比试验

, figureFileSmall=null, figureFileBig=null, tableContent=
不同方法 IoU F值 FPR/10-2 FNR
WSLCM 0.049 8 0.094 8 0.798 4 0.406 1
PSTNN 0.515 3 0.680 2 0.003 9 0.390 5
ACM 0.743 0.831 2 0.011 7 0.266 6
DeepLabv3+ 0.684 7 0.812 8 0.032 6 0.191 7
AGPCNet 0.736 1 0.848 0 0.008 9 0.153 9
文中方法 0.762 0 0.864 9 0.008 6 0.128 5
), ArticleFig(id=1168181753119256585, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, label=Table 4, caption=

Comparative experiments on self-constructed dataset

, figureFileSmall=null, figureFileBig=null, tableContent=
不同方法 IoU F FPR/10-2 FNR
WSLCM 0.063 8 0.119 7 0.382 1 0.701 1
PSTNN 0.136 1 0.239 5 0.018 2 0.858 5
ACM 0.677 1 0.807 4 0.017 4 0.181 1
DeepLabv3+ 0.685 6 0.813 5 0.035 4 0.171 1
AGPCNet 0.768 0 0.868 7 0.010 0 0.139 0
文中方法 0.771 2 0.870 8 0.010 2 0.129 1
), ArticleFig(id=1168181753291223050, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, label=表4, caption=

自行构建的数据集上的对比试验

, figureFileSmall=null, figureFileBig=null, tableContent=
不同方法 IoU F FPR/10-2 FNR
WSLCM 0.063 8 0.119 7 0.382 1 0.701 1
PSTNN 0.136 1 0.239 5 0.018 2 0.858 5
ACM 0.677 1 0.807 4 0.017 4 0.181 1
DeepLabv3+ 0.685 6 0.813 5 0.035 4 0.171 1
AGPCNet 0.768 0 0.868 7 0.010 0 0.139 0
文中方法 0.771 2 0.870 8 0.010 2 0.129 1
), ArticleFig(id=1168181753404469259, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=EN, label=Table 5, caption=

FPS comparative experiments

, figureFileSmall=null, figureFileBig=null, tableContent=
不同方法 WSLCM PSTNN ACM DeepLabv3+ AGPCNet 本文方法
FPS/(帧·s-1) 0.61 7.73 66.27 53.28 13.48 20.96
), ArticleFig(id=1168181753475772428, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149743089322996103, language=CN, label=表5, caption=

FPS的对比试验

, figureFileSmall=null, figureFileBig=null, tableContent=
不同方法 WSLCM PSTNN ACM DeepLabv3+ AGPCNet 本文方法
FPS/(帧·s-1) 0.61 7.73 66.27 53.28 13.48 20.96
)], attaches=null, journal=Journal(id=1123942128916217864, delFlag=0, nameCn=中国安全科学学报, nameEn=China Safety Science Journal, nameHistory1=null, nameHistory2=null, issn=1003-3033, eissn=, cn=11-2865/X, coden=null, periodic=0, language=CN, oaType=0, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=fkqsFM6VKlHC4gCtS5XqTw==, journalPrice=null, startedYear=null, abbrevIsoEn=Chin Saf Sci J, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754269350027, createdBy=null, updatedBy=13701087609, firstLetterCn=C, firstLetterEn=C, subjectCode=Engineering, subjectName=工程, subjectCodeEn=Engineering, subjectNameEn=null, picCn=fkqsFM6VKlHC4gCtS5XqTw==, picEn=SHn9HgqSxtJrOcAxqD++4Q==, jcr=null, cjcr=null, exts=[JournalExt(id=1159052918994595848, language=CN, name=中国安全科学学报, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.cssjj.com.cn/, createdTime=1754269350050, updatedTime=1754269350050, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.cssjj.com.cn/CN/column/item15.shtml, submissionAuthorUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionReviewUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionCeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionAeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, option={"copyright":""}), JournalExt(id=1159052919040733193, language=EN, name=China Safety Science Journal, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.cssjj.com.cn/EN/1003-3033/home.shtml, createdTime=1754269350061, updatedTime=1754269350061, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=https://synbioj.cip.com.cn/EN/column/column3.shtml, submissionAuthorUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionReviewUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionCeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionAeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, option={"copyright":""})], databaseList=null, tenantJournalId=1146031787341344770, websiteList=[Website(id=1148243202345263519, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146031787341344770, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN, language=CN, createTime=1751692112766, createBy=18614031015, updateTime=1753502583634, updateBy=18614031015, name=《中国安全科学学报》中文站点, tplId=1146099689490845704, title=中国安全科学学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148618794941046792, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=articleTextType, value=kx, createTime=1751781661020, updateTime=1751781661020, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794911686661, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=banner, value=null, createTime=1751781661012, updateTime=1751781661012, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794894909444, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=tui0IVO9FMwB61HHtX5scg==, createTime=1751781661008, updateTime=1751781661008, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794932658183, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751781661017, updateTime=1751781661017, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794924269574, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751781661015, updateTime=1751781661015, creator=18614031015, updator=18614031015)]), Website(id=1155836763751993353, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146031787341344770, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/zgaqkxxb/EN, language=EN, createTime=1753502558893, createBy=18614031015, updateTime=1753524450387, updateBy=18614031015, name=《中国安全科学学报》英文站点, tplId=1146101810881728533, title=China Safety Science Journal, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155895925743669425, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=articleTextType, value=kx, createTime=1753516664205, updateTime=1753516664205, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925722697902, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=banner, value=null, createTime=1753516664200, updateTime=1753516664200, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925714309293, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=tui0IVO9FMwB61HHtX5scg==, createTime=1753516664198, updateTime=1753516664198, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925735280816, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753516664203, updateTime=1753516664203, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925731086511, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753516664202, updateTime=1753516664202, creator=18614031015, updator=18614031015)])], journalTitle=中国安全科学学报, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=China Safety Science Journal, journalPhotoCn=fkqsFM6VKlHC4gCtS5XqTw==, journalPhotoEn=SHn9HgqSxtJrOcAxqD++4Q==, journalFirstLetter=C, journalRecommend=null, journalNew=null, journalCollection=1, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN/10.16265/j.cnki.issn1003-3033.2024.06.1565, detailUrlEn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/EN/10.16265/j.cnki.issn1003-3033.2024.06.1565, pdfUrlCn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN/PDF/10.16265/j.cnki.issn1003-3033.2024.06.1565, pdfUrlEn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/EN/PDF/10.16265/j.cnki.issn1003-3033.2024.06.1565, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
多尺度注意力特征增强融合的红外小目标检测新网络
收藏切换
PDF下载
贾桂敏 1, 2 , 程羽 1, 2 , 齐孟飞 1, 2
中国安全科学学报 | 安全工程技术 2024,34(6): 90-98
收起
收藏切换
中国安全科学学报 | 安全工程技术 2024, 34(6): 90-98
多尺度注意力特征增强融合的红外小目标检测新网络
全屏
贾桂敏1, 2 , 程羽1, 2, 齐孟飞1, 2
作者信息
  • 1 中国民航大学 天津市智能信号与图像处理重点实验室,天津 300300
  • 2 中国民航大学 电子信息与自动化学院,天津 300300
  • 贾桂敏 (1982—),女,河北沧州人,博士,副教授,主要从事光电探测与成像,模式识别等方面的研究。E-mail:

Multi-scale attention feature-enhanced fusion of a new network for infrared small object detection
Guimin JIA1, 2 , Yu CHENG1, 2, Mengfei QI1, 2
Affiliations
  • 1 Tianjin Key Lab for Advanced Signal Processing,Civil Aviation University of China,Tianjin 300300,China
  • 2 College of Electronic Information and Automation,Civil Aviation University of China,Tianjin 300300,China
出版时间: 2024-06-28 doi: 10.16265/j.cnki.issn1003-3033.2024.06.1565
文章导航
收藏切换

为提高红外成像中小目标检测的性能,提高低空空域监管能力,提出一种基于多尺度注意力特征增强融合的红外小目标检测新网络。首先,使用Resnet34提取红外图像的多尺度特征;其次,使用多尺度空间注意力特征增强模块(MFEM)来提高特征提取能力;然后,在逐级上采样过程中使用双通道注意力特征融合模块(DFFM),融合语义信息和细节信息,以更好地保护红外小目标的特征;最后,与其他方法对比,并以地/空红外弱小飞机目标视频序列检测为例测试真实场景。结果表明:新方法与现有方法相比,交互比(IoU)、F值和漏检率(FNR)的评分均获得改进;通过多尺度注意力特征增强融合可准确地定位到目标并生成精细的分割结果;MFEM能够同时利用多尺度上下文信息和空间注意力机制来突出红外小目标;DFFM通过给不同通道特征的集合赋予权重,得到最合适的特征图进行特征融合,从而提高检测性能。

红外图像  /  小目标检测  /  特征增强  /  特征融合  /  注意力机制

In order to improve the performance of small target detection in infrared imaging and the ability of low altitude airspace supervision,an infrared small target detection network based on multi-scale attention feature enhancement fusion was proposed. Firstly,Resnet34 was used to extract the multi-scale features of infrared images. Secondly,the multi-scale spatial attention feature enhancement module(MFEM) was used to improve the ability of feature extraction. Then,in the step-by-step up sampling process,the dual channel attention feature fusion module(DFFM) was used to fuse the semantic information and detail information to better protect the characteristics of infrared small targets. Finally,taking the video sequence detection of ground/air infrared dim small aircraft target as an example,the real scene test was carried out by comparing with other methods. The results show that compared with existing methods,the proposed method improves the scores of intersection over union(IoU),F-measure and false negative rate(FNR),and can accurately locate the target and generate good segmentation results. The DFFM can simultaneously use multi-scale context information and spatial attention mechanism to highlight infiared small targets. The DFFM assigns weights to sets of different channel features,thereby obtaining the most appropriate feature map for feature fusion and improving the detection performance.

infrared image  /  small target detection  /  feature enhancement  /  feature fusion  /  attention mechanism
贾桂敏, 程羽, 齐孟飞. 多尺度注意力特征增强融合的红外小目标检测新网络. 中国安全科学学报, 2024 , 34 (6) : 90 -98 . DOI: 10.16265/j.cnki.issn1003-3033.2024.06.1565
Guimin JIA, Yu CHENG, Mengfei QI. Multi-scale attention feature-enhanced fusion of a new network for infrared small object detection[J]. China Safety Science Journal, 2024 , 34 (6) : 90 -98 . DOI: 10.16265/j.cnki.issn1003-3033.2024.06.1565
红外成像在低空监视、国防防空与无人机搜索等领域有着广泛应用。它具有不易受光照变化的影响、穿透能力强和抗干扰性好等优点。然而,由于红外成像距离远、目标尺寸小、成像分辨率较低,使得在这些领域应用情况下面临的大多是红外小目标检测问题。同时,成像中背景复杂多变、目标对比度较弱,容易导致小目标错检、漏检情况出现[1-2]。因此,提高红外小目标检测性能对我国低空空域安全管控、防空安全等方面有着重要意义。
传统红外小目标检测方法分为基于背景滤波、局部特征和稀疏特征的方法。基于背景滤波的方法假设图像背景一般具有连续性,通过设计合理的滤波算法以抑制背景杂波[3]。基于局部特征的方法根据小目标与其相邻区域灰度不同的特点,构造局部特征算子检测小目标[4-6]。加权增强局部对比度算法(Weighted Strengthened Local Contrast Measure,WSLCM)[7]采用匹配滤波器和背景估计的思想,设计了一种增强的局部对比度模块。基于稀疏特征的方法根据小目标在红外图像中稀疏分布的特点,将红外图像建模成一个低秩稀疏矩阵,利用优化算法从矩阵中恢复低秩分量和稀疏分量,并从稀疏分量中恢复出小目标[8]。基于张量核范数部分和算法(Partial Sum of Tensor Nuclear Norm,PSTNN)[9],提出一种新的非凸低秩约束以更好地保护小目标并抑制背景。
深度神经网络模型通过训练自适应地调整模型参数,可更好地提取红外小目标图像特征,即便在复杂场景中也表现出较好的鲁棒性[10]。MA Tianlei等[11]提出基于尺度自适应特征提取的红外小目标检测网络,并利用二维高斯标签生成策略解决了红外图像数据较少的问题。ZHAO Bin等[12]提出一种基于生成式对抗网络的红外小目标检测算法;DAI Yimian等[13]提出了非对称上下文调制(Asymmetric Contextual Modulation,ACM)网络,并据此搭建一个自底向上的注意机制模块,用于保护深层网络中的红外小目标特征,并将其嵌入到U-Net或特征金字塔网络结构中,完成红外小目标检测;ZHANG Tianfang等[14]提出了注意力引导的金字塔背景网络(Attention-Guided Pyramid Context Networks,AGPCNet)红外小目标检测模型,设计一个上下文注意力引导模块来挖掘红外图像的全局语义信息,并采用不对称融合模块融合低级与高级语义信息。但是,现有基于深度神经网络的红外小目标检测算法在高层语义特征的提取过程中一般存在特征图下采样,容易导致小目标漏检或者错检。
因此,笔者拟提出一种基于多尺度注意力特征增强融合的红外小目标检测新网络,在提取图像深层特征过程中,将空间注意力机制和多尺度策略相结合,增强小目标的特征信息;引入双通道注意力机制,实现高层特征和低层特征的有效融合,以更好地提高红外小目标检测性能。
提出基于多尺度空间注意力特征增强融合的红外小目标检测新方法原理,如图1所示。首先,输入红外图像经过Resnet34[15]主干网络进行特征提取,得到多尺度特征图X1X2X3,其中,特征图的长宽依次减半;然后,利用分多尺度空间注意力特征增强模块(Multiscale spatial attention Feature Enhancement Module,MFEM)对高层特征图X3进行多尺度空间注意力特征增强,增强后的特征经过上采样后与低层特征经过双通道注意力特征融合模块(Dual-channel attentional Feature Fusion Module,DFFM)实现图像高级语义和图像细节特征的融合,同时生成概率置信图;最后,根据阈值检测红外小目标。
为充分提取图像空间域不同尺度的信息,并突出红外小目标在空间域的特征,将多尺度特征表示和空间注意力机制结合,使用不同膨胀率的空洞卷积提取红外图像多尺度特征,通过空间注意力机制使网络更关注特征图中红外小目标所在区域,赋予红外小目标特征更高的权重从而增强红外小目标特征。提出的MFEM原理如图2所示。MFEM模块由多尺度特征提取、空间注意力和一个1×1卷积组成。多尺度特征提取部分由一个1×1卷积、一个膨胀率为2的3×3空洞卷积和一个膨胀率为4的3×3空洞卷积组成。通过不同膨胀率的空洞卷积组合,既能提取多尺度上下文信息,又能更好地保留红外小目标细节信息。空间注意力设计部分,对一幅大小为H×W、通道维数为C的特征图,在通道维度上分别使用平均池化和最大池化聚合特征图的通道信息,得到2个尺寸为1×H×W的特征;然后,2个特征在空间维度上拼接成尺寸为2×H×W的特征;最后,接一个7×7卷积和Sigmoid激活函数,得到一个尺寸为1×H×W的空间注意力权重,将权重乘以输入特征即可得到更关注的重点区域的加权特征。由于多尺度特征和空间注意力加权特征在通道维度拼接后的通道数较大,因此,利用1×1卷积降低其通道维数,得到增强后的特征。
由于红外小目标大多小于100像素,经过网络的多次池化后,高层特征图容易丢失红外小目标特征,需要有效融合图像高级语义信息与细节信息。利用通道注意力机制将高层语义特征和低层语义特征融合,减少不同尺度特征图之间的上下文语义鸿沟。提出DFFM的原理如图3所示。
首先,输入的高层特征和低层特征在通道维度上分别经过平均池化和最大池化操作,以提取小目标的细节纹理与边缘信息;然后,将2种池化得到的特征拼接,并经过2个1×1卷积和ReLU函数实现先降维后升维,得到各个通道的重要程度,这可有效提高小目标与背景的区分度;最后,采用Sigmoid函数得到通道注意力权重,将对应的权重乘以输入特征,得到最佳的通道注意力加权特征。加权的高层特征经过1×1卷积调整通道维数后与加权的低层特征相加,得到融合后的特征。
使用单帧红外小目标检测(Single-frame InfraRed Small Target detection,SIRST)数据集[16]和一个自行构建的红外图像数据集测试试验。SIRST数据集是一个包含427幅红外小目标图像的公开数据集,其中,包含341张训练样本与86张测试样本。SIRST数据集由短波、中波和950 nm波长的红外图像组成,包含多个不同场景下的真实世界视频。自行构建的红外图像数据集在SIRST数据集的基础上进行扩充:增加123张自有红外小目标图像和1 750张地/空背景下红外图像弱小飞机目标检测跟踪数据集[17]中的红外小目标图像。自建数据集共包含2 300张红外小目标图像,其中,1 840张用于训练,460张用于测试。地/空红外弱小飞机数据集包含天空、地面和树林等场景;自有红外小目标图像主要为天空场景。
2个数据集中,自行构建的数据集约有22%的图像包含不少于2个目标,这个比率高于SIRST数据集;自行构建的数据集有15.6%是极小目标(占图像面积小于0.01%),这个比率远远高于SIRST数据集。总之,自行构建的数据集与SIRST数据集相比引入了更具挑战性的场景。
使用红外小目标检测任务常用的交互比(Intersection over Union,IoU)、F值、虚警率(False Positive Rate,FPR)和漏检率(False Negative Rate,FNR)作为模型的量化评价指标,详细定义见文献[11-14]。为更加关注小目标,试验采用SoftIoU[18]作为损失函数,使用随机梯度下降优化器,对比试验和消融试验中的动量、权重衰减系数和训练轮数分别设置为0.9、0.000 1和100。
SIRST数据集与自行构建的数据集按照8∶2比例,划分训练集与测试集。训练前,归一化所有输入图像;随机左右翻转、模糊和裁剪归一化后的图像,在输入网络前,这些图像的分辨率被调整到512×512。
开展消融试验,验证MFEM和DFFM的有效性,试验结果见表1表2。Baseline为基准模型,即不使用MFEM和DFFM,高层特征上采样2倍后直接与低层特征相加,其他试验条件不变。在基准试验模型基础上分别增加MFEM和DFFM测试模型的性能,与文中提出的方法作试验对比。试验结果表明:MFEM和DFFM能在一定程度上改进分割效果。在4种评价指标上都有不同程度的提升。DFFM引入的双通道注意力特征融合策略在文中试验中,相比MFEM,提供的多尺度特征信息对红外小目标检测结果的提升更加明显。同时,使用2个模块能达到最好的分割效果,相比于基准试验模型,在2个数据集上IoU分别提高了1.54%和1.74%,F值分别提高1.01%和1.12%,FPR分别降低0.000 9和0.001 1,FNR分别降低0.008 1和0.006 7。
选取5种具有代表性的红外小目标检测方法与提出的模型比较。在传统的红外小目标检测方法中,选取具有代表性的WSLCM[7]和PSTNN[9]作为对比模型。在基于深度学习的方法中,选取ACM[13]、AGPCNet[14]和具有代表性的目标分割深度学习模型DeepLabv3+[19]作为对比模型。分别在公开的SIRST数据集和自行构建的红外图像数据集上作对比试验,结果见表3表4。从表3表4的结果看出,提出的新网络相比于其他方法可更好地检测红外小目标。在传统方法中,PSTNN方法优于WSLCM方法,这是由于在复杂背景中使用稀疏特征构造小目标的特征比局部特征更加准确;并且在自建数据集中,传统方法的FNR相对较高,这是因为传统方法受限于人为设定的参数,在面对多样化的场景时并不能很好地提取小目标特征。与传统方法相比,基于深度学习的方法能够通过训练的方式从红外图像中提取到红外小目标更抽象、更准确的特征,从而对复杂背景有着更好的鲁棒性。因此,在SIRST数据集和文中自行构建的数据集上都取得不错的检测结果。
文中方法与其他方法相比,在2个数据集的IoU、F值和FNR等3个指标上得分最高,在FPR指标上得分次优,说明本文方法综合性能表现最好。这是因为DeepLabv3+的网络设计缺乏对红外小目标特点的考虑,这也表明了提取红外小目标需要特定的网络设计方案;ACM只融合了低级和深层的语义,缺乏对上下文信息的关注。这使得网络只感知有限的接受域,忽略了全局背景和目标之间的差异;AGPCNet虽然利用全局关联信息来区分目标位置,但是特征图中小目标信息与背景信息的极不平衡容易导致网络混淆,并不能有效突出小目标的区分性。文中设计的可以在不同尺度下学习红外小目标的特征,增强了网络的鲁棒性,同时,DFFM通过赋予特征通道不同的权重能有效地提高小目标与背景的区分度。
使用每秒传输帧数(Frames Per Second,FPS)评估模型的检测速度。在自行构建的数据集上试验,结果见表5。本文方法的FPS为20.96,与其他模型相比,在检测速度与检测精度之间取得了较好的平衡。
不同模型在实际红外小目标图像上的检测结果如图4所示。第1列为原始红外图像,第2—第7列为6种方法的检测结果,最右列为标签。从图4可以看出,传统算法更倾向于检测出高亮或点状区域,WSLCM方法的误检较多,检测效果不如PSTNN方法。这是由于在复杂背景下,基于局部特征的方法相比于基于稀疏特征的方法更容易将目标与背景混淆。在基于深度学习的方法中,本文方法相比于其他方法,具有更少的误检像素数,且分割形状更为准确。这是因为多尺度空间注意力机制挖掘了不同尺度下的红外小目标信息,突出了红外小目标并抑制了背景,同时,双通道注意力机制融合模块有效保护了网络深层的红外小目标特征。
在SIRST数据集和自行构建的数据集上,绘制不同方法的受试者工作特征(Receiver Operating Characteristic,ROC)曲线,分别如图5图6所示。ROC曲线的横轴为假阳率,表示错误预测的目标像素占总背景像素的比例;纵轴为真阳率,表示正确预测的目标像素占总目标像素的比例。由于DeepLabv3+算法输出已被分成2类的分割图,不需要再进行阈值分割,所以,未给出其ROC曲线图。从图5图6可以看出,本文新算法的ROC曲线在所有曲线的上方,具有最大的曲线下面积,检测性能最好。
为便于在检测结果中观察到杂波[14],随机选3幅图对不同方法的检测结果进行三维可视化,如图7所示。可以看出,传统方法的误检和虚警问题较严重,提出的新模型相比于其他模型与真值的结果更加相近,在目标定位和形状分割方面均表现更优。
当前,小型无人机的低空空域管控手段以视频抓拍为主,为评估本文方法针对无人机视频序列的检测泛化性能,选取地/空背景下红外图像弱小飞机目标检测跟踪数据集[17]中的data2,data8,data14视频序列图像测试,试验结果如图8所示。其中,第1,3,5行为原图,第2,4,6行为分割结果,每50帧展示一个分割结果(帧数在每张图的左下角用数字标记)。可以看出,data2与data8中的无人机目标分割准确、精细(图8中第2、4行),说明本文方法在天空与地面背景下表现较好。data14的前200帧分割较精准,但第250与第300帧发生漏检(图8中第6行)。这是由于飞机目标被图像中的高亮建筑物掩盖,难以检测。针对data14中发生的漏检问题,可通过扩充数据集,着重增加高亮建筑物背景下红外小目标的训练图像,提高本文方法的鲁棒性;还可通过在不同位置布置多个检测设备,从不同角度检测红外小目标。
1) 提出红外小目标多尺度注意力特征增强融合检测方法,该方法能够提升红外小目标检测的性能,有利于加强低空空域对无人机等小目标的管制能力。
2) 在SIRST数据集与自建数据集上,文中方法比其他红外小目标检测方法检测精度更高,同时,也能较好地平衡检测速度。在地/空红外弱小飞机视频序列检测下,文中方法具有很好的检测性能,具备实用性。
3) 后续研究包括构建较大样本数量的低空空域管控、国防防空、无人机搜索等应用背景下的红外小目标数据集,以及探索更加高效鲁棒的红外小目标检测方法。这些工作将有助于提升我国低空空域监管、国防防空、无人机搜索等领域的安全和应用能力。
  • 国家自然科学基金资助(61502498)
  • 天津市科研计划项目(2022KJ062)
  • 天津市自然基金资助(21JCYBJC00680)
  • 中央高校基本科研项目(3122019058)
参考文献 引证文献
排序方式:
[1]
张磊, 李熙蔚, 燕倩如, 等. 基于改进YOLOv5s的综采工作面人员检测算法[J]. 中国安全科学学报, 2023, 33(7):82-89.
ZHANG Lei, LI Xiwei, YAN Qianru, et al. Personnel detection algorithm in fully mechanized coal face based on improved YOLOv5s[J]. China Safety Science Journal, 2023, 33(7):82-89.
[2]
张萌, 韩豫, 刘泽峰. 深度学习下建筑工人高空安全防护装备检测方法[J]. 中国安全科学学报, 2023, 32(5):140-146.
ZHANG Meng, HAN Yu, LIU Zefeng. Detection method of high-altitude safety protective equipment for construction workers based on deep learning[J]. China Safety Science Journal, 2023, 32(5):140-146.
[3]
BAI Xiangzhi, ZHOU Fugen. Analysis of new top-hat transformation and the application for infrared dim small target detection[J]. Pattern Recognition, 2010, 43(6):2145-2156.
[4]
CHEN C L P, LI Hong, WEI Yantao, et al. A local contrast method for small infrared target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(1):574-581.
[5]
DU Peng, HAMDULLA A. Infrared small target detection using homogeneity-weighted local contrast measure[J]. IEEE Geoscience and Remote Sensing Letters, 2020, 17(3):514-518.
[6]
WEI Yantao, YOU Xinge, LI Hong. Multiscale patch-based contrast measure for small infrared target detection[J]. Pattern Recognition, 2016, 58:216-226.
[7]
HAN Jinhui, MORADI S, FARAMARZI I, et al. Infrared small target detection based on the weighted strengthened local contrast measure[J]. IEEE Geoscience and Remote Sensing Letters, 2021, 18(9):1670-1674.
[8]
GAO Chenqiang, MENG Deyu, YANG Yi, et al. Infrared patch-image model for small target detection in a single image[J]. IEEE Transactions on Image Processing, 2013, 22(12):4996-5009.
[9]
ZHANG Landan, PENG Zhenming. Infrared small target detection based on partial sum of the tensor nuclear norm[J]. Remote Sensing, 2019, 11(4): DOI: 10.3390/rs11040382.
[10]
HOU Qingyu, ZHANG Liuwei, TAN Fanjiao, et al. ISTDU-Net: infrared small-target detection U-Net[J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19:1-5.
[11]
MA Tianlei, YANG Zhen, WANG Jiaqi, et al. Infrared small target detection network with generate label and feature mapping[J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19:1-5.
[12]
ZHAO Bin, WANG Chunping, FU Qiang, et al. A novel pattern for infrared small target detection with generative adversarial network[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(5):4481-4492.
[13]
DAI Yimian, WU Yiquan, ZHOU Fei, et al. Asymmetric contextual modulation for infrared small target detection[C]. Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2021:950-959.
[14]
ZHANG Tianfang, LEI Li, CAO Siying, et al. Attention-guided pyramid context networks for detecting infrared small target under complex background[J]. IEEE Transactions on Aerospace and Electronic Systems, 2023, 59(4):4250-4261.
[15]
HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016:770-778.
[16]
DAI Yimian, WU Yiquan, ZHOU Fei, et al. Attentional local contrast networks for infrared small target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(11):9813-9824.
[17]
回丙伟, 宋志勇, 范红旗, 等. 地/空背景下红外图像弱小飞机目标检测跟踪数据集[J]. 中国科学数据, 2020, 5(3):291-302.
HUI Bingwei, SONG Zhiyong, FAN Hongqi, et al. A dataset for infrared image dim-small aircraft target detection and tracking under ground/air background[J]. Science Data Bank, 2020, 5(3):291-302.
[18]
RAHMAN M A, WANG Yang. Optimizing intersection-over-union in deep neural networks for image segmentation[C]. Proceedings of 12th International Symposium of Advances in Visual Computing (ISVC), 2016:234-244.
[19]
CHEN L C, ZHU Yukun, PAPANDREOU G, et al. Encoder-decoder with atrous separable convolution for semantic image segmentation[C]. Proceedings of the European Conference on Computer Vision (ECCV), 2018:833-851.
2024年第34卷第6期
PDF下载
396
166
引用本文
BibTeX
文章信息
doi: 10.16265/j.cnki.issn1003-3033.2024.06.1565
  • 接收时间:2024-02-21
  • 首发时间:2025-07-09
  • 出版时间:2024-06-28
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-02-21
  • 修回日期:2024-04-11
基金
国家自然科学基金资助(61502498)
天津市科研计划项目(2022KJ062)
天津市自然基金资助(21JCYBJC00680)
中央高校基本科研项目(3122019058)
作者信息
    1 中国民航大学 天津市智能信号与图像处理重点实验室,天津 300300
    2 中国民航大学 电子信息与自动化学院,天津 300300
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN/10.16265/j.cnki.issn1003-3033.2024.06.1565
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