Article(id=1241060179580612938, tenantId=1146029695717560320, journalId=1238841944844054536, issueId=1241060178263601474, articleNumber=null, orderNo=null, doi=10.12347/j.ycyk.20240119001, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1705593600000, receivedDateStr=2024-01-19, revisedDate=1709568000000, revisedDateStr=2024-03-05, acceptedDate=null, acceptedDateStr=null, onlineDate=1773821404585, onlineDateStr=2026-03-18, pubDate=null, pubDateStr=null, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1773821404585, onlineIssueDateStr=2026-03-18, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1773821404585, creator=13701087609, updateTime=1773821404585, updator=13701087609, issue=Issue{id=1241060178263601474, tenantId=1146029695717560320, journalId=1238841944844054536, year='2024', volume='45', issue='2', pageStart='1', pageEnd='123', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1773821404281, creator=13701087609, updateTime=1773821891324, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1241062221128724613, tenantId=1146029695717560320, journalId=1238841944844054536, issueId=1241060178263601474, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1241062221128724614, tenantId=1146029695717560320, journalId=1238841944844054536, issueId=1241060178263601474, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=10, endPage=17, ext={EN=ArticleExt(id=1241060179890991440, articleId=1241060179580612938, tenantId=1146029695717560320, journalId=1238841944844054536, language=EN, title=KDTGAN: Knowledge Distillation via Transformer GAN for Hyperspectral Target Detection, columnId=1241060179823882575, journalTitle=Journal of Telemetry, Tracking and Command, columnName=Artificial Intelligence Technology, runingTitle=null, highlight=null, articleAbstract=

Hyperspectral target detection is crucial in Earth observation for both military and civilian applications. However,complex backgrounds and the scarcity of target samples pose challenges in hyperspectral image analysis. In this paper, we first employ the CEM coarse detection method to extract background data. Subsequently, a novel knowledge distillation model, namely KDTGAN (implemented through Transformer-GAN), is introduced. The generator of this teacher model adopts the structure of a Transformer encoder and combines it with a multi-scale data fusion approach to accurately learn the background distribution, which in turn enables target detection by reconstructing the background information. To overcome the challenge of unstable GAN training,especially the scarcity of pure background data, we propose a new loss algorithm to reduce the negative impact of suspicious target samples on model performance. To reduce the computational burden of the model, we introduce knowledge distillation and design a new distillation loss to constrain the student model to lighten the model while improving the student model's detection accuracy. The experimental results show that KDTGAN performs better than current detection methods with higher detection accuracy and robustness.

, 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=Wen XIE, Chenchao SHAN, Zhezhe ZHANG, Jiapeng ZHANG), CN=ArticleExt(id=1241060187142943237, articleId=1241060179580612938, tenantId=1146029695717560320, journalId=1238841944844054536, language=CN, title=KDTGAN:基于Transformer-GAN和知识蒸馏的高光谱目标检测, columnId=1241060179987460434, journalTitle=遥测遥控, columnName=人工智能技术, runingTitle=null, highlight=null, articleAbstract=

高光谱目标检测在地球观测中至关重要,被广泛应用于军事和民用领域。然而,由于高光谱图像的背景复杂性和目标样本的有限性,该任务面临较大的挑战。本文首先采用CEM(约束能量最小化)粗检测方法提取背景数据。随之,引入了一种新的知识蒸馏模型,即KDTGAN(通过Transformer-GAN实现)。教师模型的生成器采用了Transformer编码器的结构,并结合多尺度数据融合的方法,能够准确地学习背景分布,进而通过重构背景信息实现目标检测。为了克服GAN(生成对抗网络)训练不稳定的挑战,特别是纯背景数据的稀缺性,本文提出了一种新的损失算法,以减小可疑目标样本对模型性能的负面影响。为了降低模型的计算负担,本文引入知识蒸馏,并设计新的蒸馏损失对学生模型加以约束,使模型轻量化的同时提高学生模型检测精度。实验结果表明:KDTGAN相较于当前检测方法表现更优,具有更高的检测精度和鲁棒性。

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=XTmdvSihmCGfBCPG0umuaA==, magXml=y83wKVUlU4l5lJQFFBMz6Q==, pdfUrl=null, pdf=0josMYpkml0A45xI5JDalQ==, pdfFileSize=5480773, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=Ih5/Kl5/UTRS8149Mi3xoQ==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=D51zOlwQ7wKTfjwz3UeVrg==, mapNumber=null, authorCompany=null, fund=null, authors=

谢雯 1989年生,副教授,研究生导师。

闪晨超 1999年生,硕士研究生。

张哲哲 1997年生,硕士研究生。

张嘉鹏 1998年生,硕士研究生。

, authorsList=谢雯, 闪晨超, 张哲哲, 张嘉鹏)}, authors=[Author(id=1241060187533013525, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, 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=1241060187637871130, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, authorId=1241060187533013525, language=EN, stringName=Wen XIE, firstName=Wen, middleName=null, lastName=XIE, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=School of Communications and Information Engineering, Xi'an University of Posts and Telecommunications, Xi'an 710121, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1241060187721757213, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, authorId=1241060187533013525, language=CN, stringName=谢雯, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=西安邮电大学通信与信息工程学院 西安 710121, bio={"content":"

谢雯 1989年生,副教授,研究生导师。

"}, bioImg=null, bioContent=

谢雯 1989年生,副教授,研究生导师。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1241060187440738829, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, xref=null, ext=[AuthorCompanyExt(id=1241060187444933134, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, companyId=1241060187440738829, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Communications and Information Engineering, Xi'an University of Posts and Telecommunications, Xi'an 710121, China), AuthorCompanyExt(id=1241060187453321743, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, companyId=1241060187440738829, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=西安邮电大学通信与信息工程学院 西安 710121)])]), Author(id=1241060187805643298, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, 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=1241060187902112297, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, authorId=1241060187805643298, language=EN, stringName=Chenchao SHAN, firstName=Chenchao, middleName=null, lastName=SHAN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=School of Communications and Information Engineering, Xi'an University of Posts and Telecommunications, Xi'an 710121, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1241060187985998381, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, authorId=1241060187805643298, language=CN, stringName=闪晨超, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=西安邮电大学通信与信息工程学院 西安 710121, bio={"content":"

闪晨超 1999年生,硕士研究生。

"}, bioImg=null, bioContent=

闪晨超 1999年生,硕士研究生。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1241060187440738829, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, xref=null, ext=[AuthorCompanyExt(id=1241060187444933134, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, companyId=1241060187440738829, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Communications and Information Engineering, Xi'an University of Posts and Telecommunications, Xi'an 710121, China), AuthorCompanyExt(id=1241060187453321743, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, companyId=1241060187440738829, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=西安邮电大学通信与信息工程学院 西安 710121)])]), Author(id=1241060188086661686, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, 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=1241060188183130683, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, authorId=1241060188086661686, language=EN, stringName=Zhezhe ZHANG, firstName=Zhezhe, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=School of Communications and Information Engineering, Xi'an University of Posts and Telecommunications, Xi'an 710121, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1241060188292182595, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, authorId=1241060188086661686, language=CN, stringName=张哲哲, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=西安邮电大学通信与信息工程学院 西安 710121, bio={"content":"

张哲哲 1997年生,硕士研究生。

"}, bioImg=null, bioContent=

张哲哲 1997年生,硕士研究生。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1241060187440738829, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, xref=null, ext=[AuthorCompanyExt(id=1241060187444933134, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, companyId=1241060187440738829, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Communications and Information Engineering, Xi'an University of Posts and Telecommunications, Xi'an 710121, China), AuthorCompanyExt(id=1241060187453321743, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, companyId=1241060187440738829, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=西安邮电大学通信与信息工程学院 西安 710121)])]), Author(id=1241060188401234506, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1241060188489314894, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, authorId=1241060188401234506, language=EN, stringName=Jiapeng ZHANG, firstName=Jiapeng, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=School of Communications and Information Engineering, Xi'an University of Posts and Telecommunications, Xi'an 710121, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1241060188560618065, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, authorId=1241060188401234506, language=CN, stringName=张嘉鹏, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=西安邮电大学通信与信息工程学院 西安 710121, bio={"content":"

张嘉鹏 1998年生,硕士研究生。

"}, bioImg=null, bioContent=

张嘉鹏 1998年生,硕士研究生。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1241060187440738829, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, xref=null, ext=[AuthorCompanyExt(id=1241060187444933134, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, companyId=1241060187440738829, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Communications and Information Engineering, Xi'an University of Posts and Telecommunications, Xi'an 710121, China), AuthorCompanyExt(id=1241060187453321743, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, companyId=1241060187440738829, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=西安邮电大学通信与信息工程学院 西安 710121)])])], keywords=[Keyword(id=1241060188636115541, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=EN, orderNo=1, keyword=Hyperspectral), Keyword(id=1241060188707418714, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=EN, orderNo=2, keyword=Target detection), Keyword(id=1241060188774527583, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=EN, orderNo=3, keyword=Knowledge distillation), Keyword(id=1241060188841636454, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=EN, orderNo=4, keyword=GAN), Keyword(id=1241060188917133929, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=EN, orderNo=5, keyword=Transformer-GAN), Keyword(id=1241060189017797228, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=CN, orderNo=1, keyword=高光谱图像), Keyword(id=1241060190502580848, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=CN, orderNo=2, keyword=目标检测), Keyword(id=1241060190615827059, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=CN, orderNo=3, keyword=知识蒸馏), Keyword(id=1241060190691324534, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=CN, orderNo=4, keyword=生成对抗网络), Keyword(id=1241060190758433401, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=CN, orderNo=5, keyword=Transformer-GAN)], refs=[Reference(id=1241060192654258895, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=Microimages, journalName=null, refType=null, unstructuredReference=Microimages. Introduction to hyperspectral imaging[EB/OL].[2024-01-19] https://www.microimages.com/documentation/Tutorials/hyprspec.pdf., articleTitle=Introduction to hyperspectral imaging, refAbstract=null), Reference(id=1241060192750727892, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=1997, volume=59, issue=1, pageStart=64, pageEnd=76, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=FARRAND W H, HARSANYI J C, journalName=Remote Sensing of Environment, refType=null, unstructuredReference=FARRAND W H, HARSANYI J C. Mapping the distribution of mine tailings in the Coeur d'Alene River Valley, Idaho, through the use of a constrained energy minimization technique[J]. Remote Sensing of Environment, 1997, 59(1): 64-76., articleTitle=Mapping the distribution of mine tailings in the Coeur d'Alene River Valley, Idaho, through the use of a constrained energy minimization technique, refAbstract=null), Reference(id=1241060192838808280, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2015, volume=48, issue=12, pageStart=3904, pageEnd=3916, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=LI W, DU Q, ZHANG B, journalName=Pattern Recognition, refType=null, unstructuredReference=LI W, DU Q, ZHANG B. Combined sparse and collaborative representation for hyperspectral target detection[J]. Pattern Recognition, 2015, 48(12): 3904-3916., articleTitle=Combined sparse and collaborative representation for hyperspectral target detection, refAbstract=null), Reference(id=1241060192910111450, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2001, volume=49, issue=1, pageStart=1, pageEnd=16, url=null, language=null, rfNumber=[4], rfOrder=3, authorNames=KRAUT S, SCHARF L L, MCWHORTER L T, journalName=IEEE Transactions on Signal Processing, refType=null, unstructuredReference=KRAUT S, SCHARF L L, MCWHORTER L T. Adaptive subspace detectors[J]. IEEE Transactions on Signal Processing, 2001, 49(1): 1-16., articleTitle=Adaptive subspace detectors, refAbstract=null), Reference(id=1241060192993997534, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=1992, volume=28, issue=1, pageStart=208, pageEnd=216, url=null, language=null, rfNumber=[5], rfOrder=4, authorNames=FUHRMANN D R, KELLY E J, NITZBERG R, journalName=IEEE Transactions on Aerospace and Electronic Systems, refType=null, unstructuredReference=FUHRMANN D R, KELLY E J, NITZBERG R. A CFAR adaptive matched filter detector[J]. IEEE Transactions on Aerospace and Electronic Systems, 1992, 28(1): 208-216., articleTitle=A CFAR adaptive matched filter detector, refAbstract=null), Reference(id=1241060193103049442, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=10.14016/j.cnki.jgzz.2022.10.001, pmid=null, pmcid=null, year=2022, volume=43, issue=10, pageStart=1, pageEnd=6, url=null, language=null, rfNumber=[6], rfOrder=5, authorNames=赵佳乐, 王广龙, 周冰, journalName=激光杂志, refType=null, unstructuredReference=赵佳乐, 王广龙, 周冰, 等. 深度学习下高光谱图像目标检测技术研究进展[J]. 激光杂志, 2022, 43(10): 1-6., articleTitle=深度学习下高光谱图像目标检测技术研究进展, refAbstract=null), Reference(id=1241060193203712744, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2014, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[7], rfOrder=6, authorNames=GOODFELLOW I, POUGET-ABADIE J, MIRZA M, journalName=null, refType=null, unstructuredReference=GOODFELLOW I, POUGET-ABADIE J, MIRZA M,et al. Generative adversarial nets[C]//NIPS'14: Procee-dings of the 27th International Conference on Neural Information Processing Systems, December 8-13, 2014, Montreal Canada. Cambridge: MIT Press, 2014., articleTitle=Generative adversarial nets, refAbstract=null), Reference(id=1241060193342124778, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2013, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[8], rfOrder=7, authorNames=KINGMA D P, WELLING M, journalName=null, refType=null, unstructuredReference=KINGMA D P, WELLING M. Auto-encoding variational Bayes[EB/OL]. arXiv preprint, 2013 arXiv:, articleTitle=Auto-encoding variational Bayes, refAbstract=null), Reference(id=1241060193421816557, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2020, volume=13, issue=null, pageStart=5887, pageEnd=5897, url=null, language=null, rfNumber=[9], rfOrder=8, authorNames=XIE Weiying, ZHANG Xin, LI Yunsong, journalName=IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, refType=null, unstructuredReference=XIE Weiying, ZHANG Xin, LI Yunsong, et al. Background learning based on target suppression constraint for hyperspectral target detection[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020, 13: 5887-5897., articleTitle=Background learning based on target suppression constraint for hyperspectral target detection, refAbstract=null), Reference(id=1241060193488925425, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=9, authorNames=SHEN Kun, XIE Weixin, TANG Haojin, journalName=null, refType=null, unstructuredReference=SHEN Kun, XIE Weixin, TANG Haojin, et al. Dense convolution siamese network for hyperspectral image target detection[C]//2022 5th International Conference on Information Communication and Signal Processing(ICICSP), November 26-28, 2022, Shenzhen, China. IEEE, 2022., articleTitle=Dense convolution siamese network for hyperspectral image target detection, refAbstract=null), Reference(id=1241060194977903349, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[11], rfOrder=10, authorNames=CHEN Xi, WANG Yulei, CHE Zongwei, journalName=null, refType=null, unstructuredReference=CHEN Xi, WANG Yulei, CHE Zongwei, et al. Contrastive learning for hyperspectral target detection[C]//IGARSS 2022-2022 IEEE International Geoscience and Remote Sensing Symposium, July 17-22, 2022, Kuala Lumpur, Malaysia. IEEE, 2022., articleTitle=Contrastive learning for hyperspectral target detection, refAbstract=null), Reference(id=1241060195078566649, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2019, volume=58, issue=2, pageStart=1463, pageEnd=1474, url=null, language=null, rfNumber=[12], rfOrder=11, authorNames=XIE Weiying, YANG Jian, LEI Jie, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=XIE Weiying, YANG Jian, LEI Jie, et al. SRUN: Spectral regularized unsupervised networks for hyperspectral target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 58(2): 1463-1474., articleTitle=SRUN: Spectral regularized unsupervised networks for hyperspectral target detection, refAbstract=null), Reference(id=1241060195154064124, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=10.27389/d.cnki.gxadu.2022.002466, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[13], rfOrder=12, authorNames=白凯玮, journalName=null, refType=null, unstructuredReference=白凯玮. 基于分布式深度学习的高光谱图像目标检测算法研究[D].西安: 西安电子科技大学, 2022., articleTitle=基于分布式深度学习的高光谱图像目标检测算法研究, refAbstract=null), Reference(id=1241060195246338814, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[14], rfOrder=13, authorNames=QIN Haonan, XIE Weiying, LI Yunsong, journalName=null, refType=null, unstructuredReference=QIN Haonan, XIE Weiying, LI Yunsong, et al. PTGAN:A proposal-weighted two-stage GAN with attention for hyperspectral target detection[C]//2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS, July 11-16, 2021, Brussels, Belgium. IEEE, 2021., articleTitle=PTGAN:A proposal-weighted two-stage GAN with attention for hyperspectral target detection, refAbstract=null), Reference(id=1241060195321836291, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=10.27241/d.cnki.gnjgu.2019.002211, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=14, authorNames=鲍亮, journalName=null, refType=null, unstructuredReference=鲍亮. 基于深度学习的高光谱遥感图像目标检测算法[D]. 南京: 南京理工大学, 2019., articleTitle=基于深度学习的高光谱遥感图像目标检测算法, refAbstract=null), Reference(id=1241060195388945159, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=10.27389/d.cnki.gxadu.2019.000351, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[16], rfOrder=15, authorNames=解静, journalName=null, refType=null, unstructuredReference=解静. 基于深度学习的高光谱图像目标检测算法研究[D]. 西安: 西安电子科技大学, 2019., articleTitle=基于深度学习的高光谱图像目标检测算法研究, refAbstract=null), Reference(id=1241060195481219853, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2020, volume=56, issue=21, pageStart=1116, pageEnd=1118, url=null, language=null, rfNumber=[17], rfOrder=16, authorNames=KIM J H, KIM J, JOUNG J, journalName=Electro-nics Letters, refType=null, unstructuredReference=KIM J H, KIM J, JOUNG J. Siamese hyperspectral target detection using synthetic training data[J]. Electro-nics Letters, 2020, 56(21): 1116-1118., articleTitle=Siamese hyperspectral target detection using synthetic training data, refAbstract=null), Reference(id=1241060195581883154, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[18], rfOrder=17, authorNames=QIN Haonan, XIE Weiying, LI Yunsong, journalName=null, refType=null, unstructuredReference=QIN Haonan, XIE Weiying, LI Yunsong, et al. HTD-VIT: Spectral-spatial joint hyperspectral target detection with vision transformer[C]//IGARSS 2022-2022 IEEE International Geoscience and Remote Sensing Symposium, July 17-22, 2022, Kuala Lumpur, Malaysia. IEEE, 2022., articleTitle=HTD-VIT: Spectral-spatial joint hyperspectral target detection with vision transformer, refAbstract=null), Reference(id=1241060195665769237, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2023, volume=61, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[19], rfOrder=18, authorNames=JIAO Jinyue, GONG Zhiqiang, ZHONG Ping, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=JIAO Jinyue, GONG Zhiqiang, ZHONG Ping. Triplet spectral-wise transformer network for hyperspectral target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 5519817., articleTitle=Triplet spectral-wise transformer network for hyperspectral target detection, refAbstract=null), Reference(id=1241060195737072408, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2023, volume=61, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[20], rfOrder=19, authorNames=SHEN Dunbin, MA Xiaorui, KONG Wenfeng, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=SHEN Dunbin, MA Xiaorui, KONG Wenfeng, et al. Hyperspectral target detection based on interpretable representation network[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 5519416., articleTitle=Hyperspectral target detection based on interpretable representation network, refAbstract=null), Reference(id=1241060195816764188, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2022, volume=60, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[21], rfOrder=20, authorNames=RAO Weiqiang, GAO Lianru, QU Ying, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=RAO Weiqiang, GAO Lianru, QU Ying, et al. Siamese transformer network for hyperspectral image target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 5526419., articleTitle=Siamese transformer network for hyperspectral image target detection, refAbstract=null), Reference(id=1241060195917427487, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2023, volume=61, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[22], rfOrder=21, authorNames=LI Yunsong, QIN Haonan, XIE Weiying, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=LI Yunsong, QIN Haonan, XIE Weiying. HTDFormer:Hyperspectral target detection based on transformer with distributed learning[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 5524715., articleTitle=HTDFormer:Hyperspectral target detection based on transformer with distributed learning, refAbstract=null), Reference(id=1241060195997119267, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[23], rfOrder=22, authorNames=KARRAS Tero, LAINE Samuli, AILA Timo, journalName=null, refType=null, unstructuredReference=KARRAS Tero, LAINE Samuli, AILA Timo. A style-based generator architecture for generative adversarial networks[C]// 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 15-20, 2019, Long Beach, CA, USA. IEEE, 2019., articleTitle=A style-based generator architecture for generative adversarial networks, refAbstract=null), Reference(id=1241060196097782567, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[24], rfOrder=23, authorNames=VASWANI A, SHAZEER N, PARMAR N, journalName=null, refType=null, unstructuredReference=VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[EB/OL]. https://doi.org/10.48550/arXiv.1706.03762., articleTitle=Attention is all you need, refAbstract=null), Reference(id=1241060196173280039, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=3967, pageEnd=3976, url=null, language=null, rfNumber=[25], rfOrder=24, authorNames=PARK W, KIM D, LU Y, journalName=null, refType=null, unstructuredReference=PARK W, KIM D, LU Y, et al. Relational knowledge distillation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 15-20, 2019, Long Beach, CA, USA. IEEE, 2019: 3967-3976., articleTitle=Relational knowledge distillation, refAbstract=null)], funds=[Fund(id=1241060192209662656, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, awardId=61901365; 62071379, language=CN, fundingSource=国家自然科学基金(61901365; 62071379), fundOrder=null, country=null), Fund(id=1241060192331297476, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, awardId=2019JQ-377, language=CN, fundingSource=陕西省自然科学基金(2019JQ-377), fundOrder=null, country=null), Fund(id=1241060192427766471, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, awardId=19JK0805, language=CN, fundingSource=陕西省教育厅专项科研计划(19JK0805), fundOrder=null, country=null), Fund(id=1241060192536818379, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, awardId=xyt2016-01, language=CN, fundingSource=西安邮电大学西邮新星团队项目(xyt2016-01), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1241060187440738829, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, xref=null, ext=[AuthorCompanyExt(id=1241060187444933134, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, companyId=1241060187440738829, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Communications and Information Engineering, Xi'an University of Posts and Telecommunications, Xi'an 710121, China), AuthorCompanyExt(id=1241060187453321743, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, companyId=1241060187440738829, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=西安邮电大学通信与信息工程学院 西安 710121)])], figs=[ArticleFig(id=1241060190892651133, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=EN, label=Fig.1, caption=Framework diagram of our proposed model, figureFileSmall=9tlN2gApxsP/FaML8koaUA==, figureFileBig=Ih5/Kl5/UTRS8149Mi3xoQ==, tableContent=null), ArticleFig(id=1241060190968148610, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=CN, label=图1, caption=KDTGAN模型总体框架图, figureFileSmall=9tlN2gApxsP/FaML8koaUA==, figureFileBig=Ih5/Kl5/UTRS8149Mi3xoQ==, tableContent=null), ArticleFig(id=1241060191224001165, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=EN, label=Fig.2, caption=Distillation of spatial and spectral information knowledge, figureFileSmall=AjLnqsWBtcbuEHVdoptq4A==, figureFileBig=oa6EgC9vKHFB2ffdvGPWww==, tableContent=null), ArticleFig(id=1241060191312081556, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=CN, label=图2, caption=空间信息和光谱信息知识蒸馏, figureFileSmall=AjLnqsWBtcbuEHVdoptq4A==, figureFileBig=oa6EgC9vKHFB2ffdvGPWww==, tableContent=null), ArticleFig(id=1241060191412744857, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=EN, label=Fig.3, caption=Comparison of different methods for detecting graphs on three data sets, figureFileSmall=qbP14plAq7ZxxBGSmLeYoA==, figureFileBig=MS9/BnxxqyFzn44MRLvWyw==, tableContent=null), ArticleFig(id=1241060191492436638, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=CN, label=图3, caption=三种数据集检测图不同方法对比, figureFileSmall=qbP14plAq7ZxxBGSmLeYoA==, figureFileBig=MS9/BnxxqyFzn44MRLvWyw==, tableContent=null), ArticleFig(id=1241060191584711331, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=EN, label=Fig.4, caption=ROC curves of different algorithms on Sandiego, EI Segundo and Synthetic, figureFileSmall=T6ojuLU2N5T2ah5aLGfmGQ==, figureFileBig=TJZCeEv2FsneMA+AS+DmZQ==, tableContent=null), ArticleFig(id=1241060191660208808, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=CN, label=图4, caption=Sandiego、EI Segundo和Synthetic数据集在不同算法上的ROC曲线, figureFileSmall=T6ojuLU2N5T2ah5aLGfmGQ==, figureFileBig=TJZCeEv2FsneMA+AS+DmZQ==, tableContent=null), ArticleFig(id=1241060191777649325, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=EN, label=Table 1, caption=

AUC values of Sandiego, EI Segundo and Synthetic data sets in different detection algorithms

, figureFileSmall=null, figureFileBig=null, tableContent=
HSIAUC values(PD, PF)
SandiegoEI SegundoSynthetic
CEM0.990 40.829 80.979 5
ACE0.978 80.837 70.969 9
MF0.990 70.748 70.975 4
SID0.811 90.630 60.552 1
SAM0.806 10.591 80.623 8
BLTSC0.985 10.983 30.993 1
KDTGANs-0.975 20.981 90.986 5
KDTGANs+0.999 90.992 60.999 4
KDTGAN0.999 90.993 00.999 6
), ArticleFig(id=1241060191878312626, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=CN, label=表1, caption=

不同检测算法在Sandiego、EI Segundo和Synthetic数据集的AUC值

, figureFileSmall=null, figureFileBig=null, tableContent=
HSIAUC values(PD, PF)
SandiegoEI SegundoSynthetic
CEM0.990 40.829 80.979 5
ACE0.978 80.837 70.969 9
MF0.990 70.748 70.975 4
SID0.811 90.630 60.552 1
SAM0.806 10.591 80.623 8
BLTSC0.985 10.983 30.993 1
KDTGANs-0.975 20.981 90.986 5
KDTGANs+0.999 90.992 60.999 4
KDTGAN0.999 90.993 00.999 6
), ArticleFig(id=1241060191970587319, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=EN, label=Table 2, caption=

KDTGAN model ablation experiment on Sandiego dataset

, figureFileSmall=null, figureFileBig=null, tableContent=
模型FIDAUC
KDTGAN-LSTT38.610.991 7
KDTGAN-多尺度40.980.986 5
KDTGAN39.190.999 9
), ArticleFig(id=1241060192071250620, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241060179580612938, language=CN, label=表2, caption=

KDTGAN模型在Sandiego数据集消融实验

, figureFileSmall=null, figureFileBig=null, tableContent=
模型FIDAUC
KDTGAN-LSTT38.610.991 7
KDTGAN-多尺度40.980.986 5
KDTGAN39.190.999 9
)], attaches=null, journal=Journal(id=1238822691516493831, delFlag=0, nameCn=遥测遥控, nameEn=Journal of Telemetry, Tracking and Command, nameHistory1=null, nameHistory2=null, issn=2095-1000, eissn=null, cn=11-1780/TP, coden=null, periodic=1, 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=uUqWBk5pV4VWPULNoAHkuQ==, journalPrice=null, startedYear=null, abbrevIsoEn=Journal of Telemetry, Tracking and Command, journalRemark=null, publicationField=null, createdTime=1773287945871, updatedTime=1773299995823, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=J, firstLetterEn=J, subjectCode=Engineering, subjectName=null, subjectCodeEn=Engineering, subjectNameEn=null, picCn=uUqWBk5pV4VWPULNoAHkuQ==, picEn=RDDslHHvDBDNnJMjk6BxdA==, jcr=null, cjcr=null, exts=[JournalExt(id=1238873232732909900, language=CN, name=遥测遥控, 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=1773299995835, updatedTime=1773299995835, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://manuscript.spacejournal.cn/ycyk, submissionEditorUrl=https://manuscript.spacejournal.cn/ycyk, submissionReviewUrl=https://manuscript.spacejournal.cn/ycyk, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1238873232779047245, language=EN, name=Journal of Telemetry, Tracking and Command, 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=1773299995846, updatedTime=1773299995846, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://manuscript.spacejournal.cn/ycyk, submissionEditorUrl=https://manuscript.spacejournal.cn/ycyk, submissionReviewUrl=https://manuscript.spacejournal.cn/ycyk, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1238841944844054536, websiteList=[Website(id=1238843830011744952, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1238841944844054536, 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/ycyk/CN, language=CN, createTime=1773292985680, createBy=18614031015, updateTime=1773293577303, updateBy=18614031015, name=遥测遥控-中文, tplId=1146099689490845704, title=遥测遥控, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1238846390617568150, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=articleTextType, value=kx, createTime=1773293596176, updateTime=1773293596176, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390596596627, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=banner, value=null, createTime=1773293596171, updateTime=1773293596171, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390638539673, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=grayFlag, value=0, createTime=1773293596181, updateTime=1773293596181, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390588208018, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=logo, value=https://castjournals.cast.org.cn/joweb/ycyk/CN/file/pic?fileId=g9S2GGWx7iZjf51o9vpmmA==, createTime=1773293596169, updateTime=1773293596169, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390651122587, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=minRunFlag, value=0, createTime=1773293596184, updateTime=1773293596184, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390613373845, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/ycyk/CN/file/pic, createTime=1773293596175, updateTime=1773293596175, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390646928282, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=silenceFlag, value=0, createTime=1773293596183, updateTime=1773293596183, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390604985236, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1773293596173, updateTime=1773293596173, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390625956759, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=themeColor, value=null, createTime=1773293596178, updateTime=1773293596178, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390634345368, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=themeStyle, value=null, createTime=1773293596180, updateTime=1773293596180, creator=18614031015, updator=18614031015)]), Website(id=1238843830078853819, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1238841944844054536, 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/ycyk/EN, language=EN, createTime=1773292985696, createBy=18614031015, updateTime=1773293566698, updateBy=18614031015, name=遥测遥控-英文, tplId=1146101810881728533, title=Journal of Telemetry, Tracking and Command, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1238846419235304352, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=articleTextType, value=kx, createTime=1773293602999, updateTime=1773293602999, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419214332829, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=banner, value=null, createTime=1773293602994, updateTime=1773293602994, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419256275875, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=grayFlag, value=0, createTime=1773293603004, updateTime=1773293603004, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419205944220, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=logo, value=https://castjournals.cast.org.cn/joweb/ycyk/EN/file/pic?fileId=g9S2GGWx7iZjf51o9vpmmA==, createTime=1773293602992, updateTime=1773293602992, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419268858789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=minRunFlag, value=0, createTime=1773293603007, updateTime=1773293603007, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419231110047, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/ycyk/EN/file/pic, createTime=1773293602998, updateTime=1773293602998, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419264664484, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=silenceFlag, value=0, createTime=1773293603006, updateTime=1773293603006, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419222721438, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1773293602996, updateTime=1773293602996, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419243692961, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=themeColor, value=null, createTime=1773293603001, updateTime=1773293603001, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419252081570, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=themeStyle, value=null, createTime=1773293603003, updateTime=1773293603003, creator=18614031015, updator=18614031015)])], journalTitle=遥测遥控, weixinUrl=null, journalUrl=https://ycyk.spacejournal.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Journal of Telemetry, Tracking and Command, journalPhotoCn=uUqWBk5pV4VWPULNoAHkuQ==, journalPhotoEn=RDDslHHvDBDNnJMjk6BxdA==, journalFirstLetter=J, 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/ycyk/CN/10.12347/j.ycyk.20240119001, detailUrlEn=https://castjournals.cast.org.cn/joweb/ycyk/EN/10.12347/j.ycyk.20240119001, pdfUrlCn=https://castjournals.cast.org.cn/joweb/ycyk/CN/PDF/10.12347/j.ycyk.20240119001, pdfUrlEn=https://castjournals.cast.org.cn/joweb/ycyk/EN/PDF/10.12347/j.ycyk.20240119001, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
KDTGAN:基于Transformer-GAN和知识蒸馏的高光谱目标检测
收藏切换
PDF下载
谢雯 , 闪晨超 , 张哲哲 , 张嘉鹏
遥测遥控 | 人工智能技术 2024,45(2): 10-17
收起
收藏切换
遥测遥控 | 人工智能技术 2024, 45(2): 10-17
KDTGAN:基于Transformer-GAN和知识蒸馏的高光谱目标检测
全屏
谢雯, 闪晨超, 张哲哲, 张嘉鹏
作者信息
  • 西安邮电大学通信与信息工程学院 西安 710121
  • 谢雯 1989年生,副教授,研究生导师。

    闪晨超 1999年生,硕士研究生。

    张哲哲 1997年生,硕士研究生。

    张嘉鹏 1998年生,硕士研究生。

KDTGAN: Knowledge Distillation via Transformer GAN for Hyperspectral Target Detection
Wen XIE, Chenchao SHAN, Zhezhe ZHANG, Jiapeng ZHANG
Affiliations
  • School of Communications and Information Engineering, Xi'an University of Posts and Telecommunications, Xi'an 710121, China
doi: 10.12347/j.ycyk.20240119001
文章导航
收藏切换

高光谱目标检测在地球观测中至关重要,被广泛应用于军事和民用领域。然而,由于高光谱图像的背景复杂性和目标样本的有限性,该任务面临较大的挑战。本文首先采用CEM(约束能量最小化)粗检测方法提取背景数据。随之,引入了一种新的知识蒸馏模型,即KDTGAN(通过Transformer-GAN实现)。教师模型的生成器采用了Transformer编码器的结构,并结合多尺度数据融合的方法,能够准确地学习背景分布,进而通过重构背景信息实现目标检测。为了克服GAN(生成对抗网络)训练不稳定的挑战,特别是纯背景数据的稀缺性,本文提出了一种新的损失算法,以减小可疑目标样本对模型性能的负面影响。为了降低模型的计算负担,本文引入知识蒸馏,并设计新的蒸馏损失对学生模型加以约束,使模型轻量化的同时提高学生模型检测精度。实验结果表明:KDTGAN相较于当前检测方法表现更优,具有更高的检测精度和鲁棒性。

高光谱图像  /  目标检测  /  知识蒸馏  /  生成对抗网络  /  Transformer-GAN

Hyperspectral target detection is crucial in Earth observation for both military and civilian applications. However,complex backgrounds and the scarcity of target samples pose challenges in hyperspectral image analysis. In this paper, we first employ the CEM coarse detection method to extract background data. Subsequently, a novel knowledge distillation model, namely KDTGAN (implemented through Transformer-GAN), is introduced. The generator of this teacher model adopts the structure of a Transformer encoder and combines it with a multi-scale data fusion approach to accurately learn the background distribution, which in turn enables target detection by reconstructing the background information. To overcome the challenge of unstable GAN training,especially the scarcity of pure background data, we propose a new loss algorithm to reduce the negative impact of suspicious target samples on model performance. To reduce the computational burden of the model, we introduce knowledge distillation and design a new distillation loss to constrain the student model to lighten the model while improving the student model's detection accuracy. The experimental results show that KDTGAN performs better than current detection methods with higher detection accuracy and robustness.

Hyperspectral  /  Target detection  /  Knowledge distillation  /  GAN  /  Transformer-GAN
谢雯, 闪晨超, 张哲哲, 张嘉鹏. KDTGAN:基于Transformer-GAN和知识蒸馏的高光谱目标检测. 遥测遥控, 2024 , 45 (2) : 10 -17 . DOI: 10.12347/j.ycyk.20240119001
Wen XIE, Chenchao SHAN, Zhezhe ZHANG, Jiapeng ZHANG. KDTGAN: Knowledge Distillation via Transformer GAN for Hyperspectral Target Detection[J]. Journal of Telemetry, Tracking and Command, 2024 , 45 (2) : 10 -17 . DOI: 10.12347/j.ycyk.20240119001
高光谱图像呈现为一个三维立方体,蕴含了光谱和空间信息。与RGB图像不同,高光谱图像的波段范围可达400 nm~2 500 nm[1],因其丰富的光谱信息,在多个应用领域发挥着重要作用。目标检测作为一项基本任务,在民用和军事等领域受到广泛关注。传统高光谱目标检测方法通常依赖于目标与背景的光谱差异。例如,约束能量最小化(Constrained Energy Minimization,CEM)[2]通过施加目标和背景的约束,构建有限脉冲响应滤波器,使得滤波器输出能量最小化。基于稀疏和协同表示(Combined Sparse and Collaborative Repre-sentation,CSCR)[3]的组合方法则利用目标和背景的表示残差生成检测输出。自适应相关估计(Adaptive Correlation Estimation,ACE)[4]和匹配滤波器(Matched Filter,MF)[5]将目标检测视为假设检验问题。然而,传统方法往往忽略了高光谱图像所覆盖的大尺度地面场景的复杂性,其中包含具有不同光谱特性的各种物质[6]
随着深度学习方法的发展,出现了许多基于生成对抗网络(Generative Adversarial Networks,GAN)[7]、自编码器(Autoencoder,AE)、变分自编码器(Variational Autoencoder,VAE)[8]和对抗自编码器(Adversarial Autoencoder,AAE)的目标检测方法[9-13]。这些方法通过利用深度学习的重构特性,对高光谱数据进行了有效的数据增强,尤其在无标签高光谱数据集上表现显著。例如,秦等人提出的PTGAN模型[14],其通过设计双阶段加权GAN模型,充分利用GAN模型的重构能力,并且借用指数非线性融合策略,结合了两个阶段的鉴别器特征来抑制背景实现高光谱目标检测任务。然而,先前的方法[15,16]主要基于CNN(Convolutional Neural Networks,卷积神经网络)捕获高光谱图像的特征,而CNN的局部感受野机制使其难以学习远程的空谱信息。因此,一些学者在高光谱目标检测中引入了Transformer模型[17-20],以关注全局信息。例如,RAO等人提出的STTD模型[21],采用孪生网络的方法,其两条支路采用相同的Transformer结构,支路之间权重共享,最后通过相似度打分得到最终检测结果。LI等人提出的HTD-Former模型[22],采用分布式学习的方法,通过多进程支路并行的方式,减少了Transformer庞大的计算量问题。但是,这些仅使用光谱信息的模型在有效区分地面目标方面存在不足。鉴于此,本文提出了KDTGAN模型,主要工作包括:
① 采用CEM粗检测方法选择背景样本进行训练,在此基础上,挖掘Transformer在高光谱目标检测任务中背景学习的潜力。为此,提出了一个基于Transformer和GAN的模型框架,并通过整合跨尺度数据,充分考虑了数据在多个尺度上的变化,提高了模型的适应性,以有效地利用背景信息和全局高光谱信息。
② 为了解决训练过程目标样本造成的负面影响,提出了目标抑制损失函数,使生成结果更趋近于纯背景样本。
③ 采用知识蒸馏的方法对模型进行轻量化处理,提出了高效的蒸馏损失函数,通过将学生模型引入来替代教师模型,从而在减少模型计算量的同时提升学生模型性能。
本文首先,详细介绍了提出的KDTGAN模型,其次,给出了实验结果和相应分析,最后是本文结论。
本文提出了一种通过背景学习的GAN来进行目标检测。所提出的方法体现在三个主要方面:首先,提取背景训练样本,基于KDTGAN对网络进行训练,以获得唯一的背景表示;然后,通过加入蒸馏损失,对学生模型加以约束,降低模型计算量同时提升学生模型重构性能;最后,用训练好的模型来重建背景,通过光谱角距离度量来实现目标检测。
图1所示,提出的KDTGAN模型包括:教师模型生成器(图1(a))、学生模型生成器(图1(b))和教师模型鉴别器(图1(c))。
由于目标训练样本相对于背景训练样本较少,因此采用以CEM算法为基础的粗检测方法提取背景训练样本。高光谱数据集XєRH×W×C可以用高度H,宽度W,光谱带数为C的三维立方体表示。通过设定阈值将其划分为目标样本和背景样本。然后,从背景样本中提取训练样本和测试样本。
教师模型采用了GAN模型结构,其中包括生成器GT(·)和鉴别器DT(·),它们通过对抗博弈方式实现数据生成任务。设计的教师模型生成器的流程图如图1(a)所示。首先,引入了Mapping Network层[23],其中包含8层全连接层,以挖掘数据的深度特征信息;接着,对输出结果进行序列化展平处理,并分别通过[4,8,16,32]尺寸在通道维度的均值池化,实现多尺度数据划分;最后,使用Transformer结构对不同尺度的数据进行特征提取,并采用数据融合模块,将中间层信息进行加权融合。本文的Transformer模型结构参考了文献[24]中的模型设计,多头自注意力机制(MHSA)可以描述为query、key、value点乘映射,其中query、key、value和最后输出都是向量形式,输出计算为value的加权和,而权重则由query和key之间的相应函数计算而来。
式(1)中的QKV分别代表的query、key、value矩阵,dk是输入数据的维度。具有m个头数的多头自注意力表示为:
MHSA(X)=Concat(h1,h2,…,hm)W
其中,W代表相应的可学习矩阵。将每阶段Transformer输出的结果,分别沿空间维度1D上采样为统一大小,再以通道维度拼接和1D反卷积实现信息融合,最后通过MLP层,以得到生成器生成的伪数据
教师模型鉴别器结构如图1(c)所示,由三层卷积层和两层全连接层组成,卷积采用1D卷积,卷积核设置为3。为了稳定鉴别器的训练过程,在鉴别器中加入了光谱归一化层,以稳定鉴别器的权重,避免出现梯度消失或梯度爆炸的问题。对于鉴别器的输入,采用经过CEM粗检测筛选的背景训练样本进行特征学习,通过GAN的交互来促使生成器学到与背景分布相关的特征。
首先,训练鉴别器网络,使其能够区分真实样本和生成器生成的伪数据,为伪数据样本设定伪标签信息,通过计算交叉熵损失,优化鉴别器的参数,以增强其对生成器输出的对抗性。同时,训练生成器网络,将背景数据作为目标,使生成器能够学到背景分布的特征。生成器的输入是从潜在空间中随机采样的噪声。通过优化生成器的参数,使得生成器能够生成与背景数据相似的合成数据,这一训练过程的目标在于通过对抗训练提高生成器生成数据的逼真度,同时使其能够捕捉背景数据的特征。
由于模型训练目的是学习背景样本分布,而提取背景训练样本过程是采用CEM算法进行粗检测得到,所以训练样本中仍可能是会存在目标样本进而影响背景分布的学习,造成负面影响。为了克服此问题,本文提出了一个损失函数LSTT,与传统的损失函数相比,该方法在处理高光谱图像时能够更有效捕捉复杂的空间特征。通过度量生成器输出的伪数据和目标先验Di={d1,d2,...,diRC之间的相似性,抑制可能存在的目标样本,损失函数LSTT如下:
表示生成器得到的伪结果,表示对取均值,di表示目标先验样本。通过余弦相似度度量二者的相似度比值,即计算出生成器生成的伪数据样本是否和真实的目标样本接近,产生由目标样本干扰而得到的结果。其中皆要减去均值,目的是使得向量更关心方向而不是幅度,加上Relu激活函数,剔除小于0的值,仅保留正值,并且通过Top函数,取前K个向量的均值,以得到损失值进而抑制目标样本的存在。
由于教师模型采用了Transformer结构,导致计算量较大,但独立使用学生模型结构的检测结果却表现较差。因此,为了在降低模型计算量的同时不损失模型的性能,采用了知识蒸馏的方法。在这个过程中,以离线蒸馏的方式协助学生网络生成器GS(·)与已训练好的DT(·)进行博弈,仅更新生成器权重而不对鉴别器权重进行更新。学生网络生成器如图1(b)所示,采用一层卷积层、两层全连接层和非线性激活函数构建。然而,蒸馏方式的选择会极大程度影响蒸馏效果,无监督模型需要从数据中学到更通用和抽象的表示,而这些表示通常存在于网络的中间层。因此,引用关系知识蒸馏(Relational Knowledge Distillation,RKD)的思想[25],输出DT(·)的(M-1)个中间层特征和最终结果,并分别按照空间维度和光谱维度将知识传递给学生网络,如图2所示。同样,对学生网络的生成器输出(M-1)个中间层特征和最终结果,将两者都输入损失函数,以实现蒸馏过程。因此,所设计的蒸馏损失如下:
公式(4)为马氏距离,其中S表示输入数据的协方差矩阵。公式(5)中,,分别表示教师网络和学生网络输出,di为目标先验,θ为设定的固定值,使得二者距离保持固定距离间隔,防止距离过小而梯度消失,(·)+代表了差值小于0时,以0取代,不算入损失。其中,将二者做差计算,确保同一类别之间的间距小于不同类别的距离,实现同类接近、异类远离。公式(6)表示将GT(·)输出的(M-1)个中间层特征和最终结果计算的损失值进行求和,以得到最终蒸馏损失。
高光谱数据集为三维立方体结构,包含空间和光谱信息,因此在知识蒸馏过程中,将教师网络的知识传递给学生时通过空间和光谱两个维度分别进行,如图2所示。通过3D卷积设置不同的卷积核,提取到不同的特征信息,其中空间维度卷积核的C表示通道维数,再经过批归一化和Relu非线性激活函数稳定数据分布,最后残差连接以提取到不同的特征信息。
由于学生生成器和教师网络鉴别器进行对抗性训练,因此在优化学生模型时加入经过Sigmoid的交叉熵损失,最终学生模型的损失函数表示为,
Ls=LBCE+Ld
在训练阶段,生成器接收随机噪声作为输入,这有助于引入一定程度的随机性和多样性,从而使生成器能够生成更具创造性的样本。鉴别器接收原始背景分布的数据,以区分生成器生成的样本与真实样本的差异。在最后检测阶段,由于生成器已经在训练中学到了背景分布的特征,将高光谱数据集X输入到已学习到背景分布的Gs(·)中,得到的重构结果可以被视为生成器尝试去除目标样本,重构出符合训练中学到的背景分布的样本,实现对目标和背景样本的重构,重构的结果反映了生成器在学习阶段所学到的背景分布特征,同时将目标样本视为“噪声”(在生成器学到的背景分布之外的信息),以产生与训练数据相符的生成结果,扩大目标样本和背景样本分布的差异性,得到重构结果与高光谱数据集X通过光谱角距离度量,获得权重值,最终再与粗检测结果进行点乘以得到检测结果图。
所有实验均在配备NVIDIA RTX 4070Ti GPU和Intel Core i9-13900K CPU的台式机上进行,并在Sandiego、EI Segundo和Synthetic高光谱数据集上测试。采用自适应时刻估计法(Adaptive Moment Estimation,Adam) 以对模型权重优化,其中epoch设置500,生成器学习率设定为0.000 1,鉴别器的学习率为0.000 01。使用曲线下的面积(Area Under the Curve, AUC)值和接收机工作特征曲线(Receiver Operating Characteristic curve, ROC)来显示检测精度和效果。
Sandiego:第一个数据集是由AVIRIS传感器获得的,覆盖了美国圣地亚哥机场范围,空间大小为200像素×200像素,包括224个光谱通道,波长从370 nm~2 510 nm。空间分辨率约为3.5 m,去除吸水带和低信噪比波段后,共使用了189个波段,三架飞机被视为目标像素占134个像素值。
EI Segundo:第二个数据集是由AVIRIS传感器捕获,它有224个光谱通道,波长从366 nm~2 496 nm。这个城市场景覆盖了美国加州的一个地区。空间大小为250像素×300像素,其中主要几个居民区,公园和炼油区被视为目标像素,占2 048像素。
Synthetic:第三个数据集是由美国地质调查局(USGS)的数字光谱库合成的高光谱数据集,空间大小为64像素×64像素,其中包含有15种矿物端元的数据合成,选择Labradorite作为目标,像素值占12。
为了评估提出模型的检测性能,本文采用CEM、ACE、MF、SID、SAM和BLTSC检测方法进行比较,如图3所示。其中,图3(a)图3(b)分别为Sandiego、EI Segundo和Synthetic数据集的伪彩图和地面真值图,图3(c)~图3(j)依次为SID、SAM、CEM、ACE、MF、BLTSC、KDTGANs-、KDTGANs+在数据集上的检测图。
特别注意的是,本文在Synthetic数据集测试时加入了信噪比为20 dB的高斯白噪声,使用KDTGAN表示教师模型,KDTGANs+表示蒸馏后的学生模型,KDTGANs-表示未蒸馏的学生模型,可以观察到我们提出的KDTGANs+模型突出了目标,与其他检测方法相比对背景有抑制作用。为了对检测性能进行详细的定性分析,本文绘制了(PD,PF)的对数尺度ROC曲线,如图4所示,以用来评估检测和背景抑制性能,并将检测结果进行了可视化操作,对应的AUC值如表1所示。从图4可以明显看出,KDTGANs+模型具有较好的背景抑制能力。
公式(8)中,Psignal表示数据的功率,Pnoise表示噪声的功率。
为评估提出的模型各模块的有效性,对教师模型KDTGAN和学生模型KDTGAN+进行了消融实验,采用弗雷谢起始距离(Fréchet Inception Distance,FID)分数来衡量GAN训练过程的重构精度受本文提出的模块的影响程度,并对不同模块减去后的检测精度进行对比,结果如表2所示。针对学生模型,如3.2.1中展示的KDTGANs+和KDTGANs-以证明本文提出的蒸馏损失,大大提高了学生模型的检测精度。
本研究提出了一种基于知识蒸馏的轻量级Transformer-GAN模型,重点关注高光谱目标检测。通过设计生成器结构,旨在使模型更专注于捕获高光谱图像中的全局信息,并构建一个损失函数来应对背景训练样本不纯的挑战。通过引入蒸馏损失,巧妙地应用了知识蒸馏策略,从光谱和空间角度促进了高层次信息向学生模型的转移,达到计算量降低的同时,重构纯背景能力不减,并实现高光谱目标检测。实验结果表明:在高光谱目标检测方面,本文提出的KDTGAN模型明显优于其他检测器。
  • 国家自然科学基金(61901365; 62071379)
  • 陕西省自然科学基金(2019JQ-377)
  • 陕西省教育厅专项科研计划(19JK0805)
  • 西安邮电大学西邮新星团队项目(xyt2016-01)
参考文献 引证文献
排序方式:
[1]
Microimages. Introduction to hyperspectral imaging[EB/OL].[2024-01-19] https://www.microimages.com/documentation/Tutorials/hyprspec.pdf.
[2]
FARRAND W H, HARSANYI J C. Mapping the distribution of mine tailings in the Coeur d'Alene River Valley, Idaho, through the use of a constrained energy minimization technique[J]. Remote Sensing of Environment, 1997, 59(1): 64-76.
[3]
LI W, DU Q, ZHANG B. Combined sparse and collaborative representation for hyperspectral target detection[J]. Pattern Recognition, 2015, 48(12): 3904-3916.
[4]
KRAUT S, SCHARF L L, MCWHORTER L T. Adaptive subspace detectors[J]. IEEE Transactions on Signal Processing, 2001, 49(1): 1-16.
[5]
FUHRMANN D R, KELLY E J, NITZBERG R. A CFAR adaptive matched filter detector[J]. IEEE Transactions on Aerospace and Electronic Systems, 1992, 28(1): 208-216.
[6]
赵佳乐, 王广龙, 周冰, 等. 深度学习下高光谱图像目标检测技术研究进展[J]. 激光杂志, 2022, 43(10): 1-6.
[7]
GOODFELLOW I, POUGET-ABADIE J, MIRZA M,et al. Generative adversarial nets[C]//NIPS'14: Procee-dings of the 27th International Conference on Neural Information Processing Systems, December 8-13, 2014, Montreal Canada. Cambridge: MIT Press, 2014.
[8]
KINGMA D P, WELLING M. Auto-encoding variational Bayes[EB/OL]. arXiv preprint, 2013 arXiv:
[9]
XIE Weiying, ZHANG Xin, LI Yunsong, et al. Background learning based on target suppression constraint for hyperspectral target detection[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020, 13: 5887-5897.
[10]
SHEN Kun, XIE Weixin, TANG Haojin, et al. Dense convolution siamese network for hyperspectral image target detection[C]//2022 5th International Conference on Information Communication and Signal Processing(ICICSP), November 26-28, 2022, Shenzhen, China. IEEE, 2022.
[11]
CHEN Xi, WANG Yulei, CHE Zongwei, et al. Contrastive learning for hyperspectral target detection[C]//IGARSS 2022-2022 IEEE International Geoscience and Remote Sensing Symposium, July 17-22, 2022, Kuala Lumpur, Malaysia. IEEE, 2022.
[12]
XIE Weiying, YANG Jian, LEI Jie, et al. SRUN: Spectral regularized unsupervised networks for hyperspectral target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 58(2): 1463-1474.
[13]
白凯玮. 基于分布式深度学习的高光谱图像目标检测算法研究[D].西安: 西安电子科技大学, 2022.
[14]
QIN Haonan, XIE Weiying, LI Yunsong, et al. PTGAN:A proposal-weighted two-stage GAN with attention for hyperspectral target detection[C]//2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS, July 11-16, 2021, Brussels, Belgium. IEEE, 2021.
[15]
鲍亮. 基于深度学习的高光谱遥感图像目标检测算法[D]. 南京: 南京理工大学, 2019.
[16]
解静. 基于深度学习的高光谱图像目标检测算法研究[D]. 西安: 西安电子科技大学, 2019.
[17]
KIM J H, KIM J, JOUNG J. Siamese hyperspectral target detection using synthetic training data[J]. Electro-nics Letters, 2020, 56(21): 1116-1118.
[18]
QIN Haonan, XIE Weiying, LI Yunsong, et al. HTD-VIT: Spectral-spatial joint hyperspectral target detection with vision transformer[C]//IGARSS 2022-2022 IEEE International Geoscience and Remote Sensing Symposium, July 17-22, 2022, Kuala Lumpur, Malaysia. IEEE, 2022.
[19]
JIAO Jinyue, GONG Zhiqiang, ZHONG Ping. Triplet spectral-wise transformer network for hyperspectral target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 5519817.
[20]
SHEN Dunbin, MA Xiaorui, KONG Wenfeng, et al. Hyperspectral target detection based on interpretable representation network[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 5519416.
[21]
RAO Weiqiang, GAO Lianru, QU Ying, et al. Siamese transformer network for hyperspectral image target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 5526419.
[22]
LI Yunsong, QIN Haonan, XIE Weiying. HTDFormer:Hyperspectral target detection based on transformer with distributed learning[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 5524715.
[23]
KARRAS Tero, LAINE Samuli, AILA Timo. A style-based generator architecture for generative adversarial networks[C]// 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 15-20, 2019, Long Beach, CA, USA. IEEE, 2019.
[24]
VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[EB/OL]. https://doi.org/10.48550/arXiv.1706.03762.
[25]
PARK W, KIM D, LU Y, et al. Relational knowledge distillation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 15-20, 2019, Long Beach, CA, USA. IEEE, 2019: 3967-3976.
2024年第45卷第2期
PDF下载
123
58
引用本文
BibTeX
文章信息
doi: 10.12347/j.ycyk.20240119001
  • 接收时间:2024-01-19
  • 首发时间:2026-03-18
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-01-19
  • 修回日期:2024-03-05
基金
国家自然科学基金(61901365; 62071379)
陕西省自然科学基金(2019JQ-377)
陕西省教育厅专项科研计划(19JK0805)
西安邮电大学西邮新星团队项目(xyt2016-01)
作者信息
    西安邮电大学通信与信息工程学院 西安 710121
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/ycyk/CN/10.12347/j.ycyk.20240119001
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