Article(id=1241757968581055430, tenantId=1146029695717560320, journalId=1238841944844054536, issueId=1241757966894945217, articleNumber=null, orderNo=null, doi=10.12347/j.ycyk.20240129001, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1706457600000, receivedDateStr=2024-01-29, revisedDate=1708272000000, revisedDateStr=2024-02-19, acceptedDate=null, acceptedDateStr=null, onlineDate=1773987770457, onlineDateStr=2026-03-20, pubDate=1720972800000, pubDateStr=2024-07-15, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1773987770457, onlineIssueDateStr=2026-03-20, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1773987770457, creator=13701087609, updateTime=1773987770457, updator=13701087609, issue=Issue{id=1241757966894945217, tenantId=1146029695717560320, journalId=1238841944844054536, year='2024', volume='45', issue='4', pageStart='1', pageEnd='136', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1773987770054, creator=13701087609, updateTime=1773991490516, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1241773571714909111, tenantId=1146029695717560320, journalId=1238841944844054536, issueId=1241757966894945217, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1241773571714909112, tenantId=1146029695717560320, journalId=1238841944844054536, issueId=1241757966894945217, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=31, endPage=44, ext={EN=ArticleExt(id=1241757968828519367, articleId=1241757968581055430, tenantId=1146029695717560320, journalId=1238841944844054536, language=EN, title=Hyperspectral Target Detection Based on Sample Enhancement and Automatic Parameter Optimization, columnId=1241060179823882575, journalTitle=Journal of Telemetry, Tracking and Command, columnName=Artificial Intelligence Technology, runingTitle=null, highlight=null, articleAbstract=

Hyperspectral target detection based on deep learning faces challenges such as insufficient quality of samples, intricate network structures, and laborious parameter adjustment. In this paper, we propose a deep learning method with data augmentation and automatic hyperparameter optimization. To tackle the issue of insufficient quality of samples, we introduce a sample augmentation strategy. The strategy utilizes endmember extraction and clustering techniques to directly acquire a large number of background pixels from hyperspectral images. By pairing these with a small number of known target pixels using a phase-reducing pixel pairing approach, we obtain a large number of labeled pure sample pairs, thereby accomplishing data augmentation. In addition, distinct from most complex deep networks, we designed a lightweight Convolutional Neural Network (CNN) comprised of 12 convolutional layers. This network is specifically engineered to efficiently and rapidly learn the mapping between input sample pairs and their corresponding labels. By incorporating the particle swarm optimization algorithm, this network possesses the capability to automatically optimize hyperparameters, overcoming the shortcomings of laborious parameter adjustment. This enables the network to automatically adjust hyperparameters based on samples from different hyperspectral images, thereby generating optimal results. For a test pixel, the input to the trained network is the spectral difference between the central pixel and its adjacent pixels. When a test pixel belongs to the target, the output score is closely align with the target label. Experimental results on five hyperspectral datasets demonstrate that our method significantly outperforms existing techniques.

, 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=Hao LIU, Mingming XU, Biaoqun SHEN, Shanwei LIU, Hui SHENG), CN=ArticleExt(id=1241757977183572114, articleId=1241757968581055430, tenantId=1146029695717560320, journalId=1238841944844054536, language=CN, title=基于样本增强和自动参数优化的高光谱目标检测方法, columnId=1241060179987460434, journalTitle=遥测遥控, columnName=人工智能技术, runingTitle=null, highlight=null, articleAbstract=

基于深度学习的高光谱目标检测面临着样本质量不足、网络结构复杂、参数调整费力等问题,本文提出了一种具有数据增强和自动超参数优化的深度学习方法。为了解决样本质量不足的问题,本文引入了一种样本扩增策略。该策略利用端元提取和聚类技术直接从高光谱图像中获取大量背景像素,通过使用相减像素配对方法将这些像素与少量已知目标像素配对,获得大量标记的纯样本对,从而实现数据增强。此外,与大多数复杂的深度网络不同,本文设计了一个由12个卷积层组成的轻量级卷积神经网络(CNN)。该网络专门设计用于高效快速地学习输入样本对与其对应标签之间的映射。结合粒子群优化算法,该网络具有超参数自动优化的能力,克服了参数调整费力的缺点,这使得网络能够根据来自不同高光谱图像的样本自动调整超参数,从而产生最优结果。对于测试像素,训练网络的输入是中心像素与其相邻像素之间的光谱差。当一个测试像素属于目标时,输出分数接近1,反之则接近0。在五个高光谱数据集上的实验结果表明:本文提出的方法明显优于现有的技术。

, correspAuthors=null, authorNote=null, correspAuthorsNote=
沈彪群()
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=0w1wacMst3iehCNVTYveHg==, magXml=QVp5dtylLMnyBbhGQvqVkw==, pdfUrl=null, pdf=+NsJzH4Gi7R8zdXQ6e6qeg==, pdfFileSize=4668672, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=SqFfVDYk9HV0/ziuiz4idQ==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=Yirkr2VZhWkBfWB1vA4P4Q==, mapNumber=null, authorCompany=null, fund=null, authors=

刘浩 1998年生,硕士研究生。

许明明 1990年生,副教授,硕士生导师。

沈彪群 1982年生,本科,高级工程师。

刘善伟 1982年生,教授,博士生导师。

盛辉 1972年生,副教授,硕士生导师。

, authorsList=刘浩, 许明明, 沈彪群, 刘善伟, 盛辉)}, authors=[Author(id=1241788818429379364, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, 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=1241788818538431271, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, authorId=1241788818429379364, language=EN, stringName=Hao LIU, firstName=Hao, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.College of Oceanography and Spatial Information, China University of Petroleum (East China), Qingdao 266580, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1241788818597151529, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, authorId=1241788818429379364, language=CN, stringName=刘浩, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1中国石油大学(华东)海洋与空间信息学院 青岛 266580, bio={"content":"

刘浩 1998年生,硕士研究生。

"}, bioImg=null, bioContent=

刘浩 1998年生,硕士研究生。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1241788818232247066, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, xref=1., ext=[AuthorCompanyExt(id=1241788818236441371, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, companyId=1241788818232247066, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.College of Oceanography and Spatial Information, China University of Petroleum (East China), Qingdao 266580, China), AuthorCompanyExt(id=1241788818244829980, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, companyId=1241788818232247066, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1中国石油大学(华东)海洋与空间信息学院 青岛 266580)])]), Author(id=1241788818681037614, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, 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=1241788818785895215, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, authorId=1241788818681037614, language=EN, stringName=Mingming XU, firstName=Mingming, middleName=null, lastName=XU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.College of Oceanography and Spatial Information, China University of Petroleum (East China), Qingdao 266580, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1241788818861392690, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, authorId=1241788818681037614, language=CN, stringName=许明明, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1中国石油大学(华东)海洋与空间信息学院 青岛 266580, bio={"content":"

许明明 1990年生,副教授,硕士生导师。

"}, bioImg=null, bioContent=

许明明 1990年生,副教授,硕士生导师。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1241788818232247066, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, xref=1., ext=[AuthorCompanyExt(id=1241788818236441371, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, companyId=1241788818232247066, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.College of Oceanography and Spatial Information, China University of Petroleum (East China), Qingdao 266580, China), AuthorCompanyExt(id=1241788818244829980, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, companyId=1241788818232247066, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1中国石油大学(华东)海洋与空间信息学院 青岛 266580)])]), Author(id=1241788818953667381, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=shenbiaoqun1982@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1241788819037553464, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, authorId=1241788818953667381, language=EN, stringName=Biaoqun SHEN, firstName=Biaoqun, middleName=null, lastName=SHEN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.Shandong Lubang Geographic Information Engineering Co., LTD. Jinan 250102, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1241788819113050938, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, authorId=1241788818953667381, language=CN, stringName=沈彪群, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2山东鲁邦地理信息工程有限公司 济南 250102, bio={"content":"

沈彪群 1982年生,本科,高级工程师。

"}, bioImg=null, bioContent=

沈彪群 1982年生,本科,高级工程师。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1241788818337104671, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, xref=2., ext=[AuthorCompanyExt(id=1241788818345493279, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, companyId=1241788818337104671, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.Shandong Lubang Geographic Information Engineering Co., LTD. Jinan 250102, China), AuthorCompanyExt(id=1241788818353881888, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, companyId=1241788818337104671, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2山东鲁邦地理信息工程有限公司 济南 250102)])]), Author(id=1241788819196937021, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, 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=1241788819280823105, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, authorId=1241788819196937021, language=EN, stringName=Shanwei LIU, firstName=Shanwei, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.College of Oceanography and Spatial Information, China University of Petroleum (East China), Qingdao 266580, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1241788819347931972, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, authorId=1241788819196937021, language=CN, stringName=刘善伟, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1中国石油大学(华东)海洋与空间信息学院 青岛 266580, bio={"content":"

刘善伟 1982年生,教授,博士生导师。

"}, bioImg=null, bioContent=

刘善伟 1982年生,教授,博士生导师。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1241788818232247066, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, xref=1., ext=[AuthorCompanyExt(id=1241788818236441371, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, companyId=1241788818232247066, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.College of Oceanography and Spatial Information, China University of Petroleum (East China), Qingdao 266580, China), AuthorCompanyExt(id=1241788818244829980, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, companyId=1241788818232247066, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1中国石油大学(华东)海洋与空间信息学院 青岛 266580)])]), Author(id=1241788819415040839, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1241788819494732618, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, authorId=1241788819415040839, language=EN, stringName=Hui SHENG, firstName=Hui, middleName=null, lastName=SHENG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.College of Oceanography and Spatial Information, China University of Petroleum (East China), Qingdao 266580, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1241788819561841483, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, authorId=1241788819415040839, language=CN, stringName=盛辉, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1中国石油大学(华东)海洋与空间信息学院 青岛 266580, bio={"content":"

盛辉 1972年生,副教授,硕士生导师。

"}, bioImg=null, bioContent=

盛辉 1972年生,副教授,硕士生导师。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1241788818232247066, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, xref=1., ext=[AuthorCompanyExt(id=1241788818236441371, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, companyId=1241788818232247066, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.College of Oceanography and Spatial Information, China University of Petroleum (East China), Qingdao 266580, China), AuthorCompanyExt(id=1241788818244829980, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, companyId=1241788818232247066, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1中国石油大学(华东)海洋与空间信息学院 青岛 266580)])])], keywords=[Keyword(id=1241788819733807953, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, orderNo=1, keyword=Hyperspectral), Keyword(id=1241788819813499732, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, orderNo=2, keyword=Target detection), Keyword(id=1241788819888997207, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, orderNo=3, keyword=Data augmentation), Keyword(id=1241788819977077594, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, orderNo=4, keyword=Convolutional neural network), Keyword(id=1241788820048380764, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, orderNo=1, keyword=高光谱), Keyword(id=1241788820123878239, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, orderNo=2, keyword=目标检测), Keyword(id=1241788820199375714, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, orderNo=3, keyword=样本增强), Keyword(id=1241788820274873188, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, orderNo=4, keyword=卷积神经网络)], refs=[Reference(id=1241788825186403249, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=10.26914/c.cnkihy.2020.050666, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=131, pageEnd=137, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=王盛铭, 王涛, 国家新闻出版广电总局中国新闻文化促进会学术期刊专业委员会, journalName=null, refType=null, unstructuredReference=王盛铭, 王涛. 高光谱图像目标检测综述[C]//国家新闻出版广电总局中国新闻文化促进会学术期刊专业委员会. 2020年第三届智慧教育与人工智能发展国际学术会议论文集(第二卷), 2020: 131-137., articleTitle=高光谱图像目标检测综述, refAbstract=null), Reference(id=1241788825253512114, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2013, volume=31, issue=1, pageStart=24, pageEnd=33, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=MANOLAKIS D, TRUSLOW E, PIEPER M, journalName=IEEE Signal Processing Magazine, refType=null, unstructuredReference=MANOLAKIS D, TRUSLOW E, PIEPER M, et al. Detection algorithms in hyperspectral imaging systems: An overview of practical algorithms[J]. IEEE Signal Processing Magazine, 2013, 31(1): 24-33., articleTitle=Detection algorithms in hyperspectral imaging systems: An overview of practical algorithms, refAbstract=null), Reference(id=1241788825312232371, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2023, volume=52, issue=6, pageStart=932, pageEnd=943, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=宋尚真, 杨怡欣, 王会峰, journalName=测绘学报, refType=null, unstructuredReference=宋尚真, 杨怡欣, 王会峰, . 高光谱图像稀疏约束与自编码器特征提取相结合的异常检测方法[J].测绘学报, 2023, 52(6): 932-943., articleTitle=高光谱图像稀疏约束与自编码器特征提取相结合的异常检测方法, refAbstract=null), Reference(id=1241788826792821686, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2023, volume=52, issue=6, pageStart=932, pageEnd=943, url=null, language=null, rfNumber=[3], rfOrder=3, authorNames=SONG Shangzhen, YANG Yixin, WANG Huifeng, journalName=Acta Geodaetica et Cartographica Sinica, refType=null, unstructuredReference=SONG Shangzhen, YANG Yixin, WANG Huifeng, et al. Hyperspectral anomaly detection combining sparse constraint and feature extraction via stacked autoencoder[J].Acta Geodaetica et Cartographica Sinica, 2023, 52(6):932-943., articleTitle=Hyperspectral anomaly detection combining sparse constraint and feature extraction via stacked autoencoder, refAbstract=null), Reference(id=1241788826893484984, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2022, volume=25, issue=null, pageStart=2710, pageEnd=2722, url=null, language=null, rfNumber=[4], rfOrder=4, authorNames=CHEN L, LIU J, CHEN W, journalName=IEEE Transactions on Multimedia, refType=null, unstructuredReference=CHEN L, LIU J, CHEN W, et al. A GLRT-based multi-pixel target detector in hyperspectral imagery[J]. IEEE Transactions on Multimedia, 2022, 25: 2710-2722., articleTitle=A GLRT-based multi-pixel target detector in hyperspectral imagery, refAbstract=null), Reference(id=1241788826968982458, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2021, volume=130, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[5], rfOrder=5, authorNames=ZHANG C, ZUO R, XIONG Y, journalName=Applied Geochemistry,, refType=null, unstructuredReference=ZHANG C, ZUO R, XIONG Y. Detection of the multivariate geochemical anomalies associated with mineralization using a deep convolutional neural network and a pixelpair feature method[J]. Applied Geochemistry, 2021, 130: 104994., articleTitle=Detection of the multivariate geochemical anomalies associated with mineralization using a deep convolutional neural network and a pixelpair feature method, refAbstract=null), Reference(id=1241788827040285626, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=10.27410/d.cnki.gxbfu.2021.000546, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[6], rfOrder=6, authorNames=陈静, journalName=null, refType=null, unstructuredReference=陈静. 基于深度学习的高光谱图像分类算法研究[D].兰州: 西北师范大学, 2021., articleTitle=基于深度学习的高光谱图像分类算法研究, refAbstract=null), Reference(id=1241788827111588796, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2020, volume=12, issue=7, pageStart=1056, pageEnd=null, url=null, language=null, rfNumber=[7], rfOrder=7, authorNames=FU X, SHANG X, SUN X, journalName=Remote Sensing, refType=null, unstructuredReference=FU X, SHANG X, SUN X, et al. Underwater hyperspectral target detection with band selection[J]. Remote Sensing, 2020, 12(7): 1056., articleTitle=Underwater hyperspectral target detection with band selection, refAbstract=null), Reference(id=1241788827233223613, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=3153, pageEnd=3161, url=null, language=null, rfNumber=[8], rfOrder=8, authorNames=ARAD B, BEN-SHAHAR O, journalName=null, refType=null, unstructuredReference=ARAD B, BEN-SHAHAR O. Filter selection for hyperspectral estimation[C]//Proceedings of the IEEE International Conference on Computer Vision. 2017: 3153-3161., articleTitle=Filter selection for hyperspectral estimation, refAbstract=null), Reference(id=1241788827338081214, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2024, volume=51, issue=2, pageStart=157, pageEnd=169, url=null, language=null, rfNumber=[9], rfOrder=9, authorNames=蔡明娇, 蒋俊正, 蔡万源, journalName=西安电子科技大学学报, refType=null, unstructuredReference=蔡明娇, 蒋俊正, 蔡万源, . 张量分解和自适应图全变分的高光谱图像去噪[J/OL].西安电子科技大学学报, 2024, 51(2): 157-169[2024-01-29]. https://doi.org/10.19665/j.issn1001-2400.20230412., articleTitle=张量分解和自适应图全变分的高光谱图像去噪, refAbstract=null), Reference(id=1241788827421967295, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=1992, volume=28, issue=1, pageStart=208, pageEnd=216, url=null, language=null, rfNumber=[10], rfOrder=10, authorNames=ROBEY F C, FUHRMANN D R, KELLY E J, journalName=IEEE Transactions on Aerospace and Electronic Systems, refType=null, unstructuredReference=ROBEY F C, FUHRMANN D R, KELLY E J, et al. 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=1241788827493270464, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2021, volume=53, issue=12, pageStart=10, pageEnd=15, url=null, language=null, rfNumber=[11], rfOrder=11, authorNames=谢少彪, 齐乃明, 秦雷, journalName=哈尔滨工业大学学报, refType=null, unstructuredReference=谢少彪, 齐乃明, 秦雷, . 高光谱亚像元图像仿真与异常检测应用[J]. 哈尔滨工业大学学报, 2021, 53(12):10-15., articleTitle=高光谱亚像元图像仿真与异常检测应用, refAbstract=null), Reference(id=1241788827564573633, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2021, volume=53, issue=12, pageStart=10, pageEnd=15, url=null, language=null, rfNumber=[11], rfOrder=12, authorNames=XIE Shaobiao, QI Naiming, QIN Lei, journalName=Journal of Harbin Institute of Technology, refType=null, unstructuredReference=XIE Shaobiao, QI Naiming, QIN Lei, et al. Hyperspectral subpixel image simulation and application to anomaly detection[J]. Journal of Harbin Institute of Technology, 2021, 53(12):10-15., articleTitle=Hyperspectral subpixel image simulation and application to anomaly detection, refAbstract=null), Reference(id=1241788827635876802, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2016, volume=20, issue=5, pageStart=1091, pageEnd=1101, url=null, language=null, rfNumber=[12], rfOrder=13, authorNames=张良培, 李家艺, journalName=遥感学报, refType=null, unstructuredReference=张良培, 李家艺. 高光谱图像稀疏信息处理综述与展望[J]. 遥感学报, 2016, 20(5): 1091-1101., articleTitle=高光谱图像稀疏信息处理综述与展望, refAbstract=null), Reference(id=1241788827698791363, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2016, volume=20, issue=5, pageStart=1091, pageEnd=1101, url=null, language=null, rfNumber=[12], rfOrder=14, authorNames=ZHANG Liangpei, LI Jiayi, journalName=Journal of Remote Sensing,, refType=null, unstructuredReference=ZHANG Liangpei, LI Jiayi. Development and prospect of sparse representation-based hyperspectral image processing and analysis[J]. Journal of Remote Sensing, 2016, 20(5): 1091-1101., articleTitle=Development and prospect of sparse representation-based hyperspectral image processing and analysis, refAbstract=null), Reference(id=1241788827761705924, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2013, volume=34, issue=9, pageStart=1171, pageEnd=1178, url=null, language=null, rfNumber=[13], rfOrder=15, authorNames=赵春晖, 李晓慧, 朱海峰, journalName=哈尔滨工程大学学报, refType=null, unstructuredReference=赵春晖, 李晓慧, 朱海峰. 空间4-邻域稀疏表示的高光谱图像目标检测[J]. 哈尔滨工程大学学报, 2013, 34(9): 1171-1178., articleTitle=空间4-邻域稀疏表示的高光谱图像目标检测, refAbstract=null), Reference(id=1241788827824620485, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2013, volume=34, issue=9, pageStart=1171, pageEnd=1178, url=null, language=null, rfNumber=[13], rfOrder=16, authorNames=ZHAO Chunhui, LI Xiaohui, ZHU Haifeng, journalName=Journal of Harbin Engineering University, refType=null, unstructuredReference=ZHAO Chunhui, LI Xiaohui, ZHU Haifeng. Hyperspectral imaging target detection algorithm based on spatial 4 neighborhoods for sparse representation[J]. Journal of Harbin Engineering University, 2013, 34(9): 1171-1178., articleTitle=Hyperspectral imaging target detection algorithm based on spatial 4 neighborhoods for sparse representation, refAbstract=null), Reference(id=1241788827904312262, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2022, volume=46, issue=2, pageStart=199, pageEnd=205, url=null, language=null, rfNumber=[14], rfOrder=17, authorNames=张帅洋, 华文深, 刘杰, journalName=激光技术, refType=null, unstructuredReference=张帅洋, 华文深, 刘杰, . 基于超像素和低秩的协同稀疏高光谱解混[J]. 激光技术, 2022, 46(2): 199-205., articleTitle=基于超像素和低秩的协同稀疏高光谱解混, refAbstract=null), Reference(id=1241788827967226823, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2022, volume=46, issue=2, pageStart=199, pageEnd=205, url=null, language=null, rfNumber=[14], rfOrder=18, authorNames=ZHANG Shuaiyang, HUA Wenshen, LIU Jie, journalName=Laser Technology, refType=null, unstructuredReference=ZHANG Shuaiyang, HUA Wenshen, LIU Jie, et al. Superpixels and low rank for collaborative sparse hyperspectral unmixing[J]. Laser Technology, 2022, 46(2):199-205., articleTitle=Superpixels and low rank for collaborative sparse hyperspectral unmixing, refAbstract=null), Reference(id=1241788828063695816, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2020, volume=59, issue=2, pageStart=1487, pageEnd=1500, url=null, language=null, rfNumber=[15], rfOrder=19, authorNames=ZHU D, DU B, ZHANG L, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=ZHU D, DU B, ZHANG L. Single-spectrum-driven binary-class sparse representation target detector for hyperspectral imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 59(2): 1487-1500., articleTitle=Single-spectrum-driven binary-class sparse representation target detector for hyperspectral imagery, refAbstract=null), Reference(id=1241788828143387593, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2008, volume=5, issue=2, pageStart=311, pageEnd=314, url=null, language=null, rfNumber=[16], rfOrder=20, authorNames=JIA X, RICHARDS J A, journalName=IEEE Geoscience and Remote Sensing Letters,, refType=null, unstructuredReference=JIA X, RICHARDS J A. Managing the spectral-spatial mix in context classification using Markov random fields[J]. IEEE Geoscience and Remote Sensing Letters, 2008, 5(2): 311-314., articleTitle=Managing the spectral-spatial mix in context classification using Markov random fields, refAbstract=null), Reference(id=1241788828202107850, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2010, volume=49, issue=1, pageStart=388, pageEnd=394, url=null, language=null, rfNumber=[17], rfOrder=21, authorNames=BAKOS K L, GAMBA P, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=BAKOS K L, GAMBA P. Hierarchical hybrid decision tree fusion of multiple hyperspectral data processing chains[J]. IEEE Transactions on Geoscience and Remote Sensing, 2010, 49(1): 388-394., articleTitle=Hierarchical hybrid decision tree fusion of multiple hyperspectral data processing chains, refAbstract=null), Reference(id=1241788828273411019, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2016, volume=43, issue=9, pageStart=11, pageEnd=17, 31, url=null, language=null, rfNumber=[18], rfOrder=22, authorNames=王振武, 孙佳骏, 于忠义, journalName=计算机科学, refType=null, unstructuredReference=王振武, 孙佳骏, 于忠义, . 基于支持向量机的遥感图像分类研究综述[J]. 计算机科学, 2016, 43(9): 11-17, 31., articleTitle=基于支持向量机的遥感图像分类研究综述, refAbstract=null), Reference(id=1241788828340519884, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2016, volume=43, issue=9, pageStart=11, pageEnd=17, 31, url=null, language=null, rfNumber=[18], rfOrder=23, authorNames=WANG Zhenwu, SUN Jiajun, YU Zhongyi, journalName=Computer Science, refType=null, unstructuredReference=WANG Zhenwu, SUN Jiajun, YU Zhongyi, et al. Review of remote sensing image classification based on support vector machine[J]. Computer Science, 2016, 43(9): 11-17, 31., articleTitle=Review of remote sensing image classification based on support vector machine, refAbstract=null), Reference(id=1241788828411823053, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2022, volume=81, issue=30, pageStart=44141, pageEnd=44206, url=null, language=null, rfNumber=[19], rfOrder=24, authorNames=SNEHA, KAUL A, journalName=Multimedia Tools and Applications, refType=null, unstructuredReference=SNEHA, KAUL A. Hyperspectral imaging and target detection algorithms: A review[J]. Multimedia Tools and Applications, 2022, 81(30): 44141-44206., articleTitle=Hyperspectral imaging and target detection algorithms: A review, refAbstract=null), Reference(id=1241788828487320526, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2017, volume=14, issue=5, pageStart=597, pageEnd=601, url=null, language=null, rfNumber=[20], rfOrder=25, authorNames=LI W, WU G, DU Q, journalName=IEEE Geoscience and Remote Sensing Letters, refType=null, unstructuredReference=LI W, WU G, DU Q. Transferred deep learning for anomaly detection in hyperspectral imagery[J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14(5):597-601., articleTitle=Transferred deep learning for anomaly detection in hyperspectral imagery, refAbstract=null), Reference(id=1241788828571206607, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=9255, pageEnd=9260, url=null, language=null, rfNumber=[21], rfOrder=26, authorNames=LIU X, WANG C, SUN Q, journalName=null, refType=null, unstructuredReference=LIU X, WANG C, SUN Q, et al. Target detection of hyperspectral image based on convolutional neural networks[C]//2018 37th Chinese Control Conference(CCC). IEEE, 2018: 9255-9260., articleTitle=Target detection of hyperspectral image based on convolutional neural networks, refAbstract=null), Reference(id=1241788828634121168, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2019, volume=16, issue=3, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[22], rfOrder=27, authorNames=FREITAS S, SILVA H, ALMEIDA J M, journalName=International Journal of Advanced Robotic Systems, refType=null, unstructuredReference=FREITAS S, SILVA H, ALMEIDA J M, et al. Convolutional neural network target detection in hyperspectral imaging for maritime surveillance[J]. International Journal of Advanced Robotic Systems, 2019, 16(3):1729881419842991., articleTitle=Convolutional neural network target detection in hyperspectral imaging for maritime surveillance, refAbstract=null), Reference(id=1241788828692841425, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=2284, pageEnd=2287, url=null, language=null, rfNumber=[23], rfOrder=28, authorNames=WANG X, CHEN X, GOU S, journalName=null, refType=null, unstructuredReference=WANG X, CHEN X, GOU S, et al. Hyperspectral target detection via deep multiple instance self-attention neural network[C]//IGARSS 2019-2019 IEEE International Geoscience and Remote Sensing Symposium. IEEE,2019: 2284-2287., articleTitle=Hyperspectral target detection via deep multiple instance self-attention neural network, refAbstract=null), Reference(id=1241788828751561682, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2020, volume=59, issue=8, pageStart=6907, pageEnd=6921, url=null, language=null, rfNumber=[24], rfOrder=29, authorNames=ZHU D, DU B, ZHANG L, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=ZHU D, DU B, ZHANG L. Two-stream convolutional networks for hyperspectral target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 59(8): 6907-6921., articleTitle=Two-stream convolutional networks for hyperspectral target detection, refAbstract=null), Reference(id=1241788828814476244, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=157, pageEnd=167, url=null, language=null, rfNumber=[25], rfOrder=30, authorNames=ANDERSON D Z, ZOLLWEG J D, SMITH B J, journalName=null, refType=null, unstructuredReference=ANDERSON D Z, ZOLLWEG J D, SMITH B J. Paired neural networks for hyperspectral target detection[C]//SPIE Optical Engineering+Applications, August 11-15,2019, San Diego, California, United States. Washington:SPIE, 2019, Proceedings Volume 11139, Applications of Machine Learning: 157-167., articleTitle=Paired neural networks for hyperspectral target detection, refAbstract=null), Reference(id=1241788828885779413, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2022, volume=60, issue=null, pageStart=1, pageEnd=13, url=null, language=null, rfNumber=[26], rfOrder=31, authorNames=WANG Y, CHEN X, WANG F, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=WANG Y, CHEN X, WANG F, et al. Meta-learning based hyperspectral target detection using Siamese network[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 1-13., articleTitle=Meta-learning based hyperspectral target detection using Siamese network, refAbstract=null), Reference(id=1241788828961276886, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=2353, pageEnd=2356, url=null, language=null, rfNumber=[27], rfOrder=32, authorNames=DONG S, QUAN Y, FENG W, journalName=null, refType=null, unstructuredReference=DONG S, QUAN Y, FENG W, et al. Ensemble CNN based on pixelpair and random feature selection for hyperspectral image classification with small-size training set[C]//2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS. IEEE, 2021: 2353-2356., articleTitle=Ensemble CNN based on pixelpair and random feature selection for hyperspectral image classification with small-size training set, refAbstract=null), Reference(id=1241788829036774359, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2018, volume=6, issue=null, pageStart=45562, pageEnd=45577, url=null, language=null, rfNumber=[28], rfOrder=33, authorNames=DU J, LI Z, journalName=IEEE Access,, refType=null, unstructuredReference=DU J, LI Z. A hyperspectral target detection framework with subtraction pixel pair features[J]. IEEE Access, 2018, 6: 45562-45577., articleTitle=A hyperspectral target detection framework with subtraction pixel pair features, refAbstract=null), Reference(id=1241788829095494616, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2022, volume=12, issue=17, pageStart=8392, pageEnd=null, url=null, language=null, rfNumber=[29], rfOrder=34, authorNames=JAIN M, SAIHJPAL V, SINGH N, journalName=Applied Sciences, refType=null, unstructuredReference=JAIN M, SAIHJPAL V, SINGH N, et al. An overview of variants and advancements of PSO algorithm[J]. Applied Sciences, 2022, 12(17): 8392., articleTitle=An overview of variants and advancements of PSO algorithm, refAbstract=null), Reference(id=1241788829158409177, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2023, volume=39, issue=4, pageStart=1232, pageEnd=1249, url=null, language=null, rfNumber=[30], rfOrder=35, authorNames=REN Hsuan, CHANG Chein I, journalName=IEEE Transactions on Aerospace and Electronic Systems, refType=null, unstructuredReference=REN Hsuan, CHANG Chein I. Automatic spectral target recognition in hyperspectral imagery[J]. IEEE Transactions on Aerospace and Electronic Systems, 2023, 39(4):1232-1249., articleTitle=Automatic spectral target recognition in hyperspectral imagery, refAbstract=null), Reference(id=1241788829229712346, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2018, volume=99, issue=null, pageStart=407, pageEnd=411, url=null, language=null, rfNumber=[31], rfOrder=36, authorNames=JAIN N K, NANGIA U, JAIN J, journalName=Journal of The Institution of Engineers (India): Series B, refType=null, unstructuredReference=JAIN N K, NANGIA U, JAIN J. A review of particle swarm optimization[J]. Journal of The Institution of Engineers (India): Series B, 2018, 99: 407-411., articleTitle=A review of particle swarm optimization, refAbstract=null), Reference(id=1241788829288432603, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=10.16660/j.cnki.1674-098x.2015.26.114, pmid=null, pmcid=null, year=2015, volume=12, issue=26, pageStart=216, pageEnd=217, url=null, language=null, rfNumber=[32], rfOrder=37, authorNames=赵乃刚, 邓景顺, journalName=科技创新导报, refType=null, unstructuredReference=赵乃刚, 邓景顺. 粒子群优化算法综述[J]. 科技创新导报, 2015, 12(26): 216-217., articleTitle=粒子群优化算法综述, refAbstract=null), Reference(id=1241788829351347164, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2019, volume=23, issue=null, pageStart=2445, pageEnd=2462, url=null, language=null, rfNumber=[33], rfOrder=38, authorNames=DENG W, YAO R, ZHAO H, journalName=Soft Computing, refType=null, unstructuredReference=DENG W, YAO R, ZHAO H, et al. A novel intelligent diagnosis method using optimal LS-SVM with improved PSO algorithm[J]. Soft Computing, 2019, 23:2445-2462., articleTitle=A novel intelligent diagnosis method using optimal LS-SVM with improved PSO algorithm, refAbstract=null), Reference(id=1241788829410067421, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2006, volume=3, issue=1, pageStart=63, pageEnd=67, url=null, language=null, rfNumber=[34], rfOrder=39, authorNames=CHANG C I, PLAZA A, journalName=IEEE Geoscience and Remote Sensing Letters, refType=null, unstructuredReference=CHANG C I, PLAZA A. A fast iterative algorithm for implementation of pixel purity index[J]. IEEE Geoscience and Remote Sensing Letters, 2006, 3(1): 63-67., articleTitle=A fast iterative algorithm for implementation of pixel purity index, refAbstract=null), Reference(id=1241788829481370590, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2019, volume=46, issue=S1, pageStart=439, pageEnd=441, 451, url=null, language=null, rfNumber=[35], rfOrder=40, authorNames=钟熙, 孙祥娥, journalName=计算机科学, refType=null, unstructuredReference=钟熙, 孙祥娥. 基于Kmeans++聚类的朴素贝叶斯集成方法研究[J]. 计算机科学, 2019, 46(S1): 439-441, 451., articleTitle=基于Kmeans++聚类的朴素贝叶斯集成方法研究, refAbstract=null), Reference(id=1241788829577839583, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2020, volume=58, issue=10, pageStart=7406, pageEnd=7417, url=null, language=null, rfNumber=[36], rfOrder=41, authorNames=LEI J, FANG S, XIE W, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=LEI J, FANG S, XIE W, et al. Discriminative reconstruction for hyperspectral anomaly detection with spectral learning[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(10): 7406-7417., articleTitle=Discriminative reconstruction for hyperspectral anomaly detection with spectral learning, refAbstract=null), Reference(id=1241788829661725664, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2017, volume=9, issue=5, pageStart=482, pageEnd=null, url=null, language=null, rfNumber=[37], rfOrder=42, authorNames=ZHANG Y, WU K, DU B, journalName=Remote Sensing, refType=null, unstructuredReference=ZHANG Y, WU K, DU B, et al. Hyperspectral target detection via adaptive joint sparse representation and multitask learning with locality information[J]. Remote Sensing, 2017, 9(5): 482., articleTitle=Hyperspectral target detection via adaptive joint sparse representation and multitask learning with locality information, refAbstract=null), Reference(id=1241788829724640225, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2020, volume=18, issue=2, pageStart=205, pageEnd=210, 217, url=null, language=null, rfNumber=[38], rfOrder=43, authorNames=王朗宁, 侯炎磐, 李彦峰, journalName=雷达科学与技术, refType=null, unstructuredReference=王朗宁, 侯炎磐, 李彦峰. 数据参数影响RCS统计特征数据可分性[J]. 雷达科学与技术, 2020, 18(2): 205-210, 217., articleTitle=数据参数影响RCS统计特征数据可分性, refAbstract=null), Reference(id=1241788829787554786, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2020, volume=18, issue=2, pageStart=205, pageEnd=210, 217, url=null, language=null, rfNumber=[38], rfOrder=44, authorNames=WANG Langning, HOU Yanpan, LI Yanfeng, journalName=Radar Science and Technology, refType=null, unstructuredReference=WANG Langning, HOU Yanpan, LI Yanfeng. Influences of sampling frequency and parameters of sliding window on radar target recognition using radar cross section features[J]. Radar Science and Technology, 2020, 18(2):205-210, 217., articleTitle=Influences of sampling frequency and parameters of sliding window on radar target recognition using radar cross section features, refAbstract=null), Reference(id=1241788831263949795, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=1982, volume=143, issue=1, pageStart=29, pageEnd=36, url=null, language=null, rfNumber=[39], rfOrder=45, authorNames=HANLEY J A, MCNEIL B J, journalName=Radiology, refType=null, unstructuredReference=HANLEY J A, MCNEIL B J. The meaning and use of the area under a receiver operating characteristic (ROC)curve[J]. Radiology, 1982, 143(1): 29-36., articleTitle=The meaning and use of the area under a receiver operating characteristic (ROC)curve, refAbstract=null), Reference(id=1241788831326864357, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=1998, volume=65, issue=3, pageStart=227, pageEnd=248, url=null, language=null, rfNumber=[40], rfOrder=46, authorNames=GREEN R O, EASTWOOD M L, SARTURE C M, journalName=Remote sensing of environment, refType=null, unstructuredReference=GREEN R O, EASTWOOD M L, SARTURE C M,et al. Imaging spectroscopy and the airborne visible/infrared imaging spectrometer (AVIRIS)[J]. Remote sensing of environment, 1998, 65(3): 227-248., articleTitle=Imaging spectroscopy and the airborne visible/infrared imaging spectrometer (AVIRIS), refAbstract=null), Reference(id=1241788831410750438, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2020, volume=13, issue=null, pageStart=2663, pageEnd=2672, url=null, language=null, rfNumber=[41], rfOrder=47, authorNames=ZHANG Y, FAN Y, XU M, journalName=IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, refType=null, unstructuredReference=ZHANG Y, FAN Y, XU M, et al. An improved low rank and sparse matrix decomposition-based anomaly target detection algorithm for hyperspectral imagery[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020, 13: 2663-2672., articleTitle=An improved low rank and sparse matrix decomposition-based anomaly target detection algorithm for hyperspectral imagery, refAbstract=null), Reference(id=1241788831494636519, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, doi=null, pmid=null, pmcid=null, year=2017, volume=55, issue=10, pageStart=5600, pageEnd=5611, url=null, language=null, rfNumber=[42], rfOrder=48, authorNames=KANG X, ZHANG X, LI S, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=KANG X, ZHANG X, LI S, et al. Hyperspectral anomaly detection with attribute and edge-preserving filters[J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(10): 5600-5611., articleTitle=Hyperspectral anomaly detection with attribute and edge-preserving filters, refAbstract=null)], funds=[Fund(id=1241788824964105132, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, awardId=41776182, language=CN, fundingSource=国家自然科学基金(41776182), fundOrder=null, country=null), Fund(id=1241788825027019694, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, awardId=2023KJ068, language=CN, fundingSource=山东省高校青年创新技术支持计划(2023KJ068), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1241788818232247066, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, xref=1., ext=[AuthorCompanyExt(id=1241788818236441371, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, companyId=1241788818232247066, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.College of Oceanography and Spatial Information, China University of Petroleum (East China), Qingdao 266580, China), AuthorCompanyExt(id=1241788818244829980, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, companyId=1241788818232247066, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1中国石油大学(华东)海洋与空间信息学院 青岛 266580)]), AuthorCompany(id=1241788818337104671, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, xref=2., ext=[AuthorCompanyExt(id=1241788818345493279, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, companyId=1241788818337104671, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.Shandong Lubang Geographic Information Engineering Co., LTD. Jinan 250102, China), AuthorCompanyExt(id=1241788818353881888, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, companyId=1241788818337104671, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2山东鲁邦地理信息工程有限公司 济南 250102)])], figs=[ArticleFig(id=1241788820455228266, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Fig. 1, caption=The flow chart of hyperspectral target detection proposed in this paper, figureFileSmall=jxTSYxVRtJzEwwozI1Licw==, figureFileBig=z9dU0MZr0crqZ6vkRBwoyA==, tableContent=null), ArticleFig(id=1241788820530725740, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=图1, caption=本文提出的高光谱目标检测流程图, figureFileSmall=jxTSYxVRtJzEwwozI1Licw==, figureFileBig=z9dU0MZr0crqZ6vkRBwoyA==, tableContent=null), ArticleFig(id=1241788820627194734, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Fig. 2, caption=Spectral curves of pixels A, B, and C, figureFileSmall=AP4o/mG9W/4AhN69lISdTQ==, figureFileBig=u91zsc46cgIyzI57eojuUw==, tableContent=null), ArticleFig(id=1241788820715275122, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=图2, caption=像素ABC的光谱曲线, figureFileSmall=AP4o/mG9W/4AhN69lISdTQ==, figureFileBig=u91zsc46cgIyzI57eojuUw==, tableContent=null), ArticleFig(id=1241788820799161204, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Fig. 3, caption=Network structure, figureFileSmall=YE1pg3/vSzvMXaOSoX9WGQ==, figureFileBig=M3vcHtcUwdN6ZCsNRRfQ+A==, tableContent=null), ArticleFig(id=1241788820870464375, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=图3, caption=网络架构, figureFileSmall=YE1pg3/vSzvMXaOSoX9WGQ==, figureFileBig=M3vcHtcUwdN6ZCsNRRfQ+A==, tableContent=null), ArticleFig(id=1241788822317499258, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Fig. 4, caption=Test process, figureFileSmall=CxOkvJNTQi+paf6zERWPPw==, figureFileBig=MPuA43ykrC1pL5b0Q+e4vA==, tableContent=null), ArticleFig(id=1241788822422356858, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=图4, caption=测试流程, figureFileSmall=CxOkvJNTQi+paf6zERWPPw==, figureFileBig=MPuA43ykrC1pL5b0Q+e4vA==, tableContent=null), ArticleFig(id=1241788822497854331, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Fig. 5, caption=Particle swarm optimization network hyperparameters, figureFileSmall=5R00qy9zZqbd8afecwQuoA==, figureFileBig=4n9tSU0B6FrXe+9CGYdA+g==, tableContent=null), ArticleFig(id=1241788822564963197, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=图5, caption=粒子群优化网络超参数, figureFileSmall=5R00qy9zZqbd8afecwQuoA==, figureFileBig=4n9tSU0B6FrXe+9CGYdA+g==, tableContent=null), ArticleFig(id=1241788822648849278, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Fig. 6, caption=The collected images by AVIRIS, figureFileSmall=zMb0AyNDfNn1BBpx8DEOmQ==, figureFileBig=VvKz+Q3lY5yL1of+GDCTIQ==, tableContent=null), ArticleFig(id=1241788822732735361, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=图6, caption=机载可见/红外成像光谱仪采集的图像, figureFileSmall=zMb0AyNDfNn1BBpx8DEOmQ==, figureFileBig=VvKz+Q3lY5yL1of+GDCTIQ==, tableContent=null), ArticleFig(id=1241788822833398658, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Fig. 7, caption=The collected images by HYDICE, figureFileSmall=Kx6F2R3wpDGXL6UxUVD1QQ==, figureFileBig=meflbQ2pxDCwZ6VLCPX4XQ==, tableContent=null), ArticleFig(id=1241788822904701827, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=图7, caption=高光谱数位影像实验仪采集的图像, figureFileSmall=Kx6F2R3wpDGXL6UxUVD1QQ==, figureFileBig=meflbQ2pxDCwZ6VLCPX4XQ==, tableContent=null), ArticleFig(id=1241788822976004997, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Fig. 7, caption=The collected images by ROSIS, figureFileSmall=S4PWNaGg0+f4bAh82ZpIRg==, figureFileBig=uZfMfb9lJsUfg66svjnV3g==, tableContent=null), ArticleFig(id=1241788823043113862, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=图7, caption=反射光学系统成像光谱仪采集的图像, figureFileSmall=S4PWNaGg0+f4bAh82ZpIRg==, figureFileBig=uZfMfb9lJsUfg66svjnV3g==, tableContent=null), ArticleFig(id=1241788823147971464, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Fig. 8, caption=The collected images by AVIRIS, figureFileSmall=l7Tj+rSQjha01lFa7IIBFg==, figureFileBig=O5rD3uy//r941UiQcyKKig==, tableContent=null), ArticleFig(id=1241788823265411978, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=图8, caption=机载可见/红外成像光谱仪采集的图像, figureFileSmall=l7Tj+rSQjha01lFa7IIBFg==, figureFileBig=O5rD3uy//r941UiQcyKKig==, tableContent=null), ArticleFig(id=1241788823370269580, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Fig. 9, caption=The detection results of five detectors in Images 1~Image 5, figureFileSmall=OLSSgAE+jICmmyv5Ipn9Cw==, figureFileBig=IFLM9+KgZRcOnJc1pllLmA==, tableContent=null), ArticleFig(id=1241788823508681614, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=图9, caption=五种检测器在图像1~图像5中的检测结果图, figureFileSmall=OLSSgAE+jICmmyv5Ipn9Cw==, figureFileBig=IFLM9+KgZRcOnJc1pllLmA==, tableContent=null), ArticleFig(id=1241788823605150607, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Fig. 10, caption=Detection performance of different detectors on image 1, figureFileSmall=z14kmbGfC44jTOiNjGeIag==, figureFileBig=q5wXRbI0vPfbg2sePJYj4w==, tableContent=null), ArticleFig(id=1241788823726785425, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=图10, caption=不同检测器在图像1上的检测性能, figureFileSmall=z14kmbGfC44jTOiNjGeIag==, figureFileBig=q5wXRbI0vPfbg2sePJYj4w==, tableContent=null), ArticleFig(id=1241788823798088594, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Fig. 11, caption=Detection performance of different detectors on image 2, figureFileSmall=SJnLxAKV27BCt24OaWsMqQ==, figureFileBig=KioKY65g1cbYou36XaTRaA==, tableContent=null), ArticleFig(id=1241788823865197460, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=图11, caption=不同检测器在图像2上的检测性能, figureFileSmall=SJnLxAKV27BCt24OaWsMqQ==, figureFileBig=KioKY65g1cbYou36XaTRaA==, tableContent=null), ArticleFig(id=1241788823957472151, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Fig. 12, caption=Detection performance of different detectors on image 3, figureFileSmall=hKZHr6JIZoGP7uo2wKjcNA==, figureFileBig=40DTcqa7RL65G3VByiPtaA==, tableContent=null), ArticleFig(id=1241788824024581017, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=图12, caption=不同检测器在图像3上的检测性能, figureFileSmall=hKZHr6JIZoGP7uo2wKjcNA==, figureFileBig=40DTcqa7RL65G3VByiPtaA==, tableContent=null), ArticleFig(id=1241788824129438617, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Fig. 13, caption=Detection performance of different detectors on image 4, figureFileSmall=MApL/WYmIvPpHVc30XpW/Q==, figureFileBig=zkHltd3FI8wUhJeXbr3wBg==, tableContent=null), ArticleFig(id=1241788824221713306, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=图13, caption=不同检测器在图像4上的检测性能, figureFileSmall=MApL/WYmIvPpHVc30XpW/Q==, figureFileBig=zkHltd3FI8wUhJeXbr3wBg==, tableContent=null), ArticleFig(id=1241788824305599388, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Fig. 14, caption=Detection performance of different detectors on image 5, figureFileSmall=MKjkz99WjzKubOzwobbBzw==, figureFileBig=DWjtcm9OncOYcbQVUgYTiw==, tableContent=null), ArticleFig(id=1241788824364319646, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=图14, caption=不同检测器在图像5上的检测性能, figureFileSmall=MKjkz99WjzKubOzwobbBzw==, figureFileBig=DWjtcm9OncOYcbQVUgYTiw==, tableContent=null), ArticleFig(id=1241788824427234208, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Fig. 15, caption=The detection results of different detectors in Images 1~Image 5, figureFileSmall=o/nMqC4gWMdwLL8BeJLIXg==, figureFileBig=wYQ5+j7pqGqoGzNHB3O59g==, tableContent=null), ArticleFig(id=1241788824502731682, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=图15, caption=不同检测器在图像1~图像5中的检测结果图, figureFileSmall=o/nMqC4gWMdwLL8BeJLIXg==, figureFileBig=wYQ5+j7pqGqoGzNHB3O59g==, tableContent=null), ArticleFig(id=1241788824561451940, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Table 1, caption=

AUC values for five detectors in different data sets

, figureFileSmall=null, figureFileBig=null, tableContent=
数据ACEOSPPPF-CNNSUB-PPF-CNNRFS-PPF-CNNProposed
Method
图像198.3597.5998.2498.4496.9599.75
图像297.8682.3497.6099.6097.5299.35
图像387.6690.1394.2196.9997.4299.91
图像493.5584.9798.9697.8898.6799.97
图像577.0087.3395.3697.7194.1699.68
), ArticleFig(id=1241788824678892455, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=表1, caption=

五种检测器在不同数据集中的AUC值

, figureFileSmall=null, figureFileBig=null, tableContent=
数据ACEOSPPPF-CNNSUB-PPF-CNNRFS-PPF-CNNProposed
Method
图像198.3597.5998.2498.4496.9599.75
图像297.8682.3497.6099.6097.5299.35
图像387.6690.1394.2196.9997.4299.91
图像493.5584.9798.9697.8898.6799.97
图像577.0087.3395.3697.7194.1699.68
), ArticleFig(id=1241788824746001320, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=EN, label=Table 2, caption=

AUC values and time for different detectors in different data sets

, figureFileSmall=null, figureFileBig=null, tableContent=
DatasetMethod1Method2Method3Proposed
Method
AUCTimeAUCTimeAUCTimeAUCTime
Image197.201 17599.4519698.177199.7573
Image298.391 02899.0318997.998199.3586
Image399.3718897.808198.863699.9134
Image499.9510499.914499.931999.9720
Image597.3817198.756897.402799.6831
), ArticleFig(id=1241788824825693097, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241757968581055430, language=CN, label=表2, caption=

不同检测器在不同数据集中的AUC值和时间

, figureFileSmall=null, figureFileBig=null, tableContent=
DatasetMethod1Method2Method3Proposed
Method
AUCTimeAUCTimeAUCTimeAUCTime
Image197.201 17599.4519698.177199.7573
Image298.391 02899.0318997.998199.3586
Image399.3718897.808198.863699.9134
Image499.9510499.914499.931999.9720
Image597.3817198.756897.402799.6831
)], 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.20240129001, detailUrlEn=https://castjournals.cast.org.cn/joweb/ycyk/EN/10.12347/j.ycyk.20240129001, pdfUrlCn=https://castjournals.cast.org.cn/joweb/ycyk/CN/PDF/10.12347/j.ycyk.20240129001, pdfUrlEn=https://castjournals.cast.org.cn/joweb/ycyk/EN/PDF/10.12347/j.ycyk.20240129001, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于样本增强和自动参数优化的高光谱目标检测方法
收藏切换
PDF下载
刘浩 1 , 许明明 1 , 沈彪群 2 , 刘善伟 1 , 盛辉 1
遥测遥控 | 人工智能技术 2024,45(4): 31-44
收起
收藏切换
遥测遥控 | 人工智能技术 2024, 45(4): 31-44
基于样本增强和自动参数优化的高光谱目标检测方法
全屏
刘浩1, 许明明1, 沈彪群2 , 刘善伟1, 盛辉1
作者信息
  • 1中国石油大学(华东)海洋与空间信息学院 青岛 266580
  • 2山东鲁邦地理信息工程有限公司 济南 250102
  • 刘浩 1998年生,硕士研究生。

    许明明 1990年生,副教授,硕士生导师。

    沈彪群 1982年生,本科,高级工程师。

    刘善伟 1982年生,教授,博士生导师。

    盛辉 1972年生,副教授,硕士生导师。

通讯作者:

Hyperspectral Target Detection Based on Sample Enhancement and Automatic Parameter Optimization
Hao LIU1, Mingming XU1, Biaoqun SHEN2 , Shanwei LIU1, Hui SHENG1
Affiliations
  • 1.College of Oceanography and Spatial Information, China University of Petroleum (East China), Qingdao 266580, China
  • 2.Shandong Lubang Geographic Information Engineering Co., LTD. Jinan 250102, China
出版时间: 2024-07-15 doi: 10.12347/j.ycyk.20240129001
文章导航
收藏切换

基于深度学习的高光谱目标检测面临着样本质量不足、网络结构复杂、参数调整费力等问题,本文提出了一种具有数据增强和自动超参数优化的深度学习方法。为了解决样本质量不足的问题,本文引入了一种样本扩增策略。该策略利用端元提取和聚类技术直接从高光谱图像中获取大量背景像素,通过使用相减像素配对方法将这些像素与少量已知目标像素配对,获得大量标记的纯样本对,从而实现数据增强。此外,与大多数复杂的深度网络不同,本文设计了一个由12个卷积层组成的轻量级卷积神经网络(CNN)。该网络专门设计用于高效快速地学习输入样本对与其对应标签之间的映射。结合粒子群优化算法,该网络具有超参数自动优化的能力,克服了参数调整费力的缺点,这使得网络能够根据来自不同高光谱图像的样本自动调整超参数,从而产生最优结果。对于测试像素,训练网络的输入是中心像素与其相邻像素之间的光谱差。当一个测试像素属于目标时,输出分数接近1,反之则接近0。在五个高光谱数据集上的实验结果表明:本文提出的方法明显优于现有的技术。

高光谱  /  目标检测  /  样本增强  /  卷积神经网络

Hyperspectral target detection based on deep learning faces challenges such as insufficient quality of samples, intricate network structures, and laborious parameter adjustment. In this paper, we propose a deep learning method with data augmentation and automatic hyperparameter optimization. To tackle the issue of insufficient quality of samples, we introduce a sample augmentation strategy. The strategy utilizes endmember extraction and clustering techniques to directly acquire a large number of background pixels from hyperspectral images. By pairing these with a small number of known target pixels using a phase-reducing pixel pairing approach, we obtain a large number of labeled pure sample pairs, thereby accomplishing data augmentation. In addition, distinct from most complex deep networks, we designed a lightweight Convolutional Neural Network (CNN) comprised of 12 convolutional layers. This network is specifically engineered to efficiently and rapidly learn the mapping between input sample pairs and their corresponding labels. By incorporating the particle swarm optimization algorithm, this network possesses the capability to automatically optimize hyperparameters, overcoming the shortcomings of laborious parameter adjustment. This enables the network to automatically adjust hyperparameters based on samples from different hyperspectral images, thereby generating optimal results. For a test pixel, the input to the trained network is the spectral difference between the central pixel and its adjacent pixels. When a test pixel belongs to the target, the output score is closely align with the target label. Experimental results on five hyperspectral datasets demonstrate that our method significantly outperforms existing techniques.

Hyperspectral  /  Target detection  /  Data augmentation  /  Convolutional neural network
刘浩, 许明明, 沈彪群, 刘善伟, 盛辉. 基于样本增强和自动参数优化的高光谱目标检测方法. 遥测遥控, 2024 , 45 (4) : 31 -44 . DOI: 10.12347/j.ycyk.20240129001
Hao LIU, Mingming XU, Biaoqun SHEN, Shanwei LIU, Hui SHENG. Hyperspectral Target Detection Based on Sample Enhancement and Automatic Parameter Optimization[J]. Journal of Telemetry, Tracking and Command, 2024 , 45 (4) : 31 -44 . DOI: 10.12347/j.ycyk.20240129001
高光谱图像具有宽波段范围和高光谱分辨率,高光谱遥感数据立方体包含丰富的图像信息和光谱信息[1]。作为一种监督方法,高光谱目标检测需要一个或多个目标的光谱作为先验光谱来检测目标[2,3]。高光谱目标探测[4]在矿产勘探[5]、军事侦察、环境监测[6]、海洋科学[7]等领域显示出巨大的应用潜力。
传统的高光谱目标检测方法包括优化光谱轮廓法(OSP)[8]、约束能量最小化法(CEM)[9]、自适应相干估计法(ACE)[10]、高光谱子空间检测法(HSD)[11]等。基于机器学习的方法也广泛应用于高光谱目标检测算法中,如核技术、稀疏与协同表示方法[1215]、马尔可夫随机场[16]、决策树[17]、支持向量机[18]等。
近年来,随着计算能力的不断提高和快速便捷的数据采集方法的出现,深度学习方法[19]逐渐被用于高光谱目标检测领域,基于CNN的方法是目前研究的主流。Li等人是第一个在这一领域利用迁移学习的人[20],他引入了一种创新的像素对方法,有效地增加了样本量。Liu等人采用卷积神经网络进行目标检测,通过参数调整和网络优化,达到了比深度信念网络和支持向量机更高的测试精度[21]。Sara Freitas等人提高了用于海上监视的高分辨率高光谱成像系统中目标识别的准确性,通过使用无人机和3D-CNN深度学习技术实现了这一目标,该技术可以协同光谱和空间信息[22]。Wang等人介绍了一种基于自注意机制的深度多实例学习方法,该方法利用最大池化和一维卷积神经网络实现端到端高光谱目标检测,无需明确学习目标特征[23]。Zhu等人提出了一种基于双流卷积网络的目标检测器,该检测器利用双流卷积网络从高光谱图像中提取丰富的光谱信息[24]。同样,基于其他网络的方法在高光谱目标检测领域也取得了重大进展。Anderson等人提出使用配对神经网络进行高光谱目标检测[25],Wang提出了一种基于元学习和Siamese网络的高光谱目标检测方法[26]
卷积神经网络通过卷积层和池化层的结合,有效地捕捉输入数据的局部空间特征。通过叠加多个卷积层和全连接层,CNN能够逐渐学习到更高级的抽象特征表示,有助于更准确地区分目标和背景。然而,由于高光谱图像的复杂性,用于高光谱目标检测的CNN往往具有复杂的结构。此外,调整CNN中的复杂超参数仍然是一项具有挑战性的任务。研究人员需要不断尝试不同的参数组合,训练和评估以获得最佳性能。这需要深入了解网络架构和超参数对性能的影响,并进行反复实验和优化,以获得最佳结果。
除了复杂的CNN网络和参数调整带来的挑战外,样本可用性问题也是基于深度学习的高光谱目标检测方法关注的关键领域。虽然深度学习方法能够提取更深层次的特征,但它们需要大量的样本支持来进行网络训练。Li等人提出利用像素对特征(PPF)来解决训练标签有限的问题。然而,这种方法只考虑了像素的光谱特征,没有考虑空间特征。Dong等人在此基础上,将随机特征选择(RFS)与像素对特征相结合,在一些图像上取得了较好的性能,但是方法的稳定性有待提高[27]。SPPF考虑到像素的空间特征,但是生成的训练样本数量相对较小,仅是配对前的8倍,在已知样本数量有限的一些场景,SPPF生成的新样本数量可能不足以满足深度学习训练过程的要求。此外,PPF和SPPF在生成样本时都简单地将目标和背景分为两类,导致对不同类别背景像素的光谱特性考虑不足。Du等人在PPF和SPPF的基础上提出了减像元对特征(SUB-PPF)[28]。该方法在检测阶段增加了检测像素周围空间信息的影响,从而考虑了像素的空间特征。此外,它还考虑到背景类,根据地理信息将其划分为多个类别。这种方法可以获得更加多样化和丰富的背景特征信息。通过将目标与多个背景类配对,生成的样本扩展了数百倍。SUB-PPF也有它的缺点,人工选择像素进行配对的过程引入了显著的随机性,导致类内方差大,样本冗余。由于高光谱图像的光谱波段较多,冗余样本的存在大大增加了训练成本和时间。
本文提出了一种基于样本增强和超参数自动优化的高光谱目标检测方法。为了解决复杂的网络结构和人工调整网络超参数的问题,首先,设计了一个具有12个卷积层的轻量级CNN架构。在保持特征提取质量的同时,显著增加了网络的非线性特征,增强了网络的表达能力,提高了网络在高光谱目标检测任务中的性能。其次,引入粒子群算法(PSO),实现了参数调整的自动化。针对样本问题,提出了一种高光谱图像中纯像元的识别和聚类方法,该方法首先提取高光谱图像中的纯像元并聚类,接着分别计算每个聚类的平均光谱曲线和已知目标像素的平均光谱曲线,通过比较它们的光谱角距离,可以确定合适的背景类别,该处理旨在获得更纯净更有效的样本集,从而减少网络训练时间。
本文提出的高光谱目标检测流程如图1所示,主要包括三个步骤(标记为红色),即基于纯像元的样本生成模型,用于训练和测试的轻量化CNN网轮以及粒子群自动优化网络超参数。
具体流程实现如下:
① 输入已知目标样本和待检测高光谱图像;
② 利用FIBBI(菲比)算法寻找纯净像元;
③ 使用光谱角距离(SAD)替代K-means(K均值)算法中的欧氏距离,对纯净像元聚类;
④ 对聚类进行筛选和剔除,获得目标类和背景类像素;
⑤ 构建相同背景类样本之间的减法像素对、不同背景类样本之间的减法像素对和目标类与背景类之间减法像素对,分别赋值为0、0、1,样本生成;
⑥ 初始化网络超参数,输入步骤⑤生成的样本,设置循环终止的阈值;
⑦ 通过粒子群优化算法优化网络的超参数,训练网络;
⑧ 对于要检测的每个像素,与周围像素生成减法像素对,放入训练良好的CNN中打分,产生检测结果图;
⑨ 计算结果的AUC(曲线下的面积)值,若不满足条件“ROC(受试者工作特征)值与期望值之间的均方误差<阈值”,则回到步骤⑦;若满足条件,则输出图像。
本文提出了一种基于纯像元的样本生成模型,首先计算高光谱图像中的纯像元指数并对得分进行排序,获得纯像元。对这些纯像元聚类并进行一定操作后,得到一些纯净背景像素类,结合已有的目标像素类,通过相减像素配对模型,即可生成基于纯像元的样本,具体分为以下三个部分。
通过计算高光谱图像中像素的纯像元指数(Pixel Purity Index,PPI)可以获得纯净像元,传统的PPI算法[29]计算复杂度较高,且对于初始值的选择十分敏感。本文选择FIPPI(快速迭代PPI)算法提取图像中的纯净像元,该算法相比传统的PPI算法收敛速度更快,且无需手动干预选择初始参数。该算法流程如下:
① 初始化并降维:通过最大噪声分数(MNF)变换对高光谱数据进行降维,并保留第一个分量,设为初始向量集,其中Nj为第j个随机生成的初始向量,通过文献[30]中的ATGP(自动目标生成方法)算法生成目标像素对应的向量,该算法能大幅加快向量的生成速度。
② 迭代规则:在迭代过程中,对每一个向量,将所有的样本向量都投影到这个特定的向量上,并找到那些处于极值位置的点来组成一个极值集合,记为。找出由步骤①定义的产生最大的样本向量,记为
③ 中止条件:如果,说明没有找到新的端元,那么算法迭代终止,将满足条件的纯像元记为
在一张大小为100×100像素的高光谱图像上,考虑到虚警率和后续实验的需要,该方法一般提取出100个纯像元。
提取纯像元之后,需要对它们进行聚类,并对于生成的类别剔除目标类,获得纯净背景像素类。
对于提取的纯像元,使用K-means算法[31]对它们进行聚类,根据高光谱图像的特征,对其进行一定的改进。在高光谱图像中,同类地物可能由于光照、分布等因素的影响,反射的能量大小不同,因此,使用传统的欧式距离表示两条光谱曲线的光谱差异是不够准确的。
图2所示,图中为ABC三个像素点的光谱曲线,三个像素均为AVRIS(机载可见光/红外成像光谱仪)传感器拍摄的加利福尼亚机场数据中的像素点,像素AB为同一类地物,像素C为另一类地物。
对于两个点P=(p1p2pn)和Q=(q1,q2qn),通过如下公式计算两点的欧氏距离:
S=sqrt((p1-q1)2+(p2-q2)2+…+(pn-qn)2)
计算结果为:像素A与像素B的欧式距离为1.79×104,像素A与像素C的欧式距离为2.83×103。像素A与像素B属于同一类地物,但是它们的欧氏距离小于像素A与像素C的欧氏距离,因此采用欧式距离判别像素光谱差异会受到干扰。
光谱角距离(Spectral Angle Distance)是用于测量两个光谱向量之间相似性的一种常用方法,常用于遥感图像处理和光谱分析[32]。光谱角距离基于光谱向量的夹角来衡量相似性,计算公式如下:
式中,AB是两个相同长度的谱向量。AiBi分别为光谱向量AB中第i个元素,使用光谱角距离替代欧氏距离能让K-means算法在聚类过程中更准确地考虑光谱数据的特征。
使用改进的K-means算法获取m个纯像元聚类之后,这些类中一般包含一个可能目标类。为了将这些类中的可能目标类剔除,进行以下操作:选择像素数量大于20的类别作为n个待定背景类,并分别计算每一类像素的平均光谱曲线。在目标检测中,一般已知一些目标像素的光谱曲线,从中随机选取20个目标像素作为已知目标类,并计算该类平均光谱曲线,通过式(2)分别计算待定背景类像素与已知目标类像素的光谱角距离,将光谱角距离最大值所对应的待定背景类判别为可能目标类并剔除,剩下的n-1个待定背景类作为背景类。
得到的1个目标类与n-1个背景类之后,为扩大样本量,采用文献[28]中提出的减法像素配对原则。一般选择10个背景类和1个目标类,每个类的像素数为15。基于配对原理,若两个像素分别来自不同的背景类,配对后产生的新背景类像素
记为Xb1,产生新样本的数量N1为:
若两个像素来自相同背景类,配对后产生的新背景类像素记为Xb2,产生新样本的数量N2为:
若两个像素来自相同背景类,配对后产生的新背景类像素记为Xt,产生新样本的数量N3为:
N3=m×m×(n-1)=15×15×10=2 250
新的背景类像素的数量为N1+N2,总共为11 175,将这些像素赋值为0;总共生成2 250个新的目标类像素,这些像素被赋值为1。基于此,生成了足够数量的低冗余样本。
由于新的训练样本全部由纯像元组成,且数量较少,特征提取更容易,因此本文设计了一个更加轻量级的CNN网络来提取深度特征。该网络各层参数及架构如图3所示,在保证模型性能的同时,节省了计算成本,提高了检测速度。
该网络由11个卷积层和一个平均池化层组成。在每个卷积层之后,使用一个线性元素层来加速随机梯度下降算法的收敛,可以保证训练的网络具有适度的稀疏性。采用两步卷积层(C4, C6, C8,C9和C11)来限制光谱维数。如图3所示,以光谱数量为224的高光谱图像为例,网络输入的特征向量为一个1×224×1张量和对应的0或1标签。第一和第二卷积层(C1和C2)主要用20个1×3×1卷积核对输入的1×224×1张量进行过滤,得到一个1×224×20张量。然后,第三个卷积层(C3)用30个1×3×1卷积核过滤输入1×224×20张量以产生1×224×30张量。从C1到C3(图3中的C1-C3),提取非线性特征。由于对样本进行了优化,使得网络更容易提取样本的高级特征,在保证网络性能的同时,大大减少了卷积层数和特征向量的数量。从C1到C3,由于步长为1,每一层的输出与原始输入向量的维数相同。当每一层在第二步进入C4时,输出30个1×112×1特征张量,将光谱维数减少一半。
按照上述相同的卷积操作,第11层卷积层(C11)的输出为20×7×1特征张量。在应用了11个卷积层之后,另一个卷积层(C12)过滤了一个1×20×1张量。这个张量是通过使用1×3×1内核对C11的输出应用一个平均池层来生成的,这获得了范围在0到1之间的1×1×1的输出分数。在C12中使用1×1×1卷积核,不仅集成了不同通道的信息,实现了降维,而且在保持特征张量尺度不变的前提下,极大地增加了非线性特征,提高了网络的表达能力。其中,epochs值设置为30,卷积核大小均为1×3×1。
在测试阶段,将测试像素分别和周围像素构建相减像素对。如图4所示,在以待测像素为中心的3×3范围内计算待测像素与周围8个像素的差向量。将它们输入经过训练的CNN网络,输出8个分数。这些分数的平均值是被测像素的最终分数,介于0到1之间。阈值一般设置为0.5,当分数大于设置的阈值时,判断为目标像素,否则为背景像素。在预测出所有像素点的得分后,通过不断改变阈值大小来计算误检率和漏检率,得到ROC曲线和AUC数值。
学习率和批处理大小是网络训练过程中两个重要的超参数[33]。学习率决定了参数更新的步长,过大或过小都会导致训练误差大,模型性能差。批处理的大小也会影响网络训练的速度、稳定性和计算资源的利用率。由于更多样本的梯度平均值可以减少单个样本对梯度更新的影响,较大的批处理大小可以减少训练过程中的噪声,但也可能导致欠拟合或过度平滑问题。此外,对于不同的数据集,需要不断调整学习率和批处理大小参数的组合,以获得更好的训练效果。粒子群优化算法[34]是一种基于群体的随机优化方法,每个潜在的解都以粒子的形式表示,粒子通过动态改变自身的速度和方向来寻找最优解,在复杂的参数调整中有着较好的效果[3537]。因此,引入粒子群优化算法对学习率和批处理大小进行优化,以找到最佳的超参数组合。过程如图5所示,具体步骤如下:
① 获取样本数据集,确定CNN网络架构,设置学习率和批处理大小的数值范围,确定最优终止条件初始化模型。
② 运行网络一次,计算并输出测试图像的AUC值。由于ROC的期望值为1,因此计算输出ROC值与1之间的均方根误差,其表达式如式(6)所示:
如果满足终止条件,执行步骤⑤;否则将该值作为粒子的适应度值,通过计算粒子的速度和方向调整方程,得到粒子的局部最优值和全局最优值。
③ 以新粒子的速度和方向形成新的粒子群,即更新网络超参数。
④ 重新计算模型输出的测试图像的ROC值与期望值之间的均方误差,如果该值小于阈值,执行步骤⑤;否则,执行步骤③。
⑤ 如果均方误差在允许范围内,则输出此时的超参数值、目标检测结果和AUC值,模型训练完成。
在本节中,验证了本文提出方法的检测性能,并选择了几种具有代表性的高光谱目标检测方法进行性能比较,包括ACE、OSP等传统方法和PPF-CNN、SUB-PPF-CNN、RFS-PPF-CNN等基于像素对特征的深度学习方法。采用统计可分性分析[38]、受试者工作特征(ROC)[39]和曲线下面积(AUC)指标作为主要评价标准进行定性和定量比较。
本文使用5个高光谱数据集来评估本文提出的算法的性能。
图像1和图像2由机载可见/红外成像光谱仪(AVIRIS)[40]传感器采集,大小为100像素×100像素,去除吸水带和低信噪比后,共保留189个波段,波长范围为0.4 μm ~1.8 μm,选择飞机作为检测目标。在图像1中,有3架飞机,由88个像素组成。图像2中有三架飞机,由58个像素组成。图像1和图像2的假彩色图和地面真值图如图6所示。
图像3来自于HYDICE(高光谱数位影像实验仪)传感器拍摄的城市地区,图像的光谱分辨率为10 nm,空间分辨率为1 m,空间大小为80像素×100像素。它由210个波段组成,去除噪声和吸水波段后,实验中使用了162个波段。伪彩色图像场景如图7(a)所示。地面真值图如图7(b)所示。汽车和房顶为目标,共21像素。
图像4是由反射光学系统成像光谱仪(ROSIS)在意大利北部帕维亚上空的飞行活动中收集的[41],该场景是帕维亚中心和大学数据集的一部分,称为Pavia C(帕维亚中心)数据集。该数据集由105像素×105像素×102波段数组成,图像中有7辆汽车,由67个像素组成。场景和地面真实情况如图7所示。
图像5由机载可见/红外成像光谱仪(AVIRIS)传感器采集,属于机场海滩城(ABU)数据集[42]。该场景覆盖了美国得克萨斯州加尔维斯顿岛的一个停车场,由100像素×100像素×207波段数组成。图像中有一些车辆,由155像素组成。图像5的图像场景和地面真值如图8所示。
本节将从两个角度讨论结果:基于实验数据的可视化分析和基于评价指标的分析。本节的实验结果数据包括图9~图14表1所示的结果。
将本文方法与其他五种方法在5个HIS(高光谱图像)数据集上的检测图与图9所示的地面真值图进行对比(图9中从上到下依次为图像1~图像5),从提取目标和抑制背景的能力两方面对每种方法的目标检测性能进行定性评价,图9从左到右分别为ACE检测图、OSP检测图、PPF-CNN检测图、SUB-PPF-CNN检测图、RFS-PPF-CNN检测图、本文提出方法的检测图以及地面真值图,对于每个像素,该值越大,黄色越深,表示该像素越有可能属于目标。高性能检测器会产生更加清晰的对比度图,突出目标像素并抑制背景像素。结合图9所示的试验结果,根据这些原理可以得出一些结论:
① ACE检测结果表明:该方法的目标提取能力和背景抑制能力相对平衡,但整体检测性能相对一般。② OSP在背景抑制方面表现较好,但其目标突出能力较弱,在所有方法中提取目标像素的效果最差。③ 与传统方法相比,PPF-CNN、SUB-PPF-CNN和RFS-PPF-CNN能更好地区分和识别目标和背景像素。在这三种方法中,PPF-CNN和RFS-PPF-CNN的识别策略更为激进,检测值普遍较大。这导致在保证目标提取质量的同时,对背景像素的抑制效果较差,特别是对背景轮廓像素的抑制效果较差。④ 相比之下,SUB-PPF-CNN在保证目标像素提取质量的前提下,可以在一定程度上降低背景像素的值。然而,SUB-PPF-CNN的背景抑制能力在不同的数据集之间差异很大,仅在某些场景下对背景轮廓有较好的抑制效果。
相比之下,本文所提出的检测方法能够以更少的噪声像素生成更显著的目标检测图。此外,可以更清晰地区分目标和背景部分,特别是对背景轮廓像素有很好的抑制效果。实验结果表明:该检测方法在提取目标像素和抑制背景像素方面是有效的。
除了可视化分析,本文还使用了一些评价指标来评价不同的比较方法,包括统计可分性分析、受试者工作特征(ROC)[42]和曲线下面积(AUC)指标作为主要评价标准,为更直观地展现方法检测结果的差异性,将ROC曲线坐标轴设置为对数。
为了验证所提方法每一步的有效性,设置了三个对比实验,流程如下:
① 使用经典SUB-PPF进行像素配对,得到训练样本。对于每个场景,手动选择3~8个背景类,每个类100像素,取场景中1/3的目标类像素作为目标类进行像素配对;使用本文提出的轻量化卷积神经网络提取样本特征;使用粒子群优化算法优化网络超参数。这个方法被称为方法1。
② 使用本文提出的样本生成方法获取训练样本;使用Du等人提出的29层深度CNN网络提取样本特征,网络epochs为30,卷积核大小为1×16×1;使用粒子群优化网络超参数。这个方法被称为方法2。
③ 使用本文提出的样本生成方法获取训练样本;使用本文提出的轻量化卷积神经网络提取样本特征;人工调整网络超参数,初始学习率为0.01,初始批处理大小为256。这个方法被称为方法3。
图15为本文提出的方法及消融实验对比方法在不同数据集上的比较结果,对于每个像素,该值越大,黄色越深,表示该像素越有可能属于目标。从左至右分别为方法1~方法3的检测图、本文提出方法的检测图以及地面真值图。表2为本文提出的方法及消融实验对比方法在不同数据集上检测结果的AUC值和单次训练所需时间(以秒为单位)。
结合图15表2可以看出:与本文提出的方法相比,方法1的检测结果图背景更加混乱,在不同数据集上的AUC值波动较大。此外,由于SUB-PPF生成的样本数量远大于本文方法生成的样本数量,方法1的单次训练时间是本文方法的5倍以上。
方法2由于其深层网络结构,在不同场景下都是稳定的。然而,由于本文提出的样本匹配方法产生的纯净样本数量较少,无法充分发挥深度网络的作用。因此,方法2在不同数据集上的综合AUC值略低于本文方法,且单次训练所需的时间是本文方法的两倍以上。
与手动调整网络的超参数相比,粒子群算法自动发现的超参数组合通常更优。首先,粒子群算法可以更全面地搜索超参数空间,通过引入群和随机性来避免局部最优解。其次,超参数通常是相互关联的,在手动调整超参数时,一般无法考虑到参数的组合。此外,对于不同的数据集,网络的最优参数组合是不同的。粒子群算法的引入可以为每个检测任务找到更好的参数组合,从而获得高精度、鲁棒性更强的检测结果。
综上所述,通过与方法1、方法2、方法3进行比较,验证了每一步的有效性。本文提出的样本生成方法产生的样品更纯、更少,当用于训练时,显著提高了检测结果中背景区域的纯度,对特征提取质量和速度均起到重要作用。此外,较小的样本量和轻量级网络的结合将单次训练时间大幅度减少。粒子群的引入不仅大大减少了重复超参数调整所需的时间,而且提高了检测结果的准确性和鲁棒性。这组消融实验证明了每个步骤对实验结果的重要贡献。
本文提出了一种基于数据增强和自动超参数优化的高光谱目标检测方法,该方法构建了自动化样本生成模型,采用SUB-PPF方法扩大样本量,实现了数据增强的同时减少了样本冗余;设计了轻量化的CNN网络,在提高训练效率的情况下,能够整合来自不同频谱通道的信息;而引入粒子群优化算法自动优化网络的超参数,不但减少了人工调整参数的重复工作,也有效地提高了目标检测的准确性和鲁棒性。综上,本文方法提高了高光谱目标检测的自动化水平,并能够适应不同的高光谱场景和样本集。
本文所提出的方法可以从以下两个方面作进一步研究:首先,可以引入残差网络到三维高光谱数据中,进一步提高网络性能,以发挥深度学习方法在提取高光谱图像中隐含的非线性特征方面的能力;其次,粒子群优化算法可以引入自适应的搜索方向调整机制,以及深度预测略等方法来提高算法的效率和准确性。
  • 国家自然科学基金(41776182)
  • 山东省高校青年创新技术支持计划(2023KJ068)
参考文献 引证文献
排序方式:
[1]
王盛铭, 王涛. 高光谱图像目标检测综述[C]//国家新闻出版广电总局中国新闻文化促进会学术期刊专业委员会. 2020年第三届智慧教育与人工智能发展国际学术会议论文集(第二卷), 2020: 131-137.
[2]
MANOLAKIS D, TRUSLOW E, PIEPER M, et al. Detection algorithms in hyperspectral imaging systems: An overview of practical algorithms[J]. IEEE Signal Processing Magazine, 2013, 31(1): 24-33.
[3]
宋尚真, 杨怡欣, 王会峰, . 高光谱图像稀疏约束与自编码器特征提取相结合的异常检测方法[J].测绘学报, 2023, 52(6): 932-943.
SONG Shangzhen, YANG Yixin, WANG Huifeng, et al. Hyperspectral anomaly detection combining sparse constraint and feature extraction via stacked autoencoder[J].Acta Geodaetica et Cartographica Sinica, 2023, 52(6):932-943.
[4]
CHEN L, LIU J, CHEN W, et al. A GLRT-based multi-pixel target detector in hyperspectral imagery[J]. IEEE Transactions on Multimedia, 2022, 25: 2710-2722.
[5]
ZHANG C, ZUO R, XIONG Y. Detection of the multivariate geochemical anomalies associated with mineralization using a deep convolutional neural network and a pixelpair feature method[J]. Applied Geochemistry, 2021, 130: 104994.
[6]
陈静. 基于深度学习的高光谱图像分类算法研究[D].兰州: 西北师范大学, 2021.
[7]
FU X, SHANG X, SUN X, et al. Underwater hyperspectral target detection with band selection[J]. Remote Sensing, 2020, 12(7): 1056.
[8]
ARAD B, BEN-SHAHAR O. Filter selection for hyperspectral estimation[C]//Proceedings of the IEEE International Conference on Computer Vision. 2017: 3153-3161.
[9]
蔡明娇, 蒋俊正, 蔡万源, . 张量分解和自适应图全变分的高光谱图像去噪[J/OL].西安电子科技大学学报, 2024, 51(2): 157-169[2024-01-29]. https://doi.org/10.19665/j.issn1001-2400.20230412.
[10]
ROBEY F C, FUHRMANN D R, KELLY E J, et al. A CFAR adaptive matched filter detector[J]. IEEE Transactions on Aerospace and Electronic Systems, 1992, 28(1):208-216.
[11]
谢少彪, 齐乃明, 秦雷, . 高光谱亚像元图像仿真与异常检测应用[J]. 哈尔滨工业大学学报, 2021, 53(12):10-15.
XIE Shaobiao, QI Naiming, QIN Lei, et al. Hyperspectral subpixel image simulation and application to anomaly detection[J]. Journal of Harbin Institute of Technology, 2021, 53(12):10-15.
[12]
张良培, 李家艺. 高光谱图像稀疏信息处理综述与展望[J]. 遥感学报, 2016, 20(5): 1091-1101.
ZHANG Liangpei, LI Jiayi. Development and prospect of sparse representation-based hyperspectral image processing and analysis[J]. Journal of Remote Sensing, 2016, 20(5): 1091-1101.
[13]
赵春晖, 李晓慧, 朱海峰. 空间4-邻域稀疏表示的高光谱图像目标检测[J]. 哈尔滨工程大学学报, 2013, 34(9): 1171-1178.
ZHAO Chunhui, LI Xiaohui, ZHU Haifeng. Hyperspectral imaging target detection algorithm based on spatial 4 neighborhoods for sparse representation[J]. Journal of Harbin Engineering University, 2013, 34(9): 1171-1178.
[14]
张帅洋, 华文深, 刘杰, . 基于超像素和低秩的协同稀疏高光谱解混[J]. 激光技术, 2022, 46(2): 199-205.
ZHANG Shuaiyang, HUA Wenshen, LIU Jie, et al. Superpixels and low rank for collaborative sparse hyperspectral unmixing[J]. Laser Technology, 2022, 46(2):199-205.
[15]
ZHU D, DU B, ZHANG L. Single-spectrum-driven binary-class sparse representation target detector for hyperspectral imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 59(2): 1487-1500.
[16]
JIA X, RICHARDS J A. Managing the spectral-spatial mix in context classification using Markov random fields[J]. IEEE Geoscience and Remote Sensing Letters, 2008, 5(2): 311-314.
[17]
BAKOS K L, GAMBA P. Hierarchical hybrid decision tree fusion of multiple hyperspectral data processing chains[J]. IEEE Transactions on Geoscience and Remote Sensing, 2010, 49(1): 388-394.
[18]
王振武, 孙佳骏, 于忠义, . 基于支持向量机的遥感图像分类研究综述[J]. 计算机科学, 2016, 43(9): 11-17, 31.
WANG Zhenwu, SUN Jiajun, YU Zhongyi, et al. Review of remote sensing image classification based on support vector machine[J]. Computer Science, 2016, 43(9): 11-17, 31.
[19]
SNEHA, KAUL A. Hyperspectral imaging and target detection algorithms: A review[J]. Multimedia Tools and Applications, 2022, 81(30): 44141-44206.
[20]
LI W, WU G, DU Q. Transferred deep learning for anomaly detection in hyperspectral imagery[J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14(5):597-601.
[21]
LIU X, WANG C, SUN Q, et al. Target detection of hyperspectral image based on convolutional neural networks[C]//2018 37th Chinese Control Conference(CCC). IEEE, 2018: 9255-9260.
[22]
FREITAS S, SILVA H, ALMEIDA J M, et al. Convolutional neural network target detection in hyperspectral imaging for maritime surveillance[J]. International Journal of Advanced Robotic Systems, 2019, 16(3):1729881419842991.
[23]
WANG X, CHEN X, GOU S, et al. Hyperspectral target detection via deep multiple instance self-attention neural network[C]//IGARSS 2019-2019 IEEE International Geoscience and Remote Sensing Symposium. IEEE,2019: 2284-2287.
[24]
ZHU D, DU B, ZHANG L. Two-stream convolutional networks for hyperspectral target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 59(8): 6907-6921.
[25]
ANDERSON D Z, ZOLLWEG J D, SMITH B J. Paired neural networks for hyperspectral target detection[C]//SPIE Optical Engineering+Applications, August 11-15,2019, San Diego, California, United States. Washington:SPIE, 2019, Proceedings Volume 11139, Applications of Machine Learning: 157-167.
[26]
WANG Y, CHEN X, WANG F, et al. Meta-learning based hyperspectral target detection using Siamese network[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 1-13.
[27]
DONG S, QUAN Y, FENG W, et al. Ensemble CNN based on pixelpair and random feature selection for hyperspectral image classification with small-size training set[C]//2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS. IEEE, 2021: 2353-2356.
[28]
DU J, LI Z. A hyperspectral target detection framework with subtraction pixel pair features[J]. IEEE Access, 2018, 6: 45562-45577.
[29]
JAIN M, SAIHJPAL V, SINGH N, et al. An overview of variants and advancements of PSO algorithm[J]. Applied Sciences, 2022, 12(17): 8392.
[30]
REN Hsuan, CHANG Chein I. Automatic spectral target recognition in hyperspectral imagery[J]. IEEE Transactions on Aerospace and Electronic Systems, 2023, 39(4):1232-1249.
[31]
JAIN N K, NANGIA U, JAIN J. A review of particle swarm optimization[J]. Journal of The Institution of Engineers (India): Series B, 2018, 99: 407-411.
[32]
赵乃刚, 邓景顺. 粒子群优化算法综述[J]. 科技创新导报, 2015, 12(26): 216-217.
[33]
DENG W, YAO R, ZHAO H, et al. A novel intelligent diagnosis method using optimal LS-SVM with improved PSO algorithm[J]. Soft Computing, 2019, 23:2445-2462.
[34]
CHANG C I, PLAZA A. A fast iterative algorithm for implementation of pixel purity index[J]. IEEE Geoscience and Remote Sensing Letters, 2006, 3(1): 63-67.
[35]
钟熙, 孙祥娥. 基于Kmeans++聚类的朴素贝叶斯集成方法研究[J]. 计算机科学, 2019, 46(S1): 439-441, 451.
[36]
LEI J, FANG S, XIE W, et al. Discriminative reconstruction for hyperspectral anomaly detection with spectral learning[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(10): 7406-7417.
[37]
ZHANG Y, WU K, DU B, et al. Hyperspectral target detection via adaptive joint sparse representation and multitask learning with locality information[J]. Remote Sensing, 2017, 9(5): 482.
[38]
王朗宁, 侯炎磐, 李彦峰. 数据参数影响RCS统计特征数据可分性[J]. 雷达科学与技术, 2020, 18(2): 205-210, 217.
WANG Langning, HOU Yanpan, LI Yanfeng. Influences of sampling frequency and parameters of sliding window on radar target recognition using radar cross section features[J]. Radar Science and Technology, 2020, 18(2):205-210, 217.
[39]
HANLEY J A, MCNEIL B J. The meaning and use of the area under a receiver operating characteristic (ROC)curve[J]. Radiology, 1982, 143(1): 29-36.
[40]
GREEN R O, EASTWOOD M L, SARTURE C M,et al. Imaging spectroscopy and the airborne visible/infrared imaging spectrometer (AVIRIS)[J]. Remote sensing of environment, 1998, 65(3): 227-248.
[41]
ZHANG Y, FAN Y, XU M, et al. An improved low rank and sparse matrix decomposition-based anomaly target detection algorithm for hyperspectral imagery[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020, 13: 2663-2672.
[42]
KANG X, ZHANG X, LI S, et al. Hyperspectral anomaly detection with attribute and edge-preserving filters[J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(10): 5600-5611.
2024年第45卷第4期
PDF下载
95
42
引用本文
BibTeX
文章信息
doi: 10.12347/j.ycyk.20240129001
  • 接收时间:2024-01-29
  • 首发时间:2026-03-20
  • 出版时间:2024-07-15
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-01-29
  • 修回日期:2024-02-19
基金
国家自然科学基金(41776182)
山东省高校青年创新技术支持计划(2023KJ068)
作者信息
    1中国石油大学(华东)海洋与空间信息学院 青岛 266580
    2山东鲁邦地理信息工程有限公司 济南 250102

通讯作者:

参考文献
分享链接
https://castjournals.cast.org.cn/joweb/ycyk/CN/10.12347/j.ycyk.20240129001
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