Article(id=1239136722274014142, tenantId=1146029695717560320, journalId=1238841944844054536, issueId=1239136718008414574, articleNumber=null, orderNo=null, doi=10.12347/j.ycyk.20250115002, pmid=null, cstr=32406.14.ycyk.20250115002, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1736870400000, receivedDateStr=2025-01-15, revisedDate=1741363200000, revisedDateStr=2025-03-08, acceptedDate=null, acceptedDateStr=null, onlineDate=1773362816637, onlineDateStr=2026-03-13, pubDate=1752508800000, pubDateStr=2025-07-15, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1773362816637, onlineIssueDateStr=2026-03-13, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1773362816637, creator=13701087609, updateTime=1773362816637, updator=13701087609, issue=Issue{id=1239136718008414574, tenantId=1146029695717560320, journalId=1238841944844054536, year='2025', volume='46', issue='4', pageStart='1', pageEnd='148', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1773362815620, creator=13701087609, updateTime=1773393299751, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1239264577775522632, tenantId=1146029695717560320, journalId=1238841944844054536, issueId=1239136718008414574, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1239264577775522633, tenantId=1146029695717560320, journalId=1238841944844054536, issueId=1239136718008414574, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=122, endPage=131, ext={EN=ArticleExt(id=1239136722936714195, articleId=1239136722274014142, tenantId=1146029695717560320, journalId=1238841944844054536, language=EN, title=Internal Wave Signature Extraction in The Andaman Sea Based on Swin-Unet, columnId=1239133500033528732, journalTitle=Journal of Telemetry, Tracking and Command, columnName=Radar and Countermeasures, runingTitle=null, highlight=null, articleAbstract=

Ocean internal wave is a common and significant ocean phenomenon that occurs between different density layers in the ocean. In this study, 92 Sentinel-1 images containing internal solitary waves, collected from 2015 to 2019, were used. Through preprocessing and visual interpretation, label data were generated to form a high-quality dataset containing 4 608 pairs of sample images. To address the inadequacy in the research of multi-scale internal solitary wave feature extraction, this paper proposes a multiscale internal solitary wave feature extraction model based on the Swin-Unet network to improve the extraction capability for internal solitary waves at different scales. The performance evaluation results show that, compared to the U-Net method, the proposed Swin-Unet model improved by 2.3% in F1 score, 2.44% in precision, and 12.12% in mean intersection over union (mIoU). The internal solitary wave extraction results at different scales in the Andaman Sea were subsequently analyzed, and the improved model was applied to Sentinel-1 full-track SAR imagery, verifying its applicability and robustness in complex marine regions. Experimental results demonstrate that the proposed model, based on the constructed dataset, can automatically extract ocean internal solitary wave features from SAR images.

, 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=Rui LONG, Junmin MENG, Lina SUN, Yonggang JI), CN=ArticleExt(id=1239136729769234567, articleId=1239136722274014142, tenantId=1146029695717560320, journalId=1238841944844054536, language=CN, title=基于Swin-Unet的安达曼海内孤立波特征信息提取, columnId=1239133500683645881, journalTitle=遥测遥控, columnName=雷达与对抗, runingTitle=null, highlight=null, articleAbstract=

海洋内波是发生在海洋中不同密度层之间的一种常见且重要的海洋现象。本研究收集了2015年至2019年期间92景包含内孤立波的哨兵一号影像,通过预处理和目视解译生成标签数据,构建了包含4 608对样本图像的高质量数据集。针对多尺度内孤立波特征提取研究不足的问题,本文提出了一种基于Swin-Unet(一种图像分割网络)网络的多尺度内孤立波特征提取模型,用于提高对不同尺度内孤立波的提取能力。模型性能评估结果显示,相比U-Net方法,本文所提的Swin-Unet模型在F1分数上提高了2.3%,精确率提高了2.44%,平均交并比提高了12.12%。同时,进一步分析了安达曼海不同尺度内孤立波的提取结果,并将改进模型应用于哨兵一号整轨SAR影像,验证了其在复杂海域的适用性和鲁棒性。实验结果表明,所提模型基于构建的数据集能够实现对SAR图像中海洋内孤立波特征的自动提取。

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=tV3g/xmJd6lXbKPSbmrqww==, magXml=XDn13wkEpWfzTpsCzKr7AQ==, pdfUrl=null, pdf=vlOTPawj81AAvrzzwDpXLg==, pdfFileSize=11325289, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=ifIiR0ItEwsHX4Hg3OqkkA==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=/wmVhTxhsd68prtgDEjZ/Q==, mapNumber=null, authorCompany=null, fund=null, authors=

龙瑞 2000年生,硕士,硕士研究生。

孟俊敏 1973年生,博士,研究员。

孙丽娜 1985年生,博士,工程师。

纪永刚 1970年生,教授,博士生导师。

, authorsList=龙瑞, 孟俊敏, 孙丽娜, 纪永刚)}, authors=[Author(id=1239263356515840268, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, 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=1239263356616503571, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, authorId=1239263356515840268, language=EN, stringName=Rui LONG, firstName=Rui, middleName=null, lastName=LONG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1.First Institute of Oceanography, Ministry of Natural Resources, Qingdao, 266061, China
2.College of Oceanography and Space Informatics, China University of Petroleum, Qingdao 266580, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1239263356704583960, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, authorId=1239263356515840268, language=CN, stringName=龙瑞, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1自然资源部第一海洋研究所 青岛 266061
2中国石油大学(华东)海洋与空间信息学院 青岛 266580, bio={"content":"

龙瑞 2000年生,硕士,硕士研究生。

"}, bioImg=null, bioContent=

龙瑞 2000年生,硕士,硕士研究生。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1239263356289347839, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, xref=1., ext=[AuthorCompanyExt(id=1239263356297736447, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, companyId=1239263356289347839, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.First Institute of Oceanography, Ministry of Natural Resources, Qingdao, 266061, China), AuthorCompanyExt(id=1239263356306125056, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, companyId=1239263356289347839, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1自然资源部第一海洋研究所 青岛 266061)]), AuthorCompany(id=1239263356402594052, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, xref=2., ext=[AuthorCompanyExt(id=1239263356410982661, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, companyId=1239263356402594052, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.College of Oceanography and Space Informatics, China University of Petroleum, Qingdao 266580, China), AuthorCompanyExt(id=1239263356419371270, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, companyId=1239263356402594052, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2中国石油大学(华东)海洋与空间信息学院 青岛 266580)])]), Author(id=1239263356775887131, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, 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=1239263356872356130, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, authorId=1239263356775887131, language=EN, stringName=Junmin MENG, firstName=Junmin, middleName=null, lastName=MENG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.First Institute of Oceanography, Ministry of Natural Resources, Qingdao, 266061, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1239263356960436516, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, authorId=1239263356775887131, language=CN, stringName=孟俊敏, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1自然资源部第一海洋研究所 青岛 266061, bio={"content":"

孟俊敏 1973年生,博士,研究员。

"}, bioImg=null, bioContent=

孟俊敏 1973年生,博士,研究员。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1239263356289347839, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, xref=1., ext=[AuthorCompanyExt(id=1239263356297736447, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, companyId=1239263356289347839, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.First Institute of Oceanography, Ministry of Natural Resources, Qingdao, 266061, China), AuthorCompanyExt(id=1239263356306125056, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, companyId=1239263356289347839, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1自然资源部第一海洋研究所 青岛 266061)])]), Author(id=1239263357040128300, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1239263357157568817, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, authorId=1239263357040128300, language=EN, stringName=Lina SUN, firstName=Lina, middleName=null, lastName=SUN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.First Institute of Oceanography, Ministry of Natural Resources, Qingdao, 266061, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1239263357237260597, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, authorId=1239263357040128300, language=CN, stringName=孙丽娜, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1自然资源部第一海洋研究所 青岛 266061, bio={"content":"

孙丽娜 1985年生,博士,工程师。

"}, bioImg=null, bioContent=

孙丽娜 1985年生,博士,工程师。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1239263356289347839, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, xref=1., ext=[AuthorCompanyExt(id=1239263356297736447, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, companyId=1239263356289347839, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.First Institute of Oceanography, Ministry of Natural Resources, Qingdao, 266061, China), AuthorCompanyExt(id=1239263356306125056, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, companyId=1239263356289347839, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1自然资源部第一海洋研究所 青岛 266061)])]), Author(id=1239263357304369467, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, 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=1239263357405032767, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, authorId=1239263357304369467, language=EN, stringName=Yonggang JI, firstName=Yonggang, middleName=null, lastName=JI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.College of Oceanography and Space Informatics, China University of Petroleum, Qingdao 266580, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1239263357484724548, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, authorId=1239263357304369467, language=CN, stringName=纪永刚, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2中国石油大学(华东)海洋与空间信息学院 青岛 266580, bio={"content":"

纪永刚 1970年生,教授,博士生导师。

"}, bioImg=null, bioContent=

纪永刚 1970年生,教授,博士生导师。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1239263356402594052, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, xref=2., ext=[AuthorCompanyExt(id=1239263356410982661, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, companyId=1239263356402594052, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.College of Oceanography and Space Informatics, China University of Petroleum, Qingdao 266580, China), AuthorCompanyExt(id=1239263356419371270, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, companyId=1239263356402594052, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2中国石油大学(华东)海洋与空间信息学院 青岛 266580)])])], keywords=[Keyword(id=1239263357665079628, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, orderNo=1, keyword=Internal solitary waves), Keyword(id=1239263357753160016, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, orderNo=2, keyword=Synthetic aperture radar), Keyword(id=1239263357816074582, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, orderNo=3, keyword=Deep learning), Keyword(id=1239263357929320794, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, orderNo=4, keyword=Swin-Unet), Keyword(id=1239263358017401183, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, orderNo=5, keyword=Feature extraction), Keyword(id=1239263358105481575, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, orderNo=1, keyword=内孤立波), Keyword(id=1239263358290030958, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, orderNo=2, keyword=合成孔径雷达), Keyword(id=1239263358386499955, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, orderNo=3, keyword=深度学习), Keyword(id=1239263358466191736, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, orderNo=4, keyword=Swin-Unet), Keyword(id=1239263358520717694, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, orderNo=5, keyword=特征提取)], refs=[Reference(id=1239263361792274975, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=1979, volume=278, issue=5702, pageStart=312, pageEnd=317, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=HAURY L R, BRISCOE M G, ORR M H, journalName=Nature, refType=null, unstructuredReference=HAURY L R, BRISCOE M G, ORR M H. Tidally generated internal wave packets in Massachusetts Bay[J]. Nature, 1979, 278(5702): 312-317., articleTitle=Tidally generated internal wave packets in Massachusetts Bay, refAbstract=null), Reference(id=1239263361880355365, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2016, volume=54, issue=11, pageStart=6674, pageEnd=6686, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=DONG D, YANG X, LI X, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=DONG D, YANG X, LI X, et al. SAR observation of eddy-induced mode-2 internal solitary waves in the south china sea[J]. IEEE Transactions on Geoscience and Remote Sensing, 2016, 54(11): 6674-6686., articleTitle=SAR observation of eddy-induced mode-2 internal solitary waves in the south china sea, refAbstract=null), Reference(id=1239263361972630056, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=1997, volume=35, issue=4, pageStart=933, pageEnd=945, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=RODENAS J A, GARELLO R, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=RODENAS J A, GARELLO R. Wavelet analysis in SAR ocean image profiles for internal wave detection and wavelength estimation[J]. IEEE Transactions on Geoscience and Remote Sensing, 1997, 35(4): 933-945., articleTitle=Wavelet analysis in SAR ocean image profiles for internal wave detection and wavelength estimation, refAbstract=null), Reference(id=1239263362102653482, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=1998, volume=36, issue=5, pageStart=1494, pageEnd=1507, url=null, language=null, rfNumber=[4], rfOrder=3, authorNames=RODENAS J A, GARELLO R, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=RODENAS J A, GARELLO R. Internal wave detection and location in SAR images using wavelet transform[J].IEEE Transactions on Geoscience and Remote Sensing,1998, 36(5): 1494-1507, articleTitle=Internal wave detection and location in SAR images using wavelet transform, refAbstract=null), Reference(id=1239263362169762346, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2001, volume=28, issue=14, pageStart=2803, pageEnd=2806, url=null, language=null, rfNumber=[5], rfOrder=4, authorNames=DOKKEN S T, OLSEN R, WAHL T, journalName=Geophysical Research Letters, refType=null, unstructuredReference=DOKKEN S T, OLSEN R, WAHL T, et al. Identification and characterization of internal waves in SAR images along the coast of Norway[J]. Geophysical Research Letters, 2001, 28(14): 2803-2806., articleTitle=Identification and characterization of internal waves in SAR images along the coast of Norway, refAbstract=null), Reference(id=1239263362245259821, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=104, pageEnd=108, url=null, language=null, rfNumber=[6], rfOrder=5, authorNames=SURAMPUDI S, SASANKA S, journalName=null, refType=null, unstructuredReference=SURAMPUDI S, SASANKA S. Internal wave detection and characterization with SAR data[C]//2019 IEEE Recent Advances in Geoscience and Remote Sensing: Technologies, Standards and Applications (TENGARSS). Kochi, Kerala, India: IEEE, 2019: 104-108., articleTitle=Internal wave detection and characterization with SAR data, refAbstract=null), Reference(id=1239263362324951601, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2009, volume=30, issue=17, pageStart=4581, pageEnd=4598, url=null, language=null, rfNumber=[7], rfOrder=6, authorNames=SIMONIN D, TATNALL A R, ROBINSON I S, journalName=International Journal of Remote Sensing, refType=null, unstructuredReference=SIMONIN D, TATNALL A R, ROBINSON I S. The auto-mated detection and recognition of internal waves[J]. International Journal of Remote Sensing, 2009, 30(17):4581-4598., articleTitle=The auto-mated detection and recognition of internal waves, refAbstract=null), Reference(id=1239263362413031988, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2018, volume=169, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[8], rfOrder=7, authorNames=MARGHANY M, journalName=IOP Conference Series: Earth and Environmental Science, refType=null, unstructuredReference=MARGHANY M. Automatic detection of internal wave using particle swarm optimization algorithm[J]. IOP Conference Series: Earth and Environmental Science, 2018,169: 012049., articleTitle=Automatic detection of internal wave using particle swarm optimization algorithm, refAbstract=null), Reference(id=1239263362505306680, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2020, volume=12, issue=15, pageStart=2476, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=8, authorNames=KUREKIN A A, LAND P E, MILLER P I, journalName=Remote Sensing, refType=null, unstructuredReference=KUREKIN A A, LAND P E, MILLER P I. Internal waves at the uk continental shelf: Automatic mapping using the ENVISAT ASAR sensor[J]. Remote Sensing, 2020, 12(15): 2476., articleTitle=Internal waves at the uk continental shelf: Automatic mapping using the ENVISAT ASAR sensor, refAbstract=null), Reference(id=1239263362589192764, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2020, volume=7, issue=10, pageStart=1584, pageEnd=1605, url=null, language=null, rfNumber=[10], rfOrder=9, authorNames=LI X, LIU B, ZHENG G, journalName=National Science Review, refType=null, unstructuredReference=LI X, LIU B, ZHENG G, et al. Deep-learning-based information mining from ocean remote-sensing imagery[J]. National Science Review, 2020, 7(10): 1584-1605., articleTitle=Deep-learning-based information mining from ocean remote-sensing imagery, refAbstract=null), Reference(id=1239263362685661760, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2020, volume=38, issue=1, pageStart=55, pageEnd=63, url=null, language=null, rfNumber=[11], rfOrder=10, authorNames=BAO S, MENG J, SUN L, journalName=Journal of Oceanology and Limnology, refType=null, unstructuredReference=BAO S, MENG J, SUN L, et al. Detection of ocean internal waves based on Faster R-CNN in SAR images[J]. Journal of Oceanology and Limnology, 2020, 38(1): 55-63., articleTitle=Detection of ocean internal waves based on Faster R-CNN in SAR images, refAbstract=null), Reference(id=1239263362756964931, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2022, volume=9, issue=12, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[12], rfOrder=11, authorNames=TAO M, XU C, GUO L, journalName=Earth and Space Science, refType=null, unstructuredReference=TAO M, XU C, GUO L, et al. An internal waves data set from sentinel-1 synthetic aperture radar imagery and preliminary detection[J]. Earth and Space Science, 2022, 9(12): e2022EA002528., articleTitle=An internal waves data set from sentinel-1 synthetic aperture radar imagery and preliminary detection, refAbstract=null), Reference(id=1239263362845045319, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2021, volume=42, issue=17, pageStart=6706, pageEnd=6724, url=null, language=null, rfNumber=[13], rfOrder=12, authorNames=ZHENG Y G, ZHANG H S, WANG Y Q, journalName=International Journal of Remote Sensing, refType=null, unstructuredReference=ZHENG Y G, ZHANG H S, WANG Y Q. Stripe detection and recognition of oceanic internal waves from synthetic aperture radar based on support vector machine and feature fusion[J]. International Journal of Remote Sensing, 2021, 42(17): 6706-6724., articleTitle=Stripe detection and recognition of oceanic internal waves from synthetic aperture radar based on support vector machine and feature fusion, refAbstract=null), Reference(id=1239263362941514317, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2022, volume=37, issue=25, pageStart=8567, pageEnd=8578, url=null, language=null, rfNumber=[14], rfOrder=13, authorNames=ZHENG Y G, ZHANG H S, QI K T, journalName=Geocarto International, refType=null, unstructuredReference=ZHENG Y G, ZHANG H S, QI K T, et al. Stripe segmentation of oceanic internal waves in SAR images based on SegNet[J]. Geocarto International, 2022, 37(25): 8567-8578., articleTitle=Stripe segmentation of oceanic internal waves in SAR images based on SegNet, refAbstract=null), Reference(id=1239263363029594705, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2023, volume=20, issue=null, pageStart=1, pageEnd=5, url=null, language=null, rfNumber=[15], rfOrder=14, authorNames=MA Y, MENG J, SUN L, journalName=IEEE Geoscience and Remote Sensing Letters, refType=null, unstructuredReference=MA Y, MENG J, SUN L, et al. Oceanic internal wave signature extraction in the sulu sea by a pixel attention U-Net: PAU-Net[J]. IEEE Geoscience and Remote Sensing Letters, 2023, 20: 1-5., articleTitle=Oceanic internal wave signature extraction in the sulu sea by a pixel attention U-Net: PAU-Net, refAbstract=null), Reference(id=1239263363105092178, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2023, volume=61, issue=null, pageStart=1, pageEnd=16, url=null, language=null, rfNumber=[16], rfOrder=15, authorNames=ZHANG S, LI X, ZHANG X, journalName=IEEE Transactions on Geoscience and Remote Sensing, refType=null, unstructuredReference=ZHANG S, LI X, ZHANG X. Internal wave signature extraction from sar and optical satellite imagery based on deep learning[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 1-16., articleTitle=Internal wave signature extraction from sar and optical satellite imagery based on deep learning, refAbstract=null), Reference(id=1239263363172201043, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=1982, volume=87, issue=C5, pageStart=3397, pageEnd=3430, url=null, language=null, rfNumber=[17], rfOrder=16, authorNames=VESECKY J F, STEWART R H, journalName=Journal of Geophysical Research: Oceans, refType=null, unstructuredReference=VESECKY J F, STEWART R H. The observation of ocean surface phenomena using imagery from the SEASAT synthetic aperture radar: An assessment[J]. Journal of Geophysical Research: Oceans, 1982, 87(C5): 3397-3430., articleTitle=The observation of ocean surface phenomena using imagery from the SEASAT synthetic aperture radar: An assessment, refAbstract=null), Reference(id=1239263363251892820, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=1988, volume=93, issue=C10, pageStart=12355, pageEnd=12369, url=null, language=null, rfNumber=[18], rfOrder=17, authorNames=KASISCHKE E S, LYZENGA D R, SHUCHMAN R A, journalName=Journal of Geophysical Research: Oceans, refType=null, unstructuredReference=KASISCHKE E S, LYZENGA D R, SHUCHMAN R A,et al. Contrast ratios of internal waves in synthetic aperture radar imagery: A comparison of SAR internal wave signature experiment observations with theory[J]. Journal of Geophysical Research: Oceans, 1988, 93(C10):12355-12369., articleTitle=Contrast ratios of internal waves in synthetic aperture radar imagery: A comparison of SAR internal wave signature experiment observations with theory, refAbstract=null), Reference(id=1239263363331584598, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2017, volume=60, issue=6, pageStart=84, pageEnd=90, url=null, language=null, rfNumber=[19], rfOrder=18, authorNames=KRIZHEVSKY A, SUTSKEVER I, HINTON G E, journalName=Communications of the ACM, refType=null, unstructuredReference=KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6): 84-90., articleTitle=ImageNet classification with deep convolutional neural networks, refAbstract=null), Reference(id=1239263363411276376, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[20], rfOrder=19, authorNames=CAO H, WANG Y, CHEN J, journalName=arXiv, refType=null, unstructuredReference=CAO H, WANG Y, CHEN J, et al. Swin-UNet: Unet-like pure transformer for medical image segmentation[EB/OL]. arXiv:2021. (2021-05-12)[2024-03-20]. https://arxiv.org/abs/2105.05537., articleTitle=Swin-UNet: Unet-like pure transformer for medical image segmentation, refAbstract=null), Reference(id=1239263363499356764, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[21], rfOrder=20, authorNames=LIU Z, LIN Y, CAO Y, journalName=arXiv, refType=null, unstructuredReference=LIU Z, LIN Y, CAO Y, et al. Swin transformer: Hierarchical vision transformer using shifted windows[EB/OL]. arXiv:2021. (2021-03-25)[2024-03-20]. https://arxiv.org/abs/2103.14030., articleTitle=Swin transformer: Hierarchical vision transformer using shifted windows, refAbstract=null), Reference(id=1239263363612602976, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2015, volume=null, issue=null, pageStart=234, pageEnd=241, url=null, language=null, rfNumber=[22], rfOrder=21, authorNames=RONNEBERGER O, FISCHER P, BROX T, NAVAB N, HORNEGGER J, WELLS W M, journalName=null, refType=null, unstructuredReference=RONNEBERGER O, FISCHER P, BROX T. U-net:Convolutional networks for biomedical image segmentation[C]//NAVAB N, HORNEGGER J, WELLS W M, et al. Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015: 18th International Conference, October 5-9, 2015, Munich, Germany. Cham:Springer International Publishing, 2015: 234-241., articleTitle=U-net:Convolutional networks for biomedical image segmentation, refAbstract=null), Reference(id=1239263363704877667, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2005, volume=null, issue=null, pageStart=345, pageEnd=359, url=null, language=null, rfNumber=[23], rfOrder=22, authorNames=GOUTTE C, GAUSSIER E, LOSADA D E, FERNÁNDEZ-LUNA J M, journalName=null, refType=null, unstructuredReference=GOUTTE C, GAUSSIER E. A probabilistic interpretation of precision, recall and f-score, with implication for evaluation[C]//LOSADA D E, FERNÁNDEZ-LUNA J M. Advances in Information Retrieval: 27th European Conference on IR Research, ECIR 2005, March 21-23,2005, Santiago de Compostela, Spain. Berlin: Springer,2005: 345-359., articleTitle=A probabilistic interpretation of precision, recall and f-score, with implication for evaluation, refAbstract=null), Reference(id=1239263363788763750, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=8024, pageEnd=8035, url=null, language=null, rfNumber=[24], rfOrder=23, authorNames=PASZKE A, GROSS S, MASSA F, journalName=null, refType=null, unstructuredReference=PASZKE A, GROSS S, MASSA F, et al. PyTorch: An imperative style, high-performance deep learning library[C]//33rd Conference on Neural Information Processing Systems (NeurIPS 2019), December 8-14, 2019, Vancouver, Canada. New York: Curran Associates, 2019:8024-8035. arXiv:, articleTitle=PyTorch: An imperative style, high-performance deep learning library, refAbstract=null)], funds=[Fund(id=1239263361565782552, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, awardId=null, language=CN, fundingSource=国家自然科学基金(低空远距离条件下海态与目标一体化雷达探测理论与方法U26207), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1239263356289347839, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, xref=1., ext=[AuthorCompanyExt(id=1239263356297736447, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, companyId=1239263356289347839, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.First Institute of Oceanography, Ministry of Natural Resources, Qingdao, 266061, China), AuthorCompanyExt(id=1239263356306125056, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, companyId=1239263356289347839, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1自然资源部第一海洋研究所 青岛 266061)]), AuthorCompany(id=1239263356402594052, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, xref=2., ext=[AuthorCompanyExt(id=1239263356410982661, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, companyId=1239263356402594052, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.College of Oceanography and Space Informatics, China University of Petroleum, Qingdao 266580, China), AuthorCompanyExt(id=1239263356419371270, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, companyId=1239263356402594052, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2中国石油大学(华东)海洋与空间信息学院 青岛 266580)])], figs=[ArticleFig(id=1239263358747210123, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, label=Fig. 1, caption=Multiscale internal solitary wave, figureFileSmall=Rgf9/7DJQy1ZtVQwgjp9Xg==, figureFileBig=VjZ7v55FiRZjA/vh+QvMvA==, tableContent=null), ArticleFig(id=1239263358822707605, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, label=图1, caption=多尺度内孤立波, figureFileSmall=Rgf9/7DJQy1ZtVQwgjp9Xg==, figureFileBig=VjZ7v55FiRZjA/vh+QvMvA==, tableContent=null), ArticleFig(id=1239263358919176605, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, label=Fig. 2, caption=Histogram of wind speed, figureFileSmall=f1MRftho/WPy489hJcu+lQ==, figureFileBig=FZHNrLm2ZOdC+LystiqJtQ==, tableContent=null), ArticleFig(id=1239263359011451298, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, label=图2, caption=风速直方图, figureFileSmall=f1MRftho/WPy489hJcu+lQ==, figureFileBig=FZHNrLm2ZOdC+LystiqJtQ==, tableContent=null), ArticleFig(id=1239263359082754473, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, label=Fig. 3, caption=Data augmentation, figureFileSmall=UeLvCx26GlMOuxCGfaK7YA==, figureFileBig=unfD1m+7sN8t7qwKKAwjhg==, tableContent=null), ArticleFig(id=1239263359166640557, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, label=图3, caption=数据增强, figureFileSmall=UeLvCx26GlMOuxCGfaK7YA==, figureFileBig=unfD1m+7sN8t7qwKKAwjhg==, tableContent=null), ArticleFig(id=1239263359263109553, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, label=Fig. 4, caption=Swin-Unet network, figureFileSmall=/GHoUumISjTFNN/G5cLW9g==, figureFileBig=onYZF8dP5gfe/wZgETuSGA==, tableContent=null), ArticleFig(id=1239263359338607030, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, label=图4, caption=Swin-Unet网络, figureFileSmall=/GHoUumISjTFNN/G5cLW9g==, figureFileBig=onYZF8dP5gfe/wZgETuSGA==, tableContent=null), ArticleFig(id=1239263359430881723, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, label=Fig. 5, caption=Multiscale internal solitary wave and detection results, figureFileSmall=+65ZyXWeRu2ON88WZil2oA==, figureFileBig=tz3P0b0UCANJSkY8NISJUw==, tableContent=null), ArticleFig(id=1239263359552516548, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, label=图5, caption=多尺度内孤立波及检测结果, figureFileSmall=+65ZyXWeRu2ON88WZil2oA==, figureFileBig=tz3P0b0UCANJSkY8NISJUw==, tableContent=null), ArticleFig(id=1239263359653179850, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, label=Fig. 6, caption=Comparison of isolated wave feature extraction results of different models in multiple scales, figureFileSmall=0IXgqH8Exv3rwHm0/kfOpw==, figureFileBig=tJ+FFVdZbv5FA6YRjRceBw==, tableContent=null), ArticleFig(id=1239263359758037460, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, label=图6, caption=不同模型在多尺度内孤立波特征提取结果对比图, figureFileSmall=0IXgqH8Exv3rwHm0/kfOpw==, figureFileBig=tJ+FFVdZbv5FA6YRjRceBw==, tableContent=null), ArticleFig(id=1239263359858700763, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, label=Fig. 7, caption=SAR image and the detection results., figureFileSmall=NIv5X/WLUu7oWjCxMyKrxw==, figureFileBig=o1o3n0Ru110Ri5c+E0Prjg==, tableContent=null), ArticleFig(id=1239263359967752671, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, label=图7, caption=影像及检测结果, figureFileSmall=NIv5X/WLUu7oWjCxMyKrxw==, figureFileBig=o1o3n0Ru110Ri5c+E0Prjg==, tableContent=null), ArticleFig(id=1239263360064221666, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, label=Fig. 8, caption=The SAR imageand the detection results., figureFileSmall=agxhiKbb3BUJJHgSRv4Ssw==, figureFileBig=3kiCZS0ZMWDrQYYxwse0FA==, tableContent=null), ArticleFig(id=1239263360164884967, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, label=图8, caption=影像及检测结果, figureFileSmall=agxhiKbb3BUJJHgSRv4Ssw==, figureFileBig=3kiCZS0ZMWDrQYYxwse0FA==, tableContent=null), ArticleFig(id=1239263360273936874, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, label=Fig. 9, caption=The image and the detection results., figureFileSmall=6a0bFnWKvxPacp9/FPXTyw==, figureFileBig=djHKTUBVJhLgD2ZsU4Znmg==, tableContent=null), ArticleFig(id=1239263360357822957, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, label=图9, caption=影像和检测结果, figureFileSmall=6a0bFnWKvxPacp9/FPXTyw==, figureFileBig=djHKTUBVJhLgD2ZsU4Znmg==, tableContent=null), ArticleFig(id=1239263360462680562, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, label=Fig. 10, caption=The image and the detection results, figureFileSmall=9mEK5IGVf98Q2tkUiVXUmg==, figureFileBig=z2j6qcRcZ4EuJlMgTugMDQ==, tableContent=null), ArticleFig(id=1239263360571732471, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, label=图10, caption=影像和检测结果, figureFileSmall=9mEK5IGVf98Q2tkUiVXUmg==, figureFileBig=z2j6qcRcZ4EuJlMgTugMDQ==, tableContent=null), ArticleFig(id=1239263360680784379, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, label=Fig. 11, caption=The image and the detection results., figureFileSmall=8pHT78hgjvwoV4fuJGSozQ==, figureFileBig=GVOUI0O5OnbFtDvnVUdVDQ==, tableContent=null), ArticleFig(id=1239263360781447683, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, label=图11, caption=影像和检测结果, figureFileSmall=8pHT78hgjvwoV4fuJGSozQ==, figureFileBig=GVOUI0O5OnbFtDvnVUdVDQ==, tableContent=null), ArticleFig(id=1239263360882110980, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, label=Fig. 12, caption=The image and the detection results., figureFileSmall=YLqEFe2aS21Ky3Q44AdbJA==, figureFileBig=RR+L/qbIjW+FYzqjESIrqA==, tableContent=null), ArticleFig(id=1239263361049883143, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, label=图12, caption=影像和检测结果, figureFileSmall=YLqEFe2aS21Ky3Q44AdbJA==, figureFileBig=RR+L/qbIjW+FYzqjESIrqA==, tableContent=null), ArticleFig(id=1239263361146352139, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, label=Table 1, caption=

Sentinel-1 dataset parameters

, figureFileSmall=null, figureFileBig=null, tableContent=
参数
传感器Sentinel-1
成像模式IW模式
波长5.55 cm(C波段)
极化方式双极化方式(VV+VH)
空间分辨率(Range×Azimuth)5 m×20 m
采集时间2015年~2019年
), ArticleFig(id=1239263361251209742, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, label=表1, caption=

哨兵1号数据集参数

, figureFileSmall=null, figureFileBig=null, tableContent=
参数
传感器Sentinel-1
成像模式IW模式
波长5.55 cm(C波段)
极化方式双极化方式(VV+VH)
空间分辨率(Range×Azimuth)5 m×20 m
采集时间2015年~2019年
), ArticleFig(id=1239263361335095826, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=EN, label=Table 2, caption=

Evaluation index of comparative experiments (%)

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 Flscore精确率召回率平均交并比
U-Net78.2%68.8%90.57%50.35%
Swin-Unet80.5%71.24%87.92%62.47%
), ArticleFig(id=1239263361414787604, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1239136722274014142, language=CN, label=表2, caption=

对比实验评估指标(%)

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 Flscore精确率召回率平均交并比
U-Net78.2%68.8%90.57%50.35%
Swin-Unet80.5%71.24%87.92%62.47%
)], 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.20250115002, detailUrlEn=https://castjournals.cast.org.cn/joweb/ycyk/EN/10.12347/j.ycyk.20250115002, pdfUrlCn=https://castjournals.cast.org.cn/joweb/ycyk/CN/PDF/10.12347/j.ycyk.20250115002, pdfUrlEn=https://castjournals.cast.org.cn/joweb/ycyk/EN/PDF/10.12347/j.ycyk.20250115002, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于Swin-Unet的安达曼海内孤立波特征信息提取
收藏切换
PDF下载
龙瑞 1, 2 , 孟俊敏 1 , 孙丽娜 1 , 纪永刚 2
遥测遥控 | 雷达与对抗 2025,46(4): 122-131
收起
收藏切换
遥测遥控 | 雷达与对抗 2025, 46(4): 122-131
基于Swin-Unet的安达曼海内孤立波特征信息提取
全屏
龙瑞1, 2, 孟俊敏1, 孙丽娜1, 纪永刚2
作者信息
  • 1自然资源部第一海洋研究所 青岛 266061
  • 2中国石油大学(华东)海洋与空间信息学院 青岛 266580
  • 龙瑞 2000年生,硕士,硕士研究生。

    孟俊敏 1973年生,博士,研究员。

    孙丽娜 1985年生,博士,工程师。

    纪永刚 1970年生,教授,博士生导师。

Internal Wave Signature Extraction in The Andaman Sea Based on Swin-Unet
Rui LONG1, 2, Junmin MENG1, Lina SUN1, Yonggang JI2
Affiliations
  • 1.First Institute of Oceanography, Ministry of Natural Resources, Qingdao, 266061, China
  • 2.College of Oceanography and Space Informatics, China University of Petroleum, Qingdao 266580, China
出版时间: 2025-07-15 doi: 10.12347/j.ycyk.20250115002
文章导航
收藏切换

海洋内波是发生在海洋中不同密度层之间的一种常见且重要的海洋现象。本研究收集了2015年至2019年期间92景包含内孤立波的哨兵一号影像,通过预处理和目视解译生成标签数据,构建了包含4 608对样本图像的高质量数据集。针对多尺度内孤立波特征提取研究不足的问题,本文提出了一种基于Swin-Unet(一种图像分割网络)网络的多尺度内孤立波特征提取模型,用于提高对不同尺度内孤立波的提取能力。模型性能评估结果显示,相比U-Net方法,本文所提的Swin-Unet模型在F1分数上提高了2.3%,精确率提高了2.44%,平均交并比提高了12.12%。同时,进一步分析了安达曼海不同尺度内孤立波的提取结果,并将改进模型应用于哨兵一号整轨SAR影像,验证了其在复杂海域的适用性和鲁棒性。实验结果表明,所提模型基于构建的数据集能够实现对SAR图像中海洋内孤立波特征的自动提取。

内孤立波  /  合成孔径雷达  /  深度学习  /  Swin-Unet  /  特征提取

Ocean internal wave is a common and significant ocean phenomenon that occurs between different density layers in the ocean. In this study, 92 Sentinel-1 images containing internal solitary waves, collected from 2015 to 2019, were used. Through preprocessing and visual interpretation, label data were generated to form a high-quality dataset containing 4 608 pairs of sample images. To address the inadequacy in the research of multi-scale internal solitary wave feature extraction, this paper proposes a multiscale internal solitary wave feature extraction model based on the Swin-Unet network to improve the extraction capability for internal solitary waves at different scales. The performance evaluation results show that, compared to the U-Net method, the proposed Swin-Unet model improved by 2.3% in F1 score, 2.44% in precision, and 12.12% in mean intersection over union (mIoU). The internal solitary wave extraction results at different scales in the Andaman Sea were subsequently analyzed, and the improved model was applied to Sentinel-1 full-track SAR imagery, verifying its applicability and robustness in complex marine regions. Experimental results demonstrate that the proposed model, based on the constructed dataset, can automatically extract ocean internal solitary wave features from SAR images.

Internal solitary waves  /  Synthetic aperture radar  /  Deep learning  /  Swin-Unet  /  Feature extraction
龙瑞, 孟俊敏, 孙丽娜, 纪永刚. 基于Swin-Unet的安达曼海内孤立波特征信息提取. 遥测遥控, 2025 , 46 (4) : 122 -131 . DOI: 10.12347/j.ycyk.20250115002
Rui LONG, Junmin MENG, Lina SUN, Yonggang JI. Internal Wave Signature Extraction in The Andaman Sea Based on Swin-Unet[J]. Journal of Telemetry, Tracking and Command, 2025 , 46 (4) : 122 -131 . DOI: 10.12347/j.ycyk.20250115002
海洋内孤立波是密度分层海洋中一种典型的非线性波动,其最大振幅位于海洋内部,通常呈现孤立的波峰或波群形态[1]。大量的实地和遥感观测表明[2],内孤立波广泛分布于边缘海域和大陆架区域,能够在传播数百公里的过程中保持波形稳定,并携带大量能量,成为海洋能量输运的重要载体,同时在航海活动和海洋工程中具有重要作用。此外,受水动力条件和海底地形的影响,不同区域内孤立波的尺度差异显著,从小尺度到大尺度的波动展现出不同的传播特性和作用机制。基于此,内孤立波的研究已成为物理海洋和海洋工程等领域的研究重点。
目前,利用遥感技术观测海洋内孤立波主要依赖光学传感器和合成孔径雷达(SAR)。其中,SAR作为一种主动传感器,具备全天候、远距离、大范围和高分辨率的显著优势,且不受云雨天气的限制[3]。SAR遥感的应用突破了传统探测手段的限制,使得对内孤立波的大范围和长期观测成为可能,在提取不同尺度内孤立波特征方面展现出明显优势。内孤立波会调制海表面流场,形成辐聚辐散现象,从而在SAR图像上呈现亮暗相间的条纹结构,这使得SAR成为观测内孤立波的有效手段。然而,SAR海洋图像纹理复杂,容易受到岛屿、船舶尾流、中尺度涡以及锋面等其他海洋现象的干扰,加之传感器特性的局限性,使得内孤立波的自动检测仍然面临较大挑战。
自上世纪九十年代以来,基于图像处理的内孤立波检测技术在遥感影像分析中得到了广泛研究与应用。早期方法主要基于小波分析和傅里叶变换。例如,Rodenas和Garello等[4,5]基于小波分析方法能够自动检测SAR图像中的内孤立波特征,Dokken等[6]在挪威海岸附近内孤立波研究中基于小波变换、傅里叶变换和表征算法进行了对比,验证了各个方法的效果。此后,Surampudi和Sasanka等[7]通过对安达曼海和莫桑比克海峡的内孤立波图像应用小波和傅里叶变换,进一步探讨了二者在不同类型内孤立波研究中的适用性。此外,Simonin等[8]结合小波分析和边缘检测,从东大西洋地区的SAR图像中提取内孤立波特征;Marghany等[9]利用粒子群优化算法和小波变换来检测安达曼海和尼科巴群岛的内孤立波;而Kurekin等[10]则结合空间处理和增强边缘检测算法,从英国大陆架的数千幅ENVISAT ASAR图像中成功分离出内孤立波特征。随着研究的深入,传统方法逐渐与机器学习技术结合,通过融合小波分析、边缘检测等方法与机器学习模型,提高了内孤立波自动检测的精度与适用性,但在面对大规模、多源遥感数据时仍存在一定的局限性,亟需更高效、自动化的解决方案以推动该领域的发展。
近年来,深度学习的快速发展为遥感数据特征提取,尤其是内孤立波特征提取研究,带来了新的契机[11]。相较于传统方法,深度学习凭借自动化、高效性等优势,广泛应用于内孤立波的检测与特征提取。研究者提出了多种基于深度学习的内孤立波检测算法:Bao等[12]基于Faster R-CNN算法检测SAR影像中的内孤立波,分析了其在不同区域和形态下的表现;Tao等[13]基于Yolo v5对哨兵一号SAR影像的内孤立波进行检测和分类;Zheng等[14,15]结合支持向量机(SVM)和Canny边缘检测技术获取内孤立波的几何特征,并进一步使用SegNet(一种深度学习模型)模型分割亮暗条纹;Ma等[16]设计了两阶段分割算法,分类内孤立波影像并结合像素注意力U-Net进行条纹分割;Zhang等[17]开发了IWE-Net(内孤立波提取网络)模型,从多源影像中提取内孤立波特征,同时并通过Tensor Voting(张量投票)算法弥补了云遮挡导致的不连续问题。尽管这些方法提高了内孤立波的检测精度,在复杂背景下也表现出较好的适用性,但仍存在局限,如泛化能力有限,以及难以精确提取复杂内孤立波特征的信息。然而,以往的研究往往针对单一尺度的内孤立波图像进行实验和验证,难以覆盖不同尺度内孤立波特征,无法满足内孤立波检测的实际应用需求。
为此,本文基于多尺度注意力机制的Swin-Unet条纹分割算法,用于从SAR图像中精确提取不同尺度内孤立波的波峰线信息,并结合Dice Loss(损失函数)和Focal Loss(焦点损失),显著改善类别不平衡问题,进而提高模型的分割精度和鲁棒性,同时处理效率较高。
图1(a)所示,其中小尺度(17~167像素,500 m~5 km),小尺度内孤立波通常较短且振幅低;如图1(b)所示,中尺度(167~1 667像素,5 km ~50 km),中尺度内孤立波具有更明显的波包结构;如图1(c)所示,大尺度(1 667~10 000像素,50 km ~300 km),大尺度内孤立波则呈周期性波阵,具有显著的传播特性。
安达曼海作为东印度洋的重要边缘海,其内孤立波呈现出多尺度分布。在安达曼海北部沿岸区域,主要观察到小尺度内孤立波,波长通常为几公里,振幅较小,且受局部水动力扰动影响显著。中尺度内孤立波主要出现在安达曼海深水区,尤其是在苏门答腊岛附近,常以波包形式存在,波长为数公里,波幅较大,波包内的主孤子常伴随次级孤子,波峰间距均匀且稳定。在安达曼海中部及其周边岛屿附近,大尺度内孤立波的分布较为显著,波长可达数十千米,通常表现为由多个孤子组成的波包,波形呈现弧形或条带状(见图1)。复杂的海域特征,如靠近岛屿或岸线的区域,可能引发内孤立波的折射或反射,进而增加其传播方向和形态的多样性,丰富了不同区域内孤立波的特征。因此,本文选取安达曼海作为研究区域,以探讨其内孤立波的特征。
哨兵一号(Sentinel-1)卫星系列包括两颗C波段SAR卫星,分别为Sentinel-1A和Sentinel-1B,分别于2014年4月和2016年4月发射。这些卫星提供多种成像模式,包括条带模式(Stripmap Mode)、干涉宽幅模式(Interferometric Wide Swath, IW Mode)、超宽幅模式(Extra Wide Swath Mode)以及波模式(Wave Mode),能够有效监测多尺度的海洋现象。Sentinel-1支持单极化和双极化操作方式,并运行在近极地太阳同步轨道,轨道高度约700 km,重访周期为12天。
Sentinel-1A/B的高精度数据为内孤立波研究提供了强有力的支持,特别是在特征变化分析及传播方向监测中表现出独特优势。这些数据在内孤立波频发区域(如南海和安达曼海)的时空分布研究中取得了重要成果[2,3]。此外,基于Sentinel-1数据的内孤立波检测与识别方法已取得显著进展 [12-16],验证了该数据源在内孤立波特征提取及应用中的可靠性和有效性。
本文选择使用Sentinel-1A/B卫星获取的干涉宽幅模式影像。该模式具备20 m的空间分辨率,使得沿岸区域的小尺度内孤立波得以清晰呈现,同时提供250 km的宽幅覆盖,既能覆盖大面积海域,又能捕捉内孤立波的整体结构及传播过程。详细参数如表1所示。高分辨率和宽覆盖范围相结合,能够有效识别波形的细节特征,为多尺度内孤立波特征提取提供了可靠的数据支撑。
本研究共收集了2015年至2019年期间含有内孤立波的影像,均为Sentinel-1A/B的IW模式下VV极化观测所得,具备20 m空间分辨率和250 km宽幅覆盖。高分辨率特性使得影像尤其适合小尺度内孤立波的识别和分析。影像数据经过预处理,构建用于内孤立波检测与分析的高质量数据集。
①从哥白尼数据空间生态系统(Copernicus Data Space Ecosystem)下载SAR影像数据;
②对原始图像进行预处理,包括辐射校准和几何校正,使用哨兵应用平台(SNAP)实现批处理;
③对影像进行1/3下采样,一定程度上减少了斑点噪声的影响,提高图像的视觉效果,可以提取更稳定和显著的内孤立波特征。同样,较小的图像尺寸可以缩短训练时间,加快模型的迭代和收敛速度。
④最后,通过LabelMe 软件对影像进行一对一的标签注释,生成PNG格式的标签数据。
通过上述预处理步骤,构建了适用于条纹分割的内孤立波数据集,最终构建了92幅包含内孤立波SAR遥感影像的样本库。同时本文采用数据增强技术提升模型精度、鲁棒性和泛化能力。SAR影像中内孤立波特征受到多种海洋环境与成像条件的影响,例如风速、水温差和斑点噪声等因素,这些因素导致内孤立波的纹理、亮度和形态呈现复杂多变的特征。
为了确保训练数据能够充分代表复杂海况条件,下载并分析了本文所用数据对应的ERA-5再分析风速数据。通过统计分析,发现所用数据的大部分风速范围在3 m/s到7 m/s之间,如图2所示。而使用SAR遥感影像观测内孤立波的最佳风速范围通常在3.5 m/s到6.5 m/s之间[17,18]。通过对风速数据的统计分析,确保数据集的多样性和代表性。虽然无法完全覆盖所有复杂海况条件,但考虑到不同风速范围和海况差异,训练数据能够有效反映多种海洋环境下的内孤立波特征。进一步采用数据增强技术可在训练过程中生成多样化的样本数据,使模型能够适应多样的内孤立波特征,从而提高复杂环境下的检测精度和模型的泛化能力。
为了提高SAR图像中内孤立波检测的精度和鲁棒性,本研究通过几何变化(如旋转、缩放和平移)调节亮度,添加噪声,来模拟不同环境条件下的内孤立波,扩大训练数据从而提升模型的鲁棒性。
几何变换:通过旋转、翻转、缩放等几何变化,模拟内孤立波在不同传播方向和尺度下的变化。
亮度和对比度:模拟SAR图像中由于不同散射强度引起的亮度和对比度变化,使模型能够适应多种海况和成像条件下的图像特征。
噪声增强:通过加入不同类型的噪声(如斑点噪声),增强模型在噪声干扰环境中的鲁棒性,从而提高模型的抗噪声能力。
通过结合数据增强的方法(如图3所示),使模型能够在更复杂的环境条件下学习到更丰富和多样的内孤立波特征。
卷积神经网络(CNN)在深度学习领域的语义分割中得到了广泛应用[19,20]。Swin-Unet [19]结合Swin Transformer [21]和传统卷积神经网络的优点,旨在实现高效且精确的图像语义分割。通过将Swin Transformer 引入传统的U-Net [22]架构,利用Swin Transformer 强大的特征学习能力及其基于窗口的自注意力机制。与传统卷积神经网络不同,Swin Transformer 能够有效捕捉长距离依赖关系,从而增强全局上下文建模能力。此外,Swin Transformer 的局部自注意力特点使其能够在不同尺度上捕捉信息,从而显著提升图像分割性能。Swin-Unet在医学图像、遥感图像以及其他高分辨率图像分割任务中已取得优异表现,尤其在需要精细分割的场景中展现出显著优势。因此,本文将其引入海洋内孤立波的特征提取任务中。
在海洋内孤立波特征提取研究中,SAR遥感影像是最常用的数据源。在SAR影像中,不同尺度的内孤立波具有显著差异。具体而言,小尺度内孤立波通常波长较短且振幅较低;中尺度内孤立波展现更明显的波包结构;而大尺度内孤立波通常以周期性波阵的形式出现,并具有显著的传播特性。这些不同尺度内孤立波在各自区域中展现出不同的特征,为深入研究内波的传播行为和动态特性提供了重要数据支撑。
而Swin Transformer 的引入可以较好满足小尺度、中尺度和大尺度内孤立波特征提取的需求。该模型不仅能够有效捕捉长距离依赖关系,还可以增强全局上下文建模能力。通过使用多层 Swin Transformer 的编码结构,Swin-Unet的分层设计在捕捉不同尺度内孤立波特征方面表现出色。具体而言,在模型编码阶段,Swin Transformer 的窗口注意力机制用于捕获局部区域的细节特征,同时其分层特性有助于提取中尺度和大尺度内孤立波的整体形态与传播趋势。在解码阶段,基于 U-Net的跳跃连接机制将编码阶段的多尺度特征逐步融合,从而恢复内孤立波的空间细节信息。小尺度内孤立波的细节特征通过 Swin Transformer 的局部窗口捕获,而中尺度和大尺度内孤立波的整体特征则通过分层注意力机制建模,实现全局与细节特征的有效交互。这种设计使模型在复杂海洋背景下能够显著提升多尺度内孤立波的分割能力。图4为本文构建的基于Swin-Unet的海洋内孤立波条纹特征提取模型示意图。
在SAR遥感影像中,内孤立波通常表现为亮暗相间的条纹特征,经过图像处理后,本文将背景区域标记为0(黑色),将内孤立波区域标记为1(白色),进而整个模型本质上就是一个二分类语义分割模型。然而,遥感图像覆盖的范围极其广泛,导致内孤立波在图像中的占比极小,这导致前景(内孤立波)与背景样本数量严重失衡,进而产生显著的样本不平衡问题。在处理样本不平衡问题时,尤其是在内孤立波的提取任务中,由于背景样本数量远多于内孤立波样本,仅使用模型本身自带的交叉熵损失函数可能导致模型偏向于预测背景,从而忽略内孤立波特征信息。为解决样本不平衡问题[23],本文采用Focal Loss和Dice Loss作为损失函数。
其中,pi是模型对第i类的预测概率,yi是实际类别的标签(1或0)。尽管交叉熵损失在许多分类任务中表现良好,而样本集中前景和背景样本严重不平衡的问题,可能导致模型偏向于预测频率较高的背景类别。为了解决这一问题,采用了Focal Loss和Dice Loss两种损失函数。Focal Loss由Lin等人在2017年提出,其公式如下:
其中,pt是模型对正样本的预测概率,αt是类别平衡因子,γ是调节因子。Focal Loss通过引入调节因子(1-pt)γ降低了易分类样本(背景类)对总损失的影响,从而增强模型对难分类样本(内孤立波)的关注,特别适用于处理样本类别严重不平衡的情况。
Dice Loss则是基于Dice系数的损失函数,广泛用于图像分割任务。其公式如下:
其中,A表示真实分割图像的像素标签,B表示模型预测分割图像的像素类别。Dice Loss通过直接优化前景和背景区域的重叠度,能够有效处理类别不平衡的问题,尤其是对小目标的分割效果更好。
通过结合Focal Loss和Dice Loss,模型在内孤立波提取过程中不仅能够更有效关注难以检测的内孤立波区域,还能通过优化前景和背景的重叠度来提升分割精度,从而缓解样本不平衡问题的影响。Focal Loss通过降低背景样本的权重,使模型更关注难分类的内孤立波区域,而Dice Loss则通过衡量前景和背景区域的重叠情况,进一步提升对内孤立波的分割精度。
深度学习模型的训练效果高度依赖于高质量的真值标签。然而,研究区域内缺乏精确的内孤立波波峰线数据产品,同时 SAR 卫星对内孤立波的观测易受其他海洋现象(如锋面、雨团等)以及成像特性的干扰,难以直接获得精确的波峰线信息。为确保标签的准确性,本研究采用目视解译的方法生成标注数据。依托Labelme 软件对遥感影像中的内孤立波区域进行标注。本研究与传统基于线条标注波峰线的方法不同,采用多边形标注方式,能够更全面地覆盖内孤立波的实际分布区域。此方法能够更准确地刻画内孤立波的形状、边界及实际面积,从而有效降低标注误差并提高数据质量。
本研究从92景经过预处理的遥感影像中提取内孤立波数据,并基于上述方法生成高质量的真值标签。通常情况下,遥感影像的分辨率极高,其尺寸通常超过 20 000×20 000像素。尽管本研究对原始影像进行了1/3下采样,但由于硬件性能和软件处理能力的限制,直接使用完整影像进行模型训练依然存在困难。因此,本文采用滑窗裁剪的方法,将影像及其对应的真值标签切分为512×512像素的小样本图像。该方法一方面能够尽可能保留内孤立波的特征信息,另一方面也更契合深度学习模型的输入需求。考虑到大尺度内孤立波的波峰线较长,裁剪后的单个样本可能无法完整覆盖内孤立波条纹区域,本文在裁剪过程中设计了128像素的重叠区域,以保证样本之间的特征连贯性和内孤立波区域的完整性。通过上述方法,最终生成了4 608对小样本图像,用于后续的模型训练和实验分析。
为确保模型评估的可靠性,数据集按照 8:2 的比例划分为训练集和验证集。训练集用于模型参数的学习,而验证集在整个训练过程中仅用于性能评估,并不参与模型训练,以确保其独立性。在实验中,验证集作为模型泛化能力的重要衡量标准,反映了模型在未见数据上的表现。尽管未单独设置测试集,本文通过严格划分数据,确保验证数据与训练数据相互独立,从而提高评估结果的可信度。
网络的训练和测试是基于Python平台使用Py-Torch框架进行编码[24],并在1张NVIDIA GeForce RTX 3080S GPU的工作站上实现。使用SGD(随机梯度下降)作为优化器,初始学习率为0.01,batch大小为8,每个实验的epoch设置为200。使用精确率作为模型结果好坏的评判标准,精确率(Precision)可以衡量模型预测为内孤立波的像素中实际为内孤立波的比例,能够有效地评估模型语义分割结果的效果。
TTP:即正样本被正确识别的数量。FFP:即误报的负样本数量。在深度学习模型分割任务中,模型的性能评估至关重要,通常依靠多种评价指标以全面衡量网络的提取效果。常用且重要的指标包括召回率、F1分数(F1score)和平均交并比(MIoU),本研究将使用这些指标来评估结果,并与其他语义分割模型进行比较。
TTN:即负样本被正确识别的数量。FFN:即漏报的正样本数量。其中召回率用来衡量模型对实际内孤立波的检测能力;F1score综合了精确率和召回率,是这两者的调和平均数,适用于样本不平衡的情况;MIoU则量化了预测区域和真实内孤立波区域的重叠程度,MIoU越高,表示分割精度越佳,总体而言,这三项指标相互补充,为全面评估模型在内孤立波提取任务中的表现提供了有力依据,验证了模型的有效性。
为了进一步探究模型在内孤立波提取任务中的表现,本文选择U-Net语义分割模型和本文所用模型进行对比。以下是对比各个模型在上述关键指标上的表现。
表2详细分析了U-Net和Swin-Unet在不同尺度内孤立波分割任务中的性能差异,基于F1分数、精准率、召回率和平均交并比等评价指标,对比了两个模型在不同尺度内孤立波图像中的表现,并且改进后的模型平均交并比达到了62.47%,显著高于U-Net。同时本文所用模型在小尺度和多尺度内孤立波图像上展现了显著优势。如图5所示。在多尺度内孤立波检测中,即使在复杂背景和噪声干扰下,也能准确分割出内孤立波条纹。
根据图6可以看出,Swin-Unet在捕捉多尺度特征方面展现出更强的适应性,能够准确区分不同尺度的内孤立波特征。通过与U-Net模型相比能够明显看出Swin-Unet在细节的捕捉上具有显著优势,它不仅能够处理细微波纹,还能在复杂背景和不同海况下,准确识别和分割出原本难以识别的内孤立波特征,从而提升了整体检测的有效性和精度。
基于Swin-Unet模型的内孤立波提取算法能够自动处理包含不同尺度内孤立波SAR图像,本文总共选择了10张SAR图像进行测试,以验证模型的有效性。
图7展示了2017年4月12日拍摄于安达曼海南部海域的SAR图像。该图像包含两组内孤立波,并表现出不同尺度的特征。从检测结果可以看出,提出的模型能够准确识别这两组波包。
图8为2024年2月24日拍摄于安达曼海中部海域的SAR图像。该图像中包含多条内孤立波,形成了两组波包。检测结果显示,模型成功分割了宽广的内孤立波条纹,并保持条纹的完整性。此外,在图像右边区域,尽管内孤立波的特征与背景信息对比度较低,模型依然能够精准提取其特征。
图9显示了2018年8月3日拍摄于南海海南岛附近的SAR图像。图像中可见两个交错的波包向西北方向传播,其中自东向西传播的波包较为清晰。模型成功提取了该区域的内孤立波特征信息。在图像中部区域,由于内孤立波的尺度较小,其边缘与背景海水高度相似,SAR图像中的条纹特征并不明显,形态结构模糊。然而,模型仍然能够有效检测该区域的内孤立波特征,进一步验证了模型的泛化能力。
为了进一步验证模型对不同SAR卫星数据的适应性,本研究选取GF-3影像(分辨率10 m),通过GF-3遥感图像测试评估模型性能。
图10展示了GF-3 SAR图像及其检测结果。从图像中可以清晰地看到内孤立波的条纹特征,模型成功提取了该区域内孤立波的特征信息。此外,模型还避免了将左侧的雨团误检为内孤立波,进一步验证了其在复杂场景下的泛化能力和鲁棒性。
为评估Swin-Unet模型在SAR影像中检测内孤立波的能力,本文选择安达曼海域的连续 SAR 数据进行测试。总体上,模型在整轨SAR影像中表现出较强的检测能力。
图11显示了2019年3月23日采集的安达曼海SAR影像及其检测结果。该影像左下区域存在显著的背景噪声干扰,导致条纹特征模糊且部分被噪声覆盖,即便如此,模型在影像左侧区域仍能准确检测内孤立波条纹结构,并在背景复杂区域保持了较高的检测能力。而在图像的左上区域,存在着不同尺度的内孤立波,不同尺度的内孤立波交错分布,但本模型依然有着较高的检测精度。
图12展示了2019年8月2日采集的另一幅SAR 影像,影像中包含朝两个方向传播的内孤立波,波包间的相互作用形成了复杂的条纹结构,并包含多个波包。尽管如此,模型依然能够准确分割波峰线,且在条纹形态发生变化时保持了较高的检测精度和条纹连贯性。此外,在靠近复杂地形(如海岸线)区域时,模型成功抑制了背景干扰,提取出较为清晰的内孤立波特征。
在处理整轨SAR影像时,模型能够有效检测并分割内孤立波条纹,且在不同影像间保持了条纹特征的连续性和一致性。
本研究收集了2015年至2019年期间覆盖安达曼海的92景哨兵一号SAR影像,通过预处理及目视解译生成了4 608对样本数据,构建了内孤立波数据集。在此基础上,基于改进的Swin-Unet网络,实现了SAR影像中内孤立波特征的自动提取,并对模型在安达曼海区域的不同尺度内孤立波特征提取效果进行了分析。尽管模型整体表现优异,但在部分检测结果中,由于雨团、锋线等海洋现象背景干扰、数据不平衡等原因,导致出现漏检和虚警目标的现象。针对这些问题,可以通过扩展数据样本、增加模型的复杂程度以及引入更强的背景抑制机制来进一步提升模型性能。本研究成果为SAR影像中内孤立波的自动化检测与分析提供了新方法,并为后续研究奠定了基础。
  • 国家自然科学基金(低空远距离条件下海态与目标一体化雷达探测理论与方法U26207)
参考文献 引证文献
排序方式:
[1]
HAURY L R, BRISCOE M G, ORR M H. Tidally generated internal wave packets in Massachusetts Bay[J]. Nature, 1979, 278(5702): 312-317.
[2]
DONG D, YANG X, LI X, et al. SAR observation of eddy-induced mode-2 internal solitary waves in the south china sea[J]. IEEE Transactions on Geoscience and Remote Sensing, 2016, 54(11): 6674-6686.
[3]
RODENAS J A, GARELLO R. Wavelet analysis in SAR ocean image profiles for internal wave detection and wavelength estimation[J]. IEEE Transactions on Geoscience and Remote Sensing, 1997, 35(4): 933-945.
[4]
RODENAS J A, GARELLO R. Internal wave detection and location in SAR images using wavelet transform[J].IEEE Transactions on Geoscience and Remote Sensing,1998, 36(5): 1494-1507
[5]
DOKKEN S T, OLSEN R, WAHL T, et al. Identification and characterization of internal waves in SAR images along the coast of Norway[J]. Geophysical Research Letters, 2001, 28(14): 2803-2806.
[6]
SURAMPUDI S, SASANKA S. Internal wave detection and characterization with SAR data[C]//2019 IEEE Recent Advances in Geoscience and Remote Sensing: Technologies, Standards and Applications (TENGARSS). Kochi, Kerala, India: IEEE, 2019: 104-108.
[7]
SIMONIN D, TATNALL A R, ROBINSON I S. The auto-mated detection and recognition of internal waves[J]. International Journal of Remote Sensing, 2009, 30(17):4581-4598.
[8]
MARGHANY M. Automatic detection of internal wave using particle swarm optimization algorithm[J]. IOP Conference Series: Earth and Environmental Science, 2018,169: 012049.
[9]
KUREKIN A A, LAND P E, MILLER P I. Internal waves at the uk continental shelf: Automatic mapping using the ENVISAT ASAR sensor[J]. Remote Sensing, 2020, 12(15): 2476.
[10]
LI X, LIU B, ZHENG G, et al. Deep-learning-based information mining from ocean remote-sensing imagery[J]. National Science Review, 2020, 7(10): 1584-1605.
[11]
BAO S, MENG J, SUN L, et al. Detection of ocean internal waves based on Faster R-CNN in SAR images[J]. Journal of Oceanology and Limnology, 2020, 38(1): 55-63.
[12]
TAO M, XU C, GUO L, et al. An internal waves data set from sentinel-1 synthetic aperture radar imagery and preliminary detection[J]. Earth and Space Science, 2022, 9(12): e2022EA002528.
[13]
ZHENG Y G, ZHANG H S, WANG Y Q. Stripe detection and recognition of oceanic internal waves from synthetic aperture radar based on support vector machine and feature fusion[J]. International Journal of Remote Sensing, 2021, 42(17): 6706-6724.
[14]
ZHENG Y G, ZHANG H S, QI K T, et al. Stripe segmentation of oceanic internal waves in SAR images based on SegNet[J]. Geocarto International, 2022, 37(25): 8567-8578.
[15]
MA Y, MENG J, SUN L, et al. Oceanic internal wave signature extraction in the sulu sea by a pixel attention U-Net: PAU-Net[J]. IEEE Geoscience and Remote Sensing Letters, 2023, 20: 1-5.
[16]
ZHANG S, LI X, ZHANG X. Internal wave signature extraction from sar and optical satellite imagery based on deep learning[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 1-16.
[17]
VESECKY J F, STEWART R H. The observation of ocean surface phenomena using imagery from the SEASAT synthetic aperture radar: An assessment[J]. Journal of Geophysical Research: Oceans, 1982, 87(C5): 3397-3430.
[18]
KASISCHKE E S, LYZENGA D R, SHUCHMAN R A,et al. Contrast ratios of internal waves in synthetic aperture radar imagery: A comparison of SAR internal wave signature experiment observations with theory[J]. Journal of Geophysical Research: Oceans, 1988, 93(C10):12355-12369.
[19]
KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6): 84-90.
[20]
CAO H, WANG Y, CHEN J, et al. Swin-UNet: Unet-like pure transformer for medical image segmentation[EB/OL]. arXiv:2021. (2021-05-12)[2024-03-20]. https://arxiv.org/abs/2105.05537.
[21]
LIU Z, LIN Y, CAO Y, et al. Swin transformer: Hierarchical vision transformer using shifted windows[EB/OL]. arXiv:2021. (2021-03-25)[2024-03-20]. https://arxiv.org/abs/2103.14030.
[22]
RONNEBERGER O, FISCHER P, BROX T. U-net:Convolutional networks for biomedical image segmentation[C]//NAVAB N, HORNEGGER J, WELLS W M, et al. Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015: 18th International Conference, October 5-9, 2015, Munich, Germany. Cham:Springer International Publishing, 2015: 234-241.
[23]
GOUTTE C, GAUSSIER E. A probabilistic interpretation of precision, recall and f-score, with implication for evaluation[C]//LOSADA D E, FERNÁNDEZ-LUNA J M. Advances in Information Retrieval: 27th European Conference on IR Research, ECIR 2005, March 21-23,2005, Santiago de Compostela, Spain. Berlin: Springer,2005: 345-359.
[24]
PASZKE A, GROSS S, MASSA F, et al. PyTorch: An imperative style, high-performance deep learning library[C]//33rd Conference on Neural Information Processing Systems (NeurIPS 2019), December 8-14, 2019, Vancouver, Canada. New York: Curran Associates, 2019:8024-8035. arXiv:
2025年第46卷第4期
PDF下载
116
51
引用本文
BibTeX
文章信息
doi: 10.12347/j.ycyk.20250115002
  • 接收时间:2025-01-15
  • 首发时间:2026-03-13
  • 出版时间:2025-07-15
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2025-01-15
  • 修回日期:2025-03-08
基金
国家自然科学基金(低空远距离条件下海态与目标一体化雷达探测理论与方法U26207)
作者信息
    1自然资源部第一海洋研究所 青岛 266061
    2中国石油大学(华东)海洋与空间信息学院 青岛 266580
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/ycyk/CN/10.12347/j.ycyk.20250115002
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