SPLASH-SegFormer Pipeline: A Transformer-Based Approach for High-Resolution and Low-Cost Laser Scanner Seafloor Mapping

High-resolution seafloor mapping continues to be challenging, primarily due to the high costs and complexity of traditional sensors. Laser scanners offer a more affordable alternative, using a monocular camera and a laser stripe. While this method provides high-resolution 3D reconstructions, it is s...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	IEEE robotics and automation letters Jg. 10; H. 8; S. 7995 - 8002
Hauptverfasser:	Fuentes-Guinez, Javiera, Troni, Giancarlo, Lobel, Hans
Format:	Journal Article
Sprache:	Englisch
Veröffentlicht:	Piscataway IEEE 01.08.2025 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:	Accuracy Cameras High resolution Image reconstruction Image segmentation Laser modes Lasers Lighting Mapping Marine robotics object detection Ocean floor Pipelines Real time Scanners Sea floor segmentation and categorization Sensors Three dimensional models Three-dimensional displays Transformers Underwater
ISSN:	2377-3766, 2377-3766
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Abstract	High-resolution seafloor mapping continues to be challenging, primarily due to the high costs and complexity of traditional sensors. Laser scanners offer a more affordable alternative, using a monocular camera and a laser stripe. While this method provides high-resolution 3D reconstructions, it is sensitive to underwater lighting and surface reflections. This work introduces SPLASH-SegFormer Pipeline, a novel method that leverages attention mechanisms and multi-level feature integration to enhance laser stripe segmentation in underwater environments, marking, to the best of our knowledge, the first application of a transformer-based architecture for this task. The proposed method is trained and evaluated using both simulated and field experimental data collected in a test tank and during multiple sea expeditions, allowing it to learn from diverse and challenging environments. A complete pipeline is constructed around this network, enabling the processing of images from a camera to a 3D point cloud reconstruction quickly and accurately. The results demonstrate that SPLASH-SegFormer significantly enhances segmentation robustness and accuracy while maintaining a reasonable speed. It outperforms traditional underwater thresholding by over 64 points in mIoU and surpasses all evaluated methods, achieving 6.6 points higher mIoU than Stripe Segmentor while running over 65% faster. This speed enhances efficiency and makes it ideal for real-time onboard applications, enabling millimeter-level underwater 3D reconstruction and high-resolution seafloor mapping using smaller, cost-effective sensors.
AbstractList	High-resolution seafloor mapping continues to be challenging, primarily due to the high costs and complexity of traditional sensors. Laser scanners offer a more affordable alternative, using a monocular camera and a laser stripe. While this method provides high-resolution 3D reconstructions, it is sensitive to underwater lighting and surface reflections. This work introduces SPLASH-SegFormer Pipeline, a novel method that leverages attention mechanisms and multi-level feature integration to enhance laser stripe segmentation in underwater environments, marking, to the best of our knowledge, the first application of a transformer-based architecture for this task. The proposed method is trained and evaluated using both simulated and field experimental data collected in a test tank and during multiple sea expeditions, allowing it to learn from diverse and challenging environments. A complete pipeline is constructed around this network, enabling the processing of images from a camera to a 3D point cloud reconstruction quickly and accurately. The results demonstrate that SPLASH-SegFormer significantly enhances segmentation robustness and accuracy while maintaining a reasonable speed. It outperforms traditional underwater thresholding by over 64 points in mIoU and surpasses all evaluated methods, achieving 6.6 points higher mIoU than Stripe Segmentor while running over 65% faster. This speed enhances efficiency and makes it ideal for real-time onboard applications, enabling millimeter-level underwater 3D reconstruction and high-resolution seafloor mapping using smaller, cost-effective sensors.
Author	Fuentes-Guinez, Javiera Lobel, Hans Troni, Giancarlo
Author_xml	– sequence: 1 givenname: Javiera orcidid: 0009-0001-2139-1448 surname: Fuentes-Guinez fullname: Fuentes-Guinez, Javiera email: jfuentes@uc.cl organization: Pontificia Universidad Catolica de Chile, Santiago, Chile – sequence: 2 givenname: Giancarlo orcidid: 0000-0002-1517-8739 surname: Troni fullname: Troni, Giancarlo email: giancarlo.troni@gmail.com organization: Monterey Bay Aquarium Research Institute, Moss Landing, CA, USA – sequence: 3 givenname: Hans orcidid: 0000-0003-3514-9414 surname: Lobel fullname: Lobel, Hans email: halobel@uc.cl organization: Pontificia Universidad Catolica de Chile, Santiago, Chile
BookMark	eNpNkM1PwkAQxTcGExG5e_CwiefF_eh2qbdKRExqJBTPzdJOoaTs1l0I8b93ERM9zeTlvTeT3zXqGWsAoVtGR4zR5CFbpCNOuRwJqZTk9AL1uVCKCBXHvX_7FRp6v6WUMsmVSGQfHfN5luYzksN6at0OHJ43HbSNgUec4qXTxtc_OnnSHiqcdp2zutzgoOJZs96QBXjbHvaNNVibCmf2SCbW73EW_A7npTbmNEHXrQ2ZN911jVnfoMtatx6Gv3OAPqbPy8mMZO8vr5M0IyWP1J4oASvJa0YroVUUi6os5ZjRUsdKqkoAJFTVOhpzJmFMJV-tWMwEFVLXiQTBxQDdn3vD258H8Ptiaw_OhJOF4DyKVSKlDC56dpXOeu-gLjrX7LT7KhgtToSLQLg4ES5-CYfI3TnSAMCfnVGuotD4DYeyd60
CODEN	IRALC6
Cites_doi	10.1016/j.measurement.2022.110837 10.3390/s24113554 10.5194/isprs-archives-xlii-2-w18-13-2019 10.1109/OCEANSSYD.2010.5603672 10.1016/j.imavis.2004.07.006 10.3389/fmars.2019.00283 10.1109/AUV.2010.5779655 10.1109/IROS.2008.4650627 10.1109/CVPR52729.2023.01215 10.3390/s16010013 10.1109/ICIIBMS60103.2023.10347750 10.1109/34.888718 10.1007/978-3-319-24574-4_28 10.7764/tesisuc/ing/73465 10.3390/s151229864 10.1016/j.optlaseng.2024.108129 10.1109/34.659930 10.1002/rob.21682 10.3390/photonics11080723 10.1109/TSMC.1979.4310076 10.1007/s00138-010-0288-6 10.1016/j.optlaseng.2024.108468 10.1109/ICMA54519.2022.9855905 10.3390/rs15071864 10.1109/34.121791 10.1088/1361-6501/ace645 10.1109/CVPR.2018.00913
ContentType	Journal Article
Copyright	Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2025
Copyright_xml	– notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2025
DBID	97E ESBDL RIA RIE AAYXX CITATION 7SC 7SP 8FD JQ2 L7M L~C L~D
DOI	10.1109/LRA.2025.3577520
DatabaseName	IEEE Xplore (IEEE) IEEE Xplore Open Access Journals IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE/IET Electronic Library (IEL) CrossRef Computer and Information Systems Abstracts Electronics & Communications Abstracts Technology Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional
DatabaseTitle	CrossRef Technology Research Database Computer and Information Systems Abstracts – Academic Electronics & Communications Abstracts ProQuest Computer Science Collection Computer and Information Systems Abstracts Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Professional
DatabaseTitleList	Technology Research Database
Database_xml	– sequence: 1 dbid: RIE name: IEEE/IET Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher
DeliveryMethod	fulltext_linktorsrc
Discipline	Engineering
EISSN	2377-3766
EndPage	8002
ExternalDocumentID	10_1109_LRA_2025_3577520 11027455
Genre	orig-research
GrantInformation_xml	– fundername: David and Lucile Packard Foundation funderid: 10.13039/100000008 – fundername: ANID BECAS/MAGISTER NACIONAL grantid: 22231821
GroupedDBID	0R~ 97E AAJGR AARMG AASAJ AAWTH ABAZT ABQJQ ABVLG ACGFS AGQYO AGSQL AHBIQ AKJIK AKQYR ALMA_UNASSIGNED_HOLDINGS ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ EBS EJD ESBDL IFIPE IPLJI JAVBF KQ8 M43 M~E O9- OCL RIA RIE AAYXX CITATION 7SC 7SP 8FD JQ2 L7M L~C L~D
ID	FETCH-LOGICAL-c247t-73eb52f10d3a7463dcc5810ca6757d3ee907fa48215e8052bb1613035af95e323
IEDL.DBID	RIE
ISICitedReferencesCount	0
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001518757500003&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	2377-3766
IngestDate	Sat Nov 22 13:40:48 EST 2025 Sat Nov 29 07:50:37 EST 2025 Wed Aug 27 01:42:25 EDT 2025
IsDoiOpenAccess	true
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Issue	8
Language	English
License	https://creativecommons.org/licenses/by-nc-nd/4.0
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c247t-73eb52f10d3a7463dcc5810ca6757d3ee907fa48215e8052bb1613035af95e323
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ORCID	0000-0002-1517-8739 0000-0003-3514-9414 0009-0001-2139-1448
OpenAccessLink	https://ieeexplore.ieee.org/document/11027455
PQID	3224679555
PQPubID	4437225
PageCount	8
ParticipantIDs	ieee_primary_11027455 proquest_journals_3224679555 crossref_primary_10_1109_LRA_2025_3577520
PublicationCentury	2000
PublicationDate	2025-08-01
PublicationDateYYYYMMDD	2025-08-01
PublicationDate_xml	– month: 08 year: 2025 text: 2025-08-01 day: 01
PublicationDecade	2020
PublicationPlace	Piscataway
PublicationPlace_xml	– name: Piscataway
PublicationTitle	IEEE robotics and automation letters
PublicationTitleAbbrev	LRA
PublicationYear	2025
Publisher	IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Publisher_xml	– name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
References	ref13 ref12 ref15 ref14 ref31 ref30 ref11 ref10 (ref25) 2024 ref2 ref1 ref17 ref16 ref19 Trnblom (ref9) 2010 ref23 ref26 (ref24) 2024 ref20 Zhou (ref32) 2018 ref22 ref21 ref28 ref27 ref29 ref8 ref7 Xie (ref18) 2021 ref4 ref3 ref6 ref5
References_xml	– ident: ref16 doi: 10.1016/j.measurement.2022.110837 – ident: ref6 doi: 10.3390/s24113554 – start-page: 12077 volume-title: Proc. Adv. Neural Inf. Process. Syst. year: 2021 ident: ref18 article-title: SegFormer: Simple and efficient design for semantic segmentation with transformers – ident: ref11 doi: 10.5194/isprs-archives-xlii-2-w18-13-2019 – ident: ref14 doi: 10.1109/OCEANSSYD.2010.5603672 – ident: ref21 doi: 10.1016/j.imavis.2004.07.006 – ident: ref1 doi: 10.3389/fmars.2019.00283 – ident: ref5 doi: 10.1109/AUV.2010.5779655 – year: 2010 ident: ref9 article-title: Underwater 3D surface scanning using structured light – year: 2024 ident: ref25 article-title: Roboflow: Organize, label, & create datasets – ident: ref4 doi: 10.1109/IROS.2008.4650627 – ident: ref26 doi: 10.1109/CVPR52729.2023.01215 – year: 2024 ident: ref24 article-title: Schmidt Ocean Institute ROV Subastian – ident: ref8 doi: 10.3390/s16010013 – ident: ref22 doi: 10.1109/ICIIBMS60103.2023.10347750 – ident: ref20 doi: 10.1109/34.888718 – ident: ref28 doi: 10.1007/978-3-319-24574-4_28 – ident: ref19 doi: 10.7764/tesisuc/ing/73465 – ident: ref7 doi: 10.3390/s151229864 – ident: ref17 doi: 10.1016/j.optlaseng.2024.108129 – ident: ref29 doi: 10.1109/34.659930 – ident: ref13 doi: 10.1002/rob.21682 – ident: ref30 doi: 10.3390/photonics11080723 – year: 2018 ident: ref32 article-title: Open3D: A modern library for 3D data processing – ident: ref27 doi: 10.1109/TSMC.1979.4310076 – ident: ref15 doi: 10.1007/s00138-010-0288-6 – ident: ref12 doi: 10.1016/j.optlaseng.2024.108468 – ident: ref3 doi: 10.1109/ICMA54519.2022.9855905 – ident: ref10 doi: 10.3390/rs15071864 – ident: ref31 doi: 10.1109/34.121791 – ident: ref2 doi: 10.1088/1361-6501/ace645 – ident: ref23 doi: 10.1109/CVPR.2018.00913
SSID	ssj0001527395
Score	2.2987597
Snippet	High-resolution seafloor mapping continues to be challenging, primarily due to the high costs and complexity of traditional sensors. Laser scanners offer a...
SourceID	proquest crossref ieee
SourceType	Aggregation Database Index Database Publisher
StartPage	7995
SubjectTerms	Accuracy Cameras High resolution Image reconstruction Image segmentation Laser modes Lasers Lighting Mapping Marine robotics object detection Ocean floor Pipelines Real time Scanners Sea floor segmentation and categorization Sensors Three dimensional models Three-dimensional displays Transformers Underwater
Title	SPLASH-SegFormer Pipeline: A Transformer-Based Approach for High-Resolution and Low-Cost Laser Scanner Seafloor Mapping
URI	https://ieeexplore.ieee.org/document/11027455 https://www.proquest.com/docview/3224679555
Volume	10
WOSCitedRecordID	wos001518757500003&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVIEE databaseName: IEEE/IET Electronic Library (IEL) customDbUrl: eissn: 2377-3766 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0001527395 issn: 2377-3766 databaseCode: RIE dateStart: 20160101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE – providerCode: PRVHPJ databaseName: ROAD: Directory of Open Access Scholarly Resources (ISSN International Center) customDbUrl: eissn: 2377-3766 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0001527395 issn: 2377-3766 databaseCode: M~E dateStart: 20160101 isFulltext: true titleUrlDefault: https://road.issn.org providerName: ISSN International Centre
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3NS-QwFH_syB7Wg-66iuMXOXjxEO00TdN4G8XBwyjiuOCtpMkbEaQdOqPe_Nt9r-2gsHjwUFpCUsp7Sd_vfQMcRmkoQopeRo52cOLopNsssTJBEnaFooGibTZhrq-z-3t70yWrN7kwiNgEn-ExPza-_FD5ZzaVnZCoIiVK6x70jEnbZK0PgwqXErN66YqM7Mn4dkgKYKyPlTZGc0fvT6Kn6aXy3w-4kSqj9W9-z29Y6-CjGLb8_gM_sNyA1U9FBf_CK4eDTC7lBB9GBEixFjePM846x1MxFHdLoIq1PCMJFsSwqyouaFRw2Idkk367IYUrgxhXr_K8mi_EmObXYkLMKPmObvpU0ZorxzUeHjbh3-ji7vxSdu0VpI8Ts5BGYaHj6SAKypkkVcF7nQ0i70iHMEEhkt48dUlGoAC58UFRDFjZUNpNrUYVqy1YKasSt0EY47UOdBGeSAbOZWhsGtjlSngoVr4PR0vK57O2ikbeaB-RzYlLOXMp77jUh02m9Me8jsh92FvyKu_O2TxXXA_PWK31zhfLduEXv72N2duDlUX9jPvw078sHuf1AfSu3i4Omo30DjrNxR8
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3Na9swFH907WDbod1Hx7J2mw697KDWsaTI2i0rCylzQ2ky6M3I0kspFLs46frv7z3boYWxww7GRkjYvCf5_d43wFEyimUcYZCJpx2sPZ10l2knNZKwKxUNlF2zCTubZVdX7qJPVm9zYRCxDT7DY35sffmxDvdsKjshUUVKlDHPYMdonSZdutajSYWLiTmzcUYm7iS_HJMKmJpjZaw13NP7ifBpu6n89Qtu5cpk7z-_6DXs9gBSjDuOv4EtrN7CqydlBd_BAweEzKdyjtcTgqTYiIubO847x29iLBYbqIqN_E4yLIpxX1dc0KjgwA_JRv1uSwpfRZHXD_K0Xq1FTvMbMSd2VHxHv7ytac255yoP1_vwa_JjcTqVfYMFGVJt19IqLE26HCZReatHKoZgsmESPGkRNipE0pyXXmcEC5BbH5TlkNUNZfzSGVSpeg_bVV3hBxDWBmMiXYQo9ND7DK0bRXa6EiJKVRjA1w3li7uujkbR6h-JK4hLBXOp6Lk0gH2m9OO8nsgDONzwquhP2qpQXBHPOmPMx38s-wIvpovzvMjPZj8P4CW_qYvgO4TtdXOPn-B5-L2-WTWf2-30B_rnxzU
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=SPLASH-SegFormer+Pipeline%3A+A+Transformer-Based+Approach+for+High-Resolution+and+Low-Cost+Laser+Scanner+Seafloor+Mapping&rft.jtitle=IEEE+robotics+and+automation+letters&rft.au=Fuentes-Gu%C3%AD%C3%B1ez%2C+Javiera&rft.au=Troni%2C+Giancarlo&rft.au=Lobel%2C+Hans&rft.date=2025-08-01&rft.issn=2377-3766&rft.eissn=2377-3766&rft.volume=10&rft.issue=8&rft.spage=7995&rft.epage=8002&rft_id=info:doi/10.1109%2FLRA.2025.3577520&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_LRA_2025_3577520
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2377-3766&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2377-3766&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2377-3766&client=summon