BenthicNet: A global compilation of seafloor images for deep learning applications
Abstract Advances in underwater imaging enable collection of extensive seafloor image datasets necessary for monitoring important benthic ecosystems. The ability to collect seafloor imagery has outpaced our capacity to analyze it, hindering mobilization of this crucial environmental information. Mac...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-02-01
|
| Series: | Scientific Data |
| Online Access: | https://doi.org/10.1038/s41597-025-04491-1 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823863319181656064 |
|---|---|
| author | Scott C. Lowe Benjamin Misiuk Isaac Xu Shakhboz Abdulazizov Amit R. Baroi Alex C. Bastos Merlin Best Vicki Ferrini Ariell Friedman Deborah Hart Ove Hoegh-Guldberg Daniel Ierodiaconou Julia Mackin-McLaughlin Kathryn Markey Pedro S. Menandro Jacquomo Monk Shreya Nemani John O’Brien Elizabeth Oh Luba Y. Reshitnyk Katleen Robert Chris M. Roelfsema Jessica A. Sameoto Alexandre C. G. Schimel Jordan A. Thomson Brittany R. Wilson Melisa C. Wong Craig J. Brown Thomas Trappenberg |
| author_facet | Scott C. Lowe Benjamin Misiuk Isaac Xu Shakhboz Abdulazizov Amit R. Baroi Alex C. Bastos Merlin Best Vicki Ferrini Ariell Friedman Deborah Hart Ove Hoegh-Guldberg Daniel Ierodiaconou Julia Mackin-McLaughlin Kathryn Markey Pedro S. Menandro Jacquomo Monk Shreya Nemani John O’Brien Elizabeth Oh Luba Y. Reshitnyk Katleen Robert Chris M. Roelfsema Jessica A. Sameoto Alexandre C. G. Schimel Jordan A. Thomson Brittany R. Wilson Melisa C. Wong Craig J. Brown Thomas Trappenberg |
| author_sort | Scott C. Lowe |
| collection | DOAJ |
| description | Abstract Advances in underwater imaging enable collection of extensive seafloor image datasets necessary for monitoring important benthic ecosystems. The ability to collect seafloor imagery has outpaced our capacity to analyze it, hindering mobilization of this crucial environmental information. Machine learning approaches provide opportunities to increase the efficiency with which seafloor imagery is analyzed, yet large and consistent datasets to support development of such approaches are scarce. Here we present BenthicNet: a global compilation of seafloor imagery designed to support the training and evaluation of large-scale image recognition models. An initial set of over 11.4 million images was collected and curated to represent a diversity of seafloor environments using a representative subset of 1.3 million images. These are accompanied by 3.1 million annotations translated to the CATAMI scheme, which span 190,000 of the images. A large deep learning model was trained on this compilation and preliminary results suggest it has utility for automating large and small-scale image analysis tasks. The compilation and model are made openly available for reuse. |
| format | Article |
| id | doaj-art-ed7567acc40a459eb6df130b64e6f979 |
| institution | Kabale University |
| issn | 2052-4463 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Data |
| spelling | doaj-art-ed7567acc40a459eb6df130b64e6f9792025-02-09T12:11:37ZengNature PortfolioScientific Data2052-44632025-02-0112112410.1038/s41597-025-04491-1BenthicNet: A global compilation of seafloor images for deep learning applicationsScott C. Lowe0Benjamin Misiuk1Isaac Xu2Shakhboz Abdulazizov3Amit R. Baroi4Alex C. Bastos5Merlin Best6Vicki Ferrini7Ariell Friedman8Deborah Hart9Ove Hoegh-Guldberg10Daniel Ierodiaconou11Julia Mackin-McLaughlin12Kathryn Markey13Pedro S. Menandro14Jacquomo Monk15Shreya Nemani16John O’Brien17Elizabeth Oh18Luba Y. Reshitnyk19Katleen Robert20Chris M. Roelfsema21Jessica A. Sameoto22Alexandre C. G. Schimel23Jordan A. Thomson24Brittany R. Wilson25Melisa C. Wong26Craig J. Brown27Thomas Trappenberg28Vector InstituteMemorial University of Newfoundland, Department of Geography, St. John’sDalhousie University, Faculty of Computer ScienceDalhousie University, Faculty of Computer ScienceDalhousie University, School for Resource and Environmental StudiesUniversidade Federal do Espírito Santo, Departamento de Oceanografia e EcologiaFisheries and Oceans Canada, Marine Spatial Ecology and Analysis Section, Institute of Ocean SciencesColumbia University, Lamont-Doherty Earth Observatory, PalisadesUniversity of Sydney, Australian Centre for Field RoboticsNational Oceanic and Atmospheric Administration Northeast Fisheries Science CenterUniversity of Queensland, School of the EnvironmentDeakin University, School of Life and Environmental SciencesOxy Occidental College, Vantuna Research GroupUniversity of Queensland, School of the EnvironmentUniversidade Federal do Espírito Santo, Departamento de Oceanografia e EcologiaUniversity of Tasmania, Institute for Marine and Antarctic StudiesFisheries and Marine Institute of Memorial University of Newfoundland, School of Ocean Technology, St. John’sFisheries and Oceans Canada, Bedford Institute of OceanographyUniversity of Tasmania, Institute for Marine and Antarctic StudiesHakai InstituteFisheries and Marine Institute of Memorial University of Newfoundland, School of Ocean Technology, St. John’sUniversity of Queensland, School of the EnvironmentFisheries and Oceans Canada, Bedford Institute of OceanographyGeological Survey of Norway (NGU)Ecology Action CentreFisheries and Oceans Canada, Bedford Institute of OceanographyFisheries and Oceans Canada, Bedford Institute of OceanographyDalhousie University, Department of OceanographyDalhousie University, Faculty of Computer ScienceAbstract Advances in underwater imaging enable collection of extensive seafloor image datasets necessary for monitoring important benthic ecosystems. The ability to collect seafloor imagery has outpaced our capacity to analyze it, hindering mobilization of this crucial environmental information. Machine learning approaches provide opportunities to increase the efficiency with which seafloor imagery is analyzed, yet large and consistent datasets to support development of such approaches are scarce. Here we present BenthicNet: a global compilation of seafloor imagery designed to support the training and evaluation of large-scale image recognition models. An initial set of over 11.4 million images was collected and curated to represent a diversity of seafloor environments using a representative subset of 1.3 million images. These are accompanied by 3.1 million annotations translated to the CATAMI scheme, which span 190,000 of the images. A large deep learning model was trained on this compilation and preliminary results suggest it has utility for automating large and small-scale image analysis tasks. The compilation and model are made openly available for reuse.https://doi.org/10.1038/s41597-025-04491-1 |
| spellingShingle | Scott C. Lowe Benjamin Misiuk Isaac Xu Shakhboz Abdulazizov Amit R. Baroi Alex C. Bastos Merlin Best Vicki Ferrini Ariell Friedman Deborah Hart Ove Hoegh-Guldberg Daniel Ierodiaconou Julia Mackin-McLaughlin Kathryn Markey Pedro S. Menandro Jacquomo Monk Shreya Nemani John O’Brien Elizabeth Oh Luba Y. Reshitnyk Katleen Robert Chris M. Roelfsema Jessica A. Sameoto Alexandre C. G. Schimel Jordan A. Thomson Brittany R. Wilson Melisa C. Wong Craig J. Brown Thomas Trappenberg BenthicNet: A global compilation of seafloor images for deep learning applications Scientific Data |
| title | BenthicNet: A global compilation of seafloor images for deep learning applications |
| title_full | BenthicNet: A global compilation of seafloor images for deep learning applications |
| title_fullStr | BenthicNet: A global compilation of seafloor images for deep learning applications |
| title_full_unstemmed | BenthicNet: A global compilation of seafloor images for deep learning applications |
| title_short | BenthicNet: A global compilation of seafloor images for deep learning applications |
| title_sort | benthicnet a global compilation of seafloor images for deep learning applications |
| url | https://doi.org/10.1038/s41597-025-04491-1 |
| work_keys_str_mv | AT scottclowe benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT benjaminmisiuk benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT isaacxu benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT shakhbozabdulazizov benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT amitrbaroi benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT alexcbastos benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT merlinbest benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT vickiferrini benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT ariellfriedman benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT deborahhart benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT ovehoeghguldberg benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT danielierodiaconou benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT juliamackinmclaughlin benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT kathrynmarkey benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT pedrosmenandro benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT jacquomomonk benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT shreyanemani benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT johnobrien benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT elizabethoh benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT lubayreshitnyk benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT katleenrobert benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT chrismroelfsema benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT jessicaasameoto benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT alexandrecgschimel benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT jordanathomson benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT brittanyrwilson benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT melisacwong benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT craigjbrown benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications AT thomastrappenberg benthicnetaglobalcompilationofseafloorimagesfordeeplearningapplications |