BenthicNet: A global compilation of seafloor images for deep learning applications
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.
Authors
- Jacquomo Monk (ORCID: https://orcid.org/0000-0002-1874-0619)
- Elizabeth Oh (ORCID: https://orcid.org/0000-0001-7061-5068)
- John T. O’Brien (ORCID: https://orcid.org/0000-0002-0837-5080)
- Ove Hoegh‐Guldberg (ORCID: https://orcid.org/0000-0001-7510-6713)
- Pedro S. Menandro (ORCID: https://orcid.org/0000-0002-9667-6851)
- Melisa C. Wong (ORCID: https://orcid.org/0000-0001-5192-5693)
- Craig J. Brown (ORCID: https://orcid.org/0000-0002-5125-978X)
- Ariell Friedman
- Jordan A. Thomson (ORCID: https://orcid.org/0000-0003-3751-9490)
- Scott Lowe (ORCID: https://orcid.org/0000-0002-5237-3867)
- Julia Mackin-McLaughlin (ORCID: https://orcid.org/0000-0003-3232-4619)
- Chris Roelfsema (ORCID: https://orcid.org/0000-0003-0182-1356)
- Shreya Nemani
- Katleen Robert (ORCID: https://orcid.org/0000-0003-0119-5359)
- Brittany R. Wilson (ORCID: https://orcid.org/0000-0002-9839-8322)
- V. L. Ferrini (ORCID: https://orcid.org/0000-0002-6054-5040)
- Alex Cardoso Bastos (ORCID: https://orcid.org/0000-0002-1272-1134)
- Daniel Ierodiaconou (ORCID: https://orcid.org/0000-0002-7832-4801)
- Alexandre C. G. Schimel (ORCID: https://orcid.org/0000-0001-9301-0803)
- Benjamin Misiuk (ORCID: https://orcid.org/0000-0001-5822-4574)
- Deborah Hart (ORCID: https://orcid.org/0000-0002-9066-4614)
- Kathryn Markey (ORCID: https://orcid.org/0000-0002-2556-3366)
- Thomas Trappenberg (ORCID: https://orcid.org/0000-0002-6144-8963)
- Shakhboz Abdulazizov
- Amit R. Baroi
- Luba Y. Reshitnyk
- Jessica A. Sameoto (ORCID: https://orcid.org/0009-0001-8170-280X)
- M. Borel Best
- Isaac Xu
Institutions
- The University of Sydney (AU)
- Geological Survey of Norway (NO)
- University of Tasmania (AU)
- Dalhousie University (CA)
- Memorial University of Newfoundland (CA)
- National Oceanic and Atmospheric Administration (US)
- Fisheries and Oceans Canada (CA)
- Deakin University (AU)
- Lamont-Doherty Earth Observatory (US)
- The University of Queensland (AU)
- Bedford Institute of Oceanography (CA)
- Tula Foundation (CA)
- NOAA National Marine Fisheries Service Northeast Fisheries Science Center (US)
- Vector Institute (CA)
- Australian Centre for Robotic Vision (AU)
- Hakai Institute
- Institute for Marine and Antarctic Studies
- Universidade Federal do Espírito Santo (BR)
- Occidental College (US)
- Columbia University (US)
Publication Details
- Journal
- Scientific Data
- Published
- 2025-02-07
- DOI
- https://doi.org/10.1038/s41597-025-04491-1
- Citations
- 11
- Primary Topic
- Underwater Acoustics Research
- Type
- article
- Field-Weighted Citation Impact
- 19.73
Funders
- Australian Government
- University of Tasmania
- Atlantic Canada Opportunities Agency
- Australian Centre for Field Robotics
- Ocean Frontier Institute
- Alliance de recherche numérique du Canada
- Hakai Institute