Deep Sea Spy: An Online Citizen Science Annotation Platform for Science and Ocean Literacy

The recent development of deep-sea observatories has enabled the acquisition of high temporal resolution imagery for the study of deep-sea communities’ dynamics from hourly to multi-decadal scales. Camera systems deployed at hydrothermal vents have acquired, since 2010, over 11 Tera bytes of data that cannot be processed by research labs only. While deep learning offers an alternative to human processing, training algorithms requires substantial annotated reference datasets. The project Deep Sea Spy enable citizens to contribute to the annotation of pictures acquired with underwater platforms. The annotation of over 45 000 images supported the development of a multi-participant data validation workflow that can be applied to similar databases. We also present the impact of the platform on the civil society, and how it can serve education and inform managers and policy makers. Deep Sea Spy and the proposed workflow has a strong potential to enhance environmental observation and monitoring.

Keyword(s)

Crowdsourcing, Ocean literacy, Image processing, EMSO-Azores, Ocean Networks Canada, Ecosystem monitoring

Full Text

FilePagesSizeAccess
Preprint
393 Mo
How to cite
Matabos Marjolaine, Cottais Pierre, Leroux Riwan, Cenatiempo Yannick, Gasne-Destaville Charlotte, Roullet Nicolas, Sarrazin Jozee, Tourolle Julie, Borremans Catherine (2024). Deep Sea Spy: An Online Citizen Science Annotation Platform for Science and Ocean Literacy. Preprint. INPRESS. https://doi.org/10.2139/ssrn.4848325, https://archimer.ifremer.fr/doc/00915/102649/

Copy this text