New tool for biological monitoring of the seafloor with help of AI and citizen science

[2021-02-24] We are happy to launch a new open-source tool combining citizens science and machine learning to extract ecological data from underwater footage.

The increasing access to remotely or autonomously operated camera technologies offers vast opportunities to collect large amounts of biological data from the ocean. Today most marine research institutes do record underwater habitats on a continuous base and hence are able to generate thousands and thousands of hours of footage of marine ecosystems. But who is going to analyze these data? Scientists alone will not be able to sieve through terabytes of underwater movies to find relevant biological information. The Koster Seafloor Observatory (KSO) combines the power of citizen science and artificial intelligence to effectively extract occurrence and abundance data for key ecological species and habitats from subsea recordings. We tested the system in a Marine Protected Area in Sweden – the Kosterhavets National Park showing that the tool can be used to monitor and discover habitat building species such as the cold water coral Desmophyllum pertusum. The tool has been developed by Ocean Data Factory [https://oceandatafactory.se/], Wildlife.ai [https://www.wildlife.ai/] and Swedish Biodiversity Data Infrastructure [https://biodiversitydata.se/].