Enhancing plankton classification in broadband echosounder data with machine learning

Financing: The Fram Centre (High North Research Centre for Climate and Environmental Research) via the Norwegian Ministry of Climate and Environment

Project lead: Pierre Priou/Akvaplan-niva

Project partners: The Institute of Marine Research and NORCE

Primary objective

To develop a machine learning model for detecting and classifying plankton layers in broadband echosounder data with limited labels. This approach will serve as a stepping stone toward the effective processing of plankton data from large acoustic datasets.

Project structure

The primary objective is attained through three specific work packages (WPs), each addressing a specific research question and hypothesis. In WP1 we identify the best way to prepare broadband echosounder data to train a ML model to classify zooplankton in WP2.

Finally, the results from our approach are summarised and disseminated in WP3.