Semi-Automated Data Processing and Semi-Supervised Machine Learning for the Detection and Classification of Water-Column Fish Schools and Gas Seeps with a Multibeam Echosounder
Multibeam echosounders are widely used for 3D bathymetric mapping, and increasingly for water column studies. However, they rapidly collect huge volumes of data, which poses a challenge for water column data processing that is often still manual and time-consuming, or affected by low efficiency and high false detection rates if automated. This research describes a comprehensive and reproducible workflow that improves efficiency and reliability of target detection and classification, by calculating metrics for target cross-sections using a commercial software before feeding into a feature-based semi-supervised machine learning framework. The method is tested with data collected from an uncalibrated multibeam echosounder around an offshore gas platform in the Adriatic Sea. It resulted in more-efficient target detection, and, although uncertainties regarding user labelled training data need to be underlined, an accuracy of 98% in target classification was reached by using a final pre-trained stacking ensemble model.
2021 -
Journal article
Sensors (Basel) 21 (2021). doi:10.3390/s21092999
Keywords:
multibeam echosounder, water column imaging, Machine learning, fish schools, gas plumes, target detection and classification
The research activities of the Institute are carried out in the context of research, development and innovation projects, both national and international, based on regional funding programs (POR FEAMPA - Regional Operational Program of the European Maritime Affairs Fisheries Fund and Aquaculture and POR FESR - Regional Operational Program of the European Regional Development Fund) or ministerial (PRIN - Projects of relevant national interest, PNRA - National Research Program in Antarctica, PO FEAMPA - National Operational Program European Maritime Affairs Fisheries and Aquaculture Fund) , programs for European Territorial Cooperation (Interreg), direct funding programs of the European Commission (Horizon2020 and Horizon Europe, Life, JPI - Joint Programming Initiatives, ERA-NET Cofund) and thematic collaboration initiatives managed by international organizations such as, for example , the FAO - GFCM (General Fisheries Commission for the Mediterranean). The Institute also develops funded projects in the context of collaborations with private companies in the sectors of the blue economy as well as technology transfer and research results. Research projects, mainly of a collaborative nature, are developed through a wide network of partners that include major Italian and foreign research institutions and universities.
