Machine learning techniques to characterize functional traits of plankton from image data

Type Article
Date 2022-08
Language English
Author(s) Orenstein Eric C.1, Ayata Sakina‐dorothée1, 2, Maps Frédéric3, 4, Becker Érica C.5, Benedetti Fabio6, Biard Tristan7, de Garidel‐thoron Thibault8, Ellen Jeffrey S.9, Ferrario Filippo3, 4, 10, Giering Sarah L. C.11, Guy‐haim Tamar12, Hoebeke Laura13, Iversen Morten Hvitfeldt14, Kiørboe Thomas15, Lalonde Jean‐françois16, Lana Arancha17, Laviale Martin18, Lombard Fabien1, Lorimer Tom19, Martini Séverine20, Meyer Albin18, Möller Klas Ove21, Niehoff Barbara14, Ohman Mark D.9, Pradalier Cédric22, Romagnan Jean-Baptiste23, Schröder Simon‐martin24, Sonnet Virginie25, Sosik Heidi M.26, Stemmann Lars S.1, Stock Michiel13, Terbiyik‐kurt Tuba27, Valcárcel‐pérez Nerea28, Vilgrain Laure1, Wacquet GuillaumeORCID29, Waite Anya M.30, Irisson Jean‐olivier1
Affiliation(s) 1 : Sorbonne Université, CNRS, Laboratoire d'Océanographie de Villefranche Villefranche‐sur‐Mer ,France
2 : Sorbonne Université, Laboratoire d'Océanographie et du Climat, Institut Pierre Simon Laplace (LOCEAN‐IPSL, SU/CNRS/IRD/MNHN) Paris ,France
3 : Département de Biologie Université Laval Québec, Canada
4 : Takuvik Joint International Laboratory Université Laval‐CNRS (UMI 3376), Québec‐Océan, Université Laval Québec ,Canada
5 : Universidade Federal de Santa Catarina (UFSC) Florianópolis Santa Catarina ,Brazil
6 : ETH Zürich Institute of Biogeochemistry and Pollutant Dynamics Zürich ,Switzerland
7 : Laboratoire d'Océanologie et de Géosciences Université du Littoral Côte d'Opale, Université de Lille, CNRS, UMR 8187 Wimereux ,France
8 : Aix‐Marseille Université, CNRS, IRD, Coll. de France, INRAE, CEREGE Aix en Provence, France
9 : Scripps Institution of Oceanography, University of California San Diego La Jolla California, USA
10 : Department of Fisheries and Oceans Maurice Lamontagne Institute Mont‐Joli Québec, Canada
11 : Ocean Biogeosciences National Oceanography Centre Southampton, UK
12 : National Institute of Oceanography, Israel Oceanographic and Limnological Research Haifa ,Israel
13 : KERMIT, Department of Data Analysis and Mathematical Modelling Ghent University Ghent ,Belgium
14 : Alfred Wegener Institute for Polar and Marine Research Bremerhaven, Germany
15 : Centre for Ocean Life, DTU‐Aqua Technical University of Denmark Kongens Lyngby, Denmark
16 : Laboratoire de Vision et Systèmes Numériques Université Laval Québec City Québec ,Canada
17 : Institut Mediterrani d'Estudis Avançats (IMEDEA, UIB‐CSIC) Balearic Islands ,Spain
18 : Université de Lorraine, CNRS, LIEC Metz ,France
19 : Eawag Dübendorf, Switzerland
20 : Aix Marseille University, Université de Toulon, CNRS, IRD, MIO UM Marseille ,France
21 : Helmholtz‐Zentrum Hereon Institute of Carbon Cycle Geesthacht,Germany
22 : GeorgiaTech Lorraine CNRS IRL GT‐CNRS Metz ,France
23 : IFREMER, Centre Atlantique, Laboratoire Ecologie et Modèles pour l'Halieutique (EMH) Unité HALGO, UMR DECOD Nantes ,France
24 : Kiel University Kiel, Germany
25 : Graduate School of Oceanography University of Rhode Island Narragansett Rhode Island ,USA
26 : Woods Hole Oceanographic Institution Woods Hole Massachusetts ,USA
27 : Department of Basic Sciences Cukurova University, Faculty of Fisheries Adana ,Turkey
28 : Centro Oceanográfico de Málaga, IEO, CSIC Fuengirola, Spain
29 : IFREMER, Laboratoire Environnement Ressources Boulogne‐sur‐Mer ,France
30 : Ocean Frontier Institute, Dalhousie University Halifax Nova Scotia, Canada
Source Limnology And Oceanography (0024-3590) (Wiley), 2022-08 , Vol. 67 , N. 8 , P. 1647-1669
DOI 10.1002/lno.12101
WOS© Times Cited 26
Abstract

Plankton imaging systems supported by automated classification and analysis have improved ecologists' ability to observe aquatic ecosystems. Today, we are on the cusp of reliably tracking plankton populations with a suite of lab-based and in situ tools, collecting imaging data at unprecedentedly fine spatial and temporal scales. But these data have potential well beyond examining the abundances of different taxa; the individual images themselves contain a wealth of information on functional traits. Here, we outline traits that could be measured from image data, suggest machine learning and computer vision approaches to extract functional trait information from the images, and discuss promising avenues for novel studies. The approaches we discuss are data agnostic and are broadly applicable to imagery of other aquatic or terrestrial organisms.
Full Text
File Pages Size Access
Publisher's official version 23 1 MB Open access
Appendix S1 Supporting Information 21 KB Open access
Top of the page

How to cite 

Orenstein Eric C., Ayata Sakina‐dorothée, Maps Frédéric, Becker Érica C., Benedetti Fabio, Biard Tristan, de Garidel‐thoron Thibault, Ellen Jeffrey S., Ferrario Filippo, Giering Sarah L. C., Guy‐haim Tamar, Hoebeke Laura, Iversen Morten Hvitfeldt, Kiørboe Thomas, Lalonde Jean‐françois, Lana Arancha, Laviale Martin, Lombard Fabien, Lorimer Tom, Martini Séverine, Meyer Albin, Möller Klas Ove, Niehoff Barbara, Ohman Mark D., Pradalier Cédric, Romagnan Jean-Baptiste, Schröder Simon‐martin, Sonnet Virginie, Sosik Heidi M., Stemmann Lars S., Stock Michiel, Terbiyik‐kurt Tuba, Valcárcel‐pérez Nerea, Vilgrain Laure, Wacquet Guillaume, Waite Anya M., Irisson Jean‐olivier (2022). Machine learning techniques to characterize functional traits of plankton from image data. Limnology And Oceanography, 67(8), 1647-1669. Publisher's official version : https://doi.org/10.1002/lno.12101 , Open Access version : https://archimer.ifremer.fr/doc/00782/89435/