Machine learning techniques to characterize functional traits of plankton from image data - Aix-Marseille Université Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2021

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

1 LOV - Laboratoire d'océanographie de Villefranche
2 ULaval - Université Laval [Québec]
3 LOG - Laboratoire d’Océanologie et de Géosciences (LOG) - UMR 8187
4 UFSC - Universidade Federal de Santa Catarina = Federal University of Santa Catarina [Florianópolis]
5 IBP - Institute of Biogeochemistry and Pollutant Dynamics [ETH Zürich]
6 CEREGE - Centre européen de recherche et d'enseignement des géosciences de l'environnement
7 SIO - UC San Diego - Scripps Institution of Oceanography
8 NOC - National Oceanography Centre [Southampton]
9 IOLR - Israel Oceanographic and Limnological Research
10 UGENT - Universiteit Gent = Ghent University
11 AWI - Alfred Wegener Institute for Polar and Marine Research
12 DTU Centre for Ocean Life
13 LVSN - Laboratoire de Vision et Systèmes Numériques
14 IMEDEA - Institut Mediterrani d'Estudis Avancats
15 LIEC - Laboratoire Interdisciplinaire des Environnements Continentaux
16 IUF - Institut universitaire de France
17 EAWAG - Swiss Federal Insitute of Aquatic Science and Technology [Dübendorf]
18 MIO - Institut méditerranéen d'océanologie
19 Helmholtz-Zentrum Hereon
20 Georgia Tech Lorraine [Metz]
21 EMH - Unité Écologie et Modèles pour l'Halieutique
22 CAU - Christian-Albrechts-Universität zu Kiel = Christian-Albrechts University of Kiel = Université Christian-Albrechts de Kiel
23 URI - University of Rhode Island
24 WHOI - Woods Hole Oceanographic Institution
25 Cukurova University
26 COM - Centro Oceanografico de Malaga [Fuengirola, Spain]
27 IFREMER - Institut Français de Recherche pour l'Exploitation de la Mer
28 Dalhousie University [Halifax]
Frédéric Maps
Cedric Pradalier
Heidi M Sosik

Résumé

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 streams 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 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 other aquatic or terrestrial organisms.
Fichier principal
Vignette du fichier
Orenstein_2021_preprint_artifactz.pdf (5.38 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03482282 , version 1 (15-12-2021)
hal-03482282 , version 2 (30-06-2022)

Licence

Paternité

Identifiants

  • HAL Id : hal-03482282 , version 1

Citer

Eric C Orenstein, Sakina-Dorothée Ayata, Frédéric Maps, Tristan Biard, Érica C Becker, et al.. Machine learning techniques to characterize functional traits of plankton from image data. 2021. ⟨hal-03482282v1⟩

Collections

LOV_COMPLEX
690 Consultations
645 Téléchargements

Partager

Gmail Facebook X LinkedIn More