FN Archimer Export Format
PT J
TI GeoAI for Marine Ecosystem Monitoring: a Complete Workflow to Generate Maps from AI Model Predictions
BT 
AF Talpaert Daudon, Justine
   Contini, Matteo
   Urbina-Barreto, Isabel
   Elliott, Brianna
   Guilhaumon, François
   Joly, Alexis
   Bonhommeau, Sylvain
   Barde, Julien
AS 1:1;2:2;3:3;4:4;5:3;6:5;7:2;8:6;
FF 1:;2:PDG-RBE-DOI;3:;4:;5:;6:;7:PDG-RBE-DOI;8:;
C1 UMR Marbec, IRD, La Reunion, France
   Ifremer DOI, La Reunion, France
   UMR Entropie (Future Maore Reefs), IRD, La Reunion, France
   Duke University, Duke Marine Lab, Beaufort, NC, USA
   INRIA Zenith, Montpellier, France
   UMR Marbec, IRD, France
C2 IRD, FRANCE
   IFREMER, FRANCE
   IRD, FRANCE
   UNIV DUKE, USA
   INRIA, FRANCE
   IRD, FRANCE
SI LA REUNION
   SETE
SE PDG-RBE-DOI
   IRD
UM MARBEC
   ENTROPIE
TC 0
UR https://archimer.ifremer.fr/doc/00844/95549/103355.pdf
LA English
DT Article
DE ;GeoAI;computer vision;deep learning;marine ecology;object detection;segmentation;geospatial;photogrammetry
AB Mapping and monitoring marine ecosystems imply several challenges for data collection and processing: water depth, restricted access to locations, instrumentation costs or weather constraints for sampling, among others. Nowadays, Artificial Intelligence (AI) and Geographic Information System (GIS) open source software can be combined in new kinds of workflows, to annotate and predict objects directly on georeferenced raster data (e.g. orthomosaics). Here, we describe and share the code of a generic method to train a deep learning model with spatial annotations and use it to directly generate model predictions as spatial features. This workflow has been tested and validated in three use cases related to marine ecosystem monitoring at different geographic scales: (i) segmentation of corals on orthomosaics made of underwater images to automate coral reef habitats mapping, (ii) detection and classification of fishing vessels on remote sensing satellite imagery to estimate a proxy of fishing effort (iii) segmentation of marine species and habitats on underwater images with a simple geolocation. Models have been successfully trained and the models predictions are displayed with maps in the three use cases. 
PY 2023
PD JUL
SO The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
SN 2194-9034
PU Copernicus GmbH
VL XLVIII-4/W7-2023
BP 223
EP 230
DI 10.5194/isprs-archives-XLVIII-4-W7-2023-223-2023
ID 95549
ER

EF