Flying Robots Teach Floating Robots—A Machine Learning Approach for Marine Habitat Mapping Based on Combined Datasets

Unmanned aerial and autonomous surface vehicles (UAVs and ASVs, respectively) are two emerging technologies for the mapping of coastal and marine environments. Using UAV photogrammetry, the sea-bottom composition can be resolved with very high fidelity in shallow waters. At greater depths, acoustic...

Full description

Saved in:
Bibliographic Details
Main Authors: Zacharias Kapelonis, Georgios Chatzigeorgiou, Manolis Ntoumas, Panos Grigoriou, Manos Pettas, Spyros Michelinakis, Ricardo Correia, Catarina Rasquilha Lemos, Luis Menezes Pinheiro, Caio Lomba, João Fortuna, Rui Loureiro, André Santos, Eva Chatzinikolaou
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Journal of Marine Science and Engineering
Subjects:
Online Access:https://www.mdpi.com/2077-1312/13/3/611
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850279467651432448
author Zacharias Kapelonis
Georgios Chatzigeorgiou
Manolis Ntoumas
Panos Grigoriou
Manos Pettas
Spyros Michelinakis
Ricardo Correia
Catarina Rasquilha Lemos
Luis Menezes Pinheiro
Caio Lomba
João Fortuna
Rui Loureiro
André Santos
Eva Chatzinikolaou
author_facet Zacharias Kapelonis
Georgios Chatzigeorgiou
Manolis Ntoumas
Panos Grigoriou
Manos Pettas
Spyros Michelinakis
Ricardo Correia
Catarina Rasquilha Lemos
Luis Menezes Pinheiro
Caio Lomba
João Fortuna
Rui Loureiro
André Santos
Eva Chatzinikolaou
author_sort Zacharias Kapelonis
collection DOAJ
description Unmanned aerial and autonomous surface vehicles (UAVs and ASVs, respectively) are two emerging technologies for the mapping of coastal and marine environments. Using UAV photogrammetry, the sea-bottom composition can be resolved with very high fidelity in shallow waters. At greater depths, acoustic methodologies have far better propagation properties compared to optics; therefore, ASVs equipped with multibeam echosounders (MBES) are better-suited for mapping applications in deeper waters. In this work, a sea-bottom classification methodology is presented for mapping the protected habitat of Mediterranean seagrass <i>Posidonia oceanica</i> (habitat code 1120) in a coastal subregion of Heraklion (Crete, Greece). The methodology implements a machine learning scheme, where knowledge obtained from UAV imagery is embedded (through training) into a classifier that utilizes acoustic backscatter intensity and features derived from the MBES data provided by an ASV. Accuracy and precision scores of greater than 85% compared with visual census ground-truth data for both optical and acoustic classifiers indicate that this hybrid mapping approach is promising to mitigate the depth-induced bias in UAV-only models. The latter is especially interesting in cases where the studied habitat boundaries extend beyond depths that can be studied via aerial devices’ optics, as is the case with <i>P. oceanica</i> meadows.
format Article
id doaj-art-c7346bfa08884abc8ae68fd6852bc670
institution OA Journals
issn 2077-1312
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Journal of Marine Science and Engineering
spelling doaj-art-c7346bfa08884abc8ae68fd6852bc6702025-08-20T01:49:04ZengMDPI AGJournal of Marine Science and Engineering2077-13122025-03-0113361110.3390/jmse13030611Flying Robots Teach Floating Robots—A Machine Learning Approach for Marine Habitat Mapping Based on Combined DatasetsZacharias Kapelonis0Georgios Chatzigeorgiou1Manolis Ntoumas2Panos Grigoriou3Manos Pettas4Spyros Michelinakis5Ricardo Correia6Catarina Rasquilha Lemos7Luis Menezes Pinheiro8Caio Lomba9João Fortuna10Rui Loureiro11André Santos12Eva Chatzinikolaou13Institute of Marine Biological Resources and Inland Waters (IMBRIW), Hellenic Centre for Marine Research (HCMR), 16452 Athens, GreeceInstitute of Marine Biology, Biotechnology and Aquaculture (IMBBC), Hellenic Centre for Marine Research (HCMR), 71500 Heraklion, GreeceInstitute of Oceanography (IO), Hellenic Centre for Marine Research (HCMR), 71500 Heraklion, GreeceCretaquarium, Hellenic Centre for Marine Research (HCMR), 71500 Heraklion, GreeceInstitute of Oceanography (IO), Hellenic Centre for Marine Research (HCMR), 71500 Heraklion, GreeceInstitute of Oceanography (IO), Hellenic Centre for Marine Research (HCMR), 71500 Heraklion, GreeceCESAM—Centre for Environmental and Marine Studies, Geosciences Department, University of Aveiro (UA), Campus Universitário de Santiago, 3810-193 Aveiro, PortugalCentre of Engineering and Product Development (CEiiA), 4450-017 Matosinhos, PortugalCESAM—Centre for Environmental and Marine Studies, Geosciences Department, University of Aveiro (UA), Campus Universitário de Santiago, 3810-193 Aveiro, PortugalCentre of Engineering and Product Development (CEiiA), 4450-017 Matosinhos, PortugalCentre of Engineering and Product Development (CEiiA), 4450-017 Matosinhos, PortugalCentre of Engineering and Product Development (CEiiA), 4450-017 Matosinhos, PortugalCentre of Engineering and Product Development (CEiiA), 4450-017 Matosinhos, PortugalInstitute of Marine Biology, Biotechnology and Aquaculture (IMBBC), Hellenic Centre for Marine Research (HCMR), 71500 Heraklion, GreeceUnmanned aerial and autonomous surface vehicles (UAVs and ASVs, respectively) are two emerging technologies for the mapping of coastal and marine environments. Using UAV photogrammetry, the sea-bottom composition can be resolved with very high fidelity in shallow waters. At greater depths, acoustic methodologies have far better propagation properties compared to optics; therefore, ASVs equipped with multibeam echosounders (MBES) are better-suited for mapping applications in deeper waters. In this work, a sea-bottom classification methodology is presented for mapping the protected habitat of Mediterranean seagrass <i>Posidonia oceanica</i> (habitat code 1120) in a coastal subregion of Heraklion (Crete, Greece). The methodology implements a machine learning scheme, where knowledge obtained from UAV imagery is embedded (through training) into a classifier that utilizes acoustic backscatter intensity and features derived from the MBES data provided by an ASV. Accuracy and precision scores of greater than 85% compared with visual census ground-truth data for both optical and acoustic classifiers indicate that this hybrid mapping approach is promising to mitigate the depth-induced bias in UAV-only models. The latter is especially interesting in cases where the studied habitat boundaries extend beyond depths that can be studied via aerial devices’ optics, as is the case with <i>P. oceanica</i> meadows.https://www.mdpi.com/2077-1312/13/3/611mobile platformsunmanned and autonomous vehiclesmultibeam echosoundermarine habitat mappingaerial mappingtransfer learning
spellingShingle Zacharias Kapelonis
Georgios Chatzigeorgiou
Manolis Ntoumas
Panos Grigoriou
Manos Pettas
Spyros Michelinakis
Ricardo Correia
Catarina Rasquilha Lemos
Luis Menezes Pinheiro
Caio Lomba
João Fortuna
Rui Loureiro
André Santos
Eva Chatzinikolaou
Flying Robots Teach Floating Robots—A Machine Learning Approach for Marine Habitat Mapping Based on Combined Datasets
Journal of Marine Science and Engineering
mobile platforms
unmanned and autonomous vehicles
multibeam echosounder
marine habitat mapping
aerial mapping
transfer learning
title Flying Robots Teach Floating Robots—A Machine Learning Approach for Marine Habitat Mapping Based on Combined Datasets
title_full Flying Robots Teach Floating Robots—A Machine Learning Approach for Marine Habitat Mapping Based on Combined Datasets
title_fullStr Flying Robots Teach Floating Robots—A Machine Learning Approach for Marine Habitat Mapping Based on Combined Datasets
title_full_unstemmed Flying Robots Teach Floating Robots—A Machine Learning Approach for Marine Habitat Mapping Based on Combined Datasets
title_short Flying Robots Teach Floating Robots—A Machine Learning Approach for Marine Habitat Mapping Based on Combined Datasets
title_sort flying robots teach floating robots a machine learning approach for marine habitat mapping based on combined datasets
topic mobile platforms
unmanned and autonomous vehicles
multibeam echosounder
marine habitat mapping
aerial mapping
transfer learning
url https://www.mdpi.com/2077-1312/13/3/611
work_keys_str_mv AT zachariaskapelonis flyingrobotsteachfloatingrobotsamachinelearningapproachformarinehabitatmappingbasedoncombineddatasets
AT georgioschatzigeorgiou flyingrobotsteachfloatingrobotsamachinelearningapproachformarinehabitatmappingbasedoncombineddatasets
AT manolisntoumas flyingrobotsteachfloatingrobotsamachinelearningapproachformarinehabitatmappingbasedoncombineddatasets
AT panosgrigoriou flyingrobotsteachfloatingrobotsamachinelearningapproachformarinehabitatmappingbasedoncombineddatasets
AT manospettas flyingrobotsteachfloatingrobotsamachinelearningapproachformarinehabitatmappingbasedoncombineddatasets
AT spyrosmichelinakis flyingrobotsteachfloatingrobotsamachinelearningapproachformarinehabitatmappingbasedoncombineddatasets
AT ricardocorreia flyingrobotsteachfloatingrobotsamachinelearningapproachformarinehabitatmappingbasedoncombineddatasets
AT catarinarasquilhalemos flyingrobotsteachfloatingrobotsamachinelearningapproachformarinehabitatmappingbasedoncombineddatasets
AT luismenezespinheiro flyingrobotsteachfloatingrobotsamachinelearningapproachformarinehabitatmappingbasedoncombineddatasets
AT caiolomba flyingrobotsteachfloatingrobotsamachinelearningapproachformarinehabitatmappingbasedoncombineddatasets
AT joaofortuna flyingrobotsteachfloatingrobotsamachinelearningapproachformarinehabitatmappingbasedoncombineddatasets
AT ruiloureiro flyingrobotsteachfloatingrobotsamachinelearningapproachformarinehabitatmappingbasedoncombineddatasets
AT andresantos flyingrobotsteachfloatingrobotsamachinelearningapproachformarinehabitatmappingbasedoncombineddatasets
AT evachatzinikolaou flyingrobotsteachfloatingrobotsamachinelearningapproachformarinehabitatmappingbasedoncombineddatasets