Show simple item record

dc.contributor.author
Rousset, Guillaume
dc.contributor.author
Despinoy, Marc
dc.contributor.author
Schindler, Konrad
dc.contributor.author
Mangeas, Morgan
dc.date.accessioned
2021-07-05T14:34:59Z
dc.date.available
2021-06-28T04:05:27Z
dc.date.available
2021-07-05T14:34:59Z
dc.date.issued
2021-06-02
dc.identifier.issn
2072-4292
dc.identifier.other
10.3390/rs13122257
en_US
dc.identifier.uri
http://hdl.handle.net/20.500.11850/491577
dc.identifier.doi
10.3929/ethz-b-000491577
dc.description.abstract
Land use (LU) and land cover (LC) are two complementary pieces of cartographic information used for urban planning and environmental monitoring. In the context of New Caledonia, a biodiversity hotspot, the availability of up-to-date LULC maps is essential to monitor the impact of extreme events such as cyclones and human activities on the environment. With the democratization of satellite data and the development of high-performance deep learning techniques, it is possible to create these data automatically. This work aims at determining the best current deep learning configuration (pixel-wise vs. semantic labelling architectures, data augmentation, image prepossessing, …), to perform LULC mapping in a complex, subtropical environment. For this purpose, a specific data set based on SPOT6 satellite data was created and made available for the scientific community as an LULC benchmark in a tropical, complex environment using five representative areas of New Caledonia labelled by a human operator: four used as training sets, and the fifth as a test set. Several architectures were trained and the resulting classification was compared with a state-of-the-art machine learning technique: XGboost. We also assessed the relevance of popular neo-channels derived from the raw observations in the context of deep learning. The deep learning approach showed comparable results to XGboost for LC detection and over-performed it on the LU detection task (61.45% vs. 51.56% of overall accuracy). Finally, adding LC classification output of the dedicated deep learning architecture to the raw channels input significantly improved the overall accuracy of the deep learning LU classification task (63.61% of overall accuracy). All the data used in this study are available on line for the remote sensing community and for assessing other LULC detection techniques.
en_US
dc.format
application/pdf
en_US
dc.language.iso
en
en_US
dc.publisher
MDPI
en_US
dc.rights.uri
http://creativecommons.org/licenses/by/4.0/
dc.subject
New Caledonia
en_US
dc.subject
remote sensing
en_US
dc.subject
land use
en_US
dc.subject
land cover
en_US
dc.subject
deep learning
en_US
dc.subject
XGBoost
en_US
dc.subject
neural network
en_US
dc.subject
neo-channels
en_US
dc.title
Assessment of Deep Learning Techniques for Land Use Land Cover Classification in Southern New Caledonia
en_US
dc.type
Journal Article
dc.rights.license
Creative Commons Attribution 4.0 International
dc.date.published
2021-06-09
ethz.journal.title
Remote Sensing
ethz.journal.volume
13
en_US
ethz.journal.issue
12
en_US
ethz.journal.abbreviated
Remote sens.
ethz.pages.start
2257
en_US
ethz.size
22 p.
en_US
ethz.version.deposit
publishedVersion
en_US
ethz.identifier.wos
ethz.identifier.scopus
ethz.publication.place
Basel
en_US
ethz.publication.status
published
en_US
ethz.date.deposited
2021-06-28T04:05:36Z
ethz.source
SCOPUS
ethz.eth
yes
en_US
ethz.availability
Open access
en_US
ethz.rosetta.installDate
2021-07-05T14:35:07Z
ethz.rosetta.lastUpdated
2021-07-05T14:35:07Z
ethz.rosetta.exportRequired
true
ethz.rosetta.versionExported
true
ethz.COinS
ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.atitle=Assessment%20of%20Deep%20Learning%20Techniques%20for%20Land%20Use%20Land%20Cover%20Classification%20in%20Southern%20New%20Caledonia&rft.jtitle=Remote%20Sensing&rft.date=2021-06-02&rft.volume=13&rft.issue=12&rft.spage=2257&rft.issn=2072-4292&rft.au=Rousset,%20Guillaume&Despinoy,%20Marc&Schindler,%20Konrad&Mangeas,%20Morgan&rft.genre=article&rft_id=info:doi/10.3390/rs13122257&
 Search print copy at ETH Library

Files in this item

Thumbnail

Publication type

Show simple item record