Weakly supervised learning for landcover mapping of satellite image timeseries via attention-based CNN
The unprecedented possibility to acquire high resolution Satellite Image Time Series (SITS) data is opening new opportunities to monitor the different aspects of the Earth Surface but, at the same time, it is raising up new challenges in term of suitable methods to analyze and exploit such huge amount of rich image data. One of the main tasks associated to SITS data analysis is related to land cover mapping. Due to operational constraints, the collected label information is often limited in volume and obtained at coarse granularity level carrying out inexact and weak knowledge that can affect the whole process. To cope with such issues, in the context of object-based SITS land cover mapping, we propose a new deep learning framework, named TASSEL (aTtentive weAkly Supervised Satellite image time sEries cLassifier), to deal with the weak supervision provided by the coarse granularity labels. Our framework exploits the multifaceted information conveyed by the object-based representation considering object components instead of aggregated object statistics. Furthermore, our framework also produces an additional outcome that supports the model interpretability. Quantitative and qualitative experimental evaluations are carried out on two real-world scenarios. Results indicate that not only TASSEL outperforms the competing approaches in terms of predictive performances, but it also produces valuable extra information that can be practically exploited to interpret model decisions.
Main Authors: | , , , |
---|---|
Format: | article biblioteca |
Language: | eng |
Subjects: | U30 - Méthodes de recherche, U10 - Informatique, mathématiques et statistiques, P31 - Levés et cartographie des sols, cartographie de l'occupation du sol, imagerie par satellite, analyse de séries chronologiques, imagerie, analyse de données, analyse d'image, http://aims.fao.org/aos/agrovoc/c_9000094, http://aims.fao.org/aos/agrovoc/c_36761, http://aims.fao.org/aos/agrovoc/c_28778, http://aims.fao.org/aos/agrovoc/c_36760, http://aims.fao.org/aos/agrovoc/c_15962, http://aims.fao.org/aos/agrovoc/c_36762, |
Online Access: | http://agritrop.cirad.fr/597782/ http://agritrop.cirad.fr/597782/7/597782.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
dig-cirad-fr-597782 |
---|---|
record_format |
koha |
spelling |
dig-cirad-fr-5977822024-01-29T03:23:32Z http://agritrop.cirad.fr/597782/ http://agritrop.cirad.fr/597782/ Weakly supervised learning for landcover mapping of satellite image timeseries via attention-based CNN. Ienco Dino, Gbodjo Yawogan Jean Eudes, Gaetano Raffaele, Interdonato Roberto. 2020. IEEE Access, 8 : 179547-179560.https://doi.org/10.1109/ACCESS.2020.3024133 <https://doi.org/10.1109/ACCESS.2020.3024133> Weakly supervised learning for landcover mapping of satellite image timeseries via attention-based CNN Ienco, Dino Gbodjo, Yawogan Jean Eudes Gaetano, Raffaele Interdonato, Roberto eng 2020 IEEE Access U30 - Méthodes de recherche U10 - Informatique, mathématiques et statistiques P31 - Levés et cartographie des sols cartographie de l'occupation du sol imagerie par satellite analyse de séries chronologiques imagerie analyse de données analyse d'image http://aims.fao.org/aos/agrovoc/c_9000094 http://aims.fao.org/aos/agrovoc/c_36761 http://aims.fao.org/aos/agrovoc/c_28778 http://aims.fao.org/aos/agrovoc/c_36760 http://aims.fao.org/aos/agrovoc/c_15962 http://aims.fao.org/aos/agrovoc/c_36762 The unprecedented possibility to acquire high resolution Satellite Image Time Series (SITS) data is opening new opportunities to monitor the different aspects of the Earth Surface but, at the same time, it is raising up new challenges in term of suitable methods to analyze and exploit such huge amount of rich image data. One of the main tasks associated to SITS data analysis is related to land cover mapping. Due to operational constraints, the collected label information is often limited in volume and obtained at coarse granularity level carrying out inexact and weak knowledge that can affect the whole process. To cope with such issues, in the context of object-based SITS land cover mapping, we propose a new deep learning framework, named TASSEL (aTtentive weAkly Supervised Satellite image time sEries cLassifier), to deal with the weak supervision provided by the coarse granularity labels. Our framework exploits the multifaceted information conveyed by the object-based representation considering object components instead of aggregated object statistics. Furthermore, our framework also produces an additional outcome that supports the model interpretability. Quantitative and qualitative experimental evaluations are carried out on two real-world scenarios. Results indicate that not only TASSEL outperforms the competing approaches in terms of predictive performances, but it also produces valuable extra information that can be practically exploited to interpret model decisions. article info:eu-repo/semantics/article Journal Article info:eu-repo/semantics/publishedVersion http://agritrop.cirad.fr/597782/7/597782.pdf text cc_by info:eu-repo/semantics/openAccess https://creativecommons.org/licenses/by/4.0/ https://doi.org/10.1109/ACCESS.2020.3024133 10.1109/ACCESS.2020.3024133 info:eu-repo/semantics/altIdentifier/doi/10.1109/ACCESS.2020.3024133 info:eu-repo/semantics/altIdentifier/purl/https://doi.org/10.1109/ACCESS.2020.3024133 |
institution |
CIRAD FR |
collection |
DSpace |
country |
Francia |
countrycode |
FR |
component |
Bibliográfico |
access |
En linea |
databasecode |
dig-cirad-fr |
tag |
biblioteca |
region |
Europa del Oeste |
libraryname |
Biblioteca del CIRAD Francia |
language |
eng |
topic |
U30 - Méthodes de recherche U10 - Informatique, mathématiques et statistiques P31 - Levés et cartographie des sols cartographie de l'occupation du sol imagerie par satellite analyse de séries chronologiques imagerie analyse de données analyse d'image http://aims.fao.org/aos/agrovoc/c_9000094 http://aims.fao.org/aos/agrovoc/c_36761 http://aims.fao.org/aos/agrovoc/c_28778 http://aims.fao.org/aos/agrovoc/c_36760 http://aims.fao.org/aos/agrovoc/c_15962 http://aims.fao.org/aos/agrovoc/c_36762 U30 - Méthodes de recherche U10 - Informatique, mathématiques et statistiques P31 - Levés et cartographie des sols cartographie de l'occupation du sol imagerie par satellite analyse de séries chronologiques imagerie analyse de données analyse d'image http://aims.fao.org/aos/agrovoc/c_9000094 http://aims.fao.org/aos/agrovoc/c_36761 http://aims.fao.org/aos/agrovoc/c_28778 http://aims.fao.org/aos/agrovoc/c_36760 http://aims.fao.org/aos/agrovoc/c_15962 http://aims.fao.org/aos/agrovoc/c_36762 |
spellingShingle |
U30 - Méthodes de recherche U10 - Informatique, mathématiques et statistiques P31 - Levés et cartographie des sols cartographie de l'occupation du sol imagerie par satellite analyse de séries chronologiques imagerie analyse de données analyse d'image http://aims.fao.org/aos/agrovoc/c_9000094 http://aims.fao.org/aos/agrovoc/c_36761 http://aims.fao.org/aos/agrovoc/c_28778 http://aims.fao.org/aos/agrovoc/c_36760 http://aims.fao.org/aos/agrovoc/c_15962 http://aims.fao.org/aos/agrovoc/c_36762 U30 - Méthodes de recherche U10 - Informatique, mathématiques et statistiques P31 - Levés et cartographie des sols cartographie de l'occupation du sol imagerie par satellite analyse de séries chronologiques imagerie analyse de données analyse d'image http://aims.fao.org/aos/agrovoc/c_9000094 http://aims.fao.org/aos/agrovoc/c_36761 http://aims.fao.org/aos/agrovoc/c_28778 http://aims.fao.org/aos/agrovoc/c_36760 http://aims.fao.org/aos/agrovoc/c_15962 http://aims.fao.org/aos/agrovoc/c_36762 Ienco, Dino Gbodjo, Yawogan Jean Eudes Gaetano, Raffaele Interdonato, Roberto Weakly supervised learning for landcover mapping of satellite image timeseries via attention-based CNN |
description |
The unprecedented possibility to acquire high resolution Satellite Image Time Series (SITS) data is opening new opportunities to monitor the different aspects of the Earth Surface but, at the same time, it is raising up new challenges in term of suitable methods to analyze and exploit such huge amount of rich image data. One of the main tasks associated to SITS data analysis is related to land cover mapping. Due to operational constraints, the collected label information is often limited in volume and obtained at coarse granularity level carrying out inexact and weak knowledge that can affect the whole process. To cope with such issues, in the context of object-based SITS land cover mapping, we propose a new deep learning framework, named TASSEL (aTtentive weAkly Supervised Satellite image time sEries cLassifier), to deal with the weak supervision provided by the coarse granularity labels. Our framework exploits the multifaceted information conveyed by the object-based representation considering object components instead of aggregated object statistics. Furthermore, our framework also produces an additional outcome that supports the model interpretability. Quantitative and qualitative experimental evaluations are carried out on two real-world scenarios. Results indicate that not only TASSEL outperforms the competing approaches in terms of predictive performances, but it also produces valuable extra information that can be practically exploited to interpret model decisions. |
format |
article |
topic_facet |
U30 - Méthodes de recherche U10 - Informatique, mathématiques et statistiques P31 - Levés et cartographie des sols cartographie de l'occupation du sol imagerie par satellite analyse de séries chronologiques imagerie analyse de données analyse d'image http://aims.fao.org/aos/agrovoc/c_9000094 http://aims.fao.org/aos/agrovoc/c_36761 http://aims.fao.org/aos/agrovoc/c_28778 http://aims.fao.org/aos/agrovoc/c_36760 http://aims.fao.org/aos/agrovoc/c_15962 http://aims.fao.org/aos/agrovoc/c_36762 |
author |
Ienco, Dino Gbodjo, Yawogan Jean Eudes Gaetano, Raffaele Interdonato, Roberto |
author_facet |
Ienco, Dino Gbodjo, Yawogan Jean Eudes Gaetano, Raffaele Interdonato, Roberto |
author_sort |
Ienco, Dino |
title |
Weakly supervised learning for landcover mapping of satellite image timeseries via attention-based CNN |
title_short |
Weakly supervised learning for landcover mapping of satellite image timeseries via attention-based CNN |
title_full |
Weakly supervised learning for landcover mapping of satellite image timeseries via attention-based CNN |
title_fullStr |
Weakly supervised learning for landcover mapping of satellite image timeseries via attention-based CNN |
title_full_unstemmed |
Weakly supervised learning for landcover mapping of satellite image timeseries via attention-based CNN |
title_sort |
weakly supervised learning for landcover mapping of satellite image timeseries via attention-based cnn |
url |
http://agritrop.cirad.fr/597782/ http://agritrop.cirad.fr/597782/7/597782.pdf |
work_keys_str_mv |
AT iencodino weaklysupervisedlearningforlandcovermappingofsatelliteimagetimeseriesviaattentionbasedcnn AT gbodjoyawoganjeaneudes weaklysupervisedlearningforlandcovermappingofsatelliteimagetimeseriesviaattentionbasedcnn AT gaetanoraffaele weaklysupervisedlearningforlandcovermappingofsatelliteimagetimeseriesviaattentionbasedcnn AT interdonatoroberto weaklysupervisedlearningforlandcovermappingofsatelliteimagetimeseriesviaattentionbasedcnn |
_version_ |
1792500120851316736 |