Centre dedocumentation
scientifique

Multi-nomenclature, multi-resolution joint translation: an application to land-cover mapping / Luc Baudoux in International journal of geographical information science IJGIS, vol 37 n° 2 (February 2023)  

Public [article]  inInternational journal of geographical information science IJGIS > vol 37 n° 2 (February 2023) . - pp ? Titre : Multi-nomenclature, multi-resolution joint translation: an application to land-cover mapping Type de document : Article/Communication Auteurs : Luc Baudoux , Auteur ; Jordi Inglada, Auteur ; Clément Mallet , Auteur Année de publication : 2023 Projets : AI4GEO / Article en page(s) : pp ? Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Cartographie thématique [Termes IGN] apprentissage profond [Termes IGN] carte d'occupation du sol [Termes IGN] carte d'utilisation du sol [Termes IGN] carte thématique [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] données d'entrainement (apprentissage automatique) [Termes IGN] harmonisation des données [Termes IGN] nomenclature [Termes IGN] pouvoir de résolution géométrique Résumé : (auteur) Land-use/land-cover (LULC) maps describe the Earth’s surface with discrete classes at a specific spatial resolution. The chosen classes and resolution highly depend on peculiar uses, making it mandatory to develop methods to adapt these characteristics for a large range of applications. Recently, a convolutional neural network (CNN)-based method was introduced to take into account both spatial and geographical context to translate a LULC map into another one. However, this model only works for two maps: one source and one target. Inspired by natural language translation using multiple-language models, this article explores how to translate one LULC map into several targets with distinct nomenclatures and spatial resolutions. We first propose a new data set based on six open access LULC maps to train our CNN-based encoder-decoder framework. We then apply such a framework to convert each of these six maps into each of the others using our Multi-Landcover Translation network (MLCT-Net). Extensive experiments are conducted at a country scale (namely France). The results reveal that our MLCT-Net outperforms its semantic counterparts and gives on par results with mono-LULC models when evaluated on areas similar to those used for training. Furthermore, it outperforms the mono-LULC models when applied to totally new landscapes. Numéro de notice : A2023-075 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Thématique : GEOMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/13658816.2022.2120996 Date de publication en ligne : 10/10/2022 En ligne : https://doi.org/10.1080/13658816.2022.2120996 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101797 [article]

Hyperspectral imagery and urban areas: results of the HYEP project / Christiane Weber in Revue Française de Photogrammétrie et de Télédétection, n° 224 (2022)  

Public [article]  inRevue Française de Photogrammétrie et de Télédétection > n° 224 (2022) . - pp 75 - 92 Titre : Hyperspectral imagery and urban areas: results of the HYEP project Type de document : Article/Communication Auteurs : Christiane Weber, Auteur ; Xavier Briottet , Auteur ; Thomas Houet, Auteur ; Sébastien Gadal, Auteur ; Rahim Aguejdad, Auteur ; Yannick Deville, Auteur ; Mauro Dalla Mura, Auteur ; Clément Mallet , Auteur ; Arnaud Le Bris , Auteur ; et al., Auteur Année de publication : 2022 Projets : HYEP / Weber, Christiane Article en page(s) : pp 75 - 92 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image [Termes IGN] analyse comparative [Termes IGN] détection d'objet [Termes IGN] fusion d'images [Termes IGN] image hyperspectrale [Termes IGN] Lituanie [Termes IGN] milieu urbain [Termes IGN] panneau photovoltaïque [Termes IGN] surface imperméable [Termes IGN] Toulouse Résumé : (Auteur) The HYEP project (ANR HYEP 14-CE22-0016-01 Hyperspectral imagery for Environmental urban Planning - Mobility and Urban Systems Programme - 2014) confirmed the interest of a global approach to the urban environment by remote sensing and in particular by using hyperspectral imaging (HI). The interest of hyperspectral images lies in the range of information provided over wavelengths from 0.4 to 4 μm; they thus provide access to spectral quantities of interest and to chemical or biophysical parameters of the surface. HYEP's objective was to specify this and to propose a panel of methods and treatments taking into account the characteristics of other existing sensors in order to compare their performance. The developments carried out were applied and evaluated on hyperspectral airborne images acquired in Toulouse and Kaunas (Lithuania), also used to synthesize space systems: Sentinel-2, Hypxim/Biodiversity and Pleiades. Among the locks identified were those related to improving the spatial capabilities of the sensors and spatial scale changes, which were partially overcome by fusion and sharpening approaches, which proved to be successful. After a description of our hyperspectral data set acquired over Toulouse, an analysis is conducted on several existing and accessible spectral databases. Then, the chosen methods are presented. They include extraction, fusion and classification methods, which are then applied on our dataset synthetized at different spatial resolution to evaluate the benefits and the complementarity of hyperspectral imagery in comparison with other traditional sensors. Some specific applications are investigated of interest for urban planners: impervious soil map, vegetation species cartography and detection of solar panels. Finally, discussion and perspectives are presented. Numéro de notice : A2022-941 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Autre URL associée : Hal Thématique : IMAGERIE/URBANISME Nature : Article nature-HAL : ArtAvecCL-RevueNat DOI : 10.52638/rfpt.2022.589 Date de publication en ligne : 22/12/2022 En ligne : https://dx.doi.org/10.52638/rfpt.2022.589 Format de la ressource électronique : URL Article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=102831 [article]

Predicting vegetation stratum occupancy from airborne LiDAR data with deep learning / Ekaterina Kalinicheva in International journal of applied Earth observation and geoinformation, vol 112 (August 2022)    

Public [article]  inInternational journal of applied Earth observation and geoinformation > vol 112 (August 2022) . - n° 102863 Titre : Predicting vegetation stratum occupancy from airborne LiDAR data with deep learning Type de document : Article/Communication Auteurs : Ekaterina Kalinicheva , Auteur ; Loïc Landrieu , Auteur ; Clément Mallet , Auteur ; Nesrine Chehata , Auteur Année de publication : 2022 Projets : TOSCA-FRISBEE / Weber, Christiane Article en page(s) : n° 102863 Note générale : bibliographie This study has been co-funded by CNES (TOSCA FRISBEE Project, convention no200769/00) and CONFETTI Project (Nouvelle Aquitaine Region project, France). Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] apprentissage profond [Termes IGN] apprentissage semi-dirigé [Termes IGN] données lidar [Termes IGN] données localisées 3D [Termes IGN] parcelle agricole [Termes IGN] régression [Termes IGN] semis de points [Termes IGN] strate végétale Résumé : (auteur) We propose a new deep learning-based method for estimating the occupancy of vegetation strata from airborne 3D LiDAR point clouds. Our model predicts rasterized occupancy maps for three vegetation strata corresponding to lower, medium, and higher cover. Our weakly-supervised training scheme allows our network to only be supervised with vegetation occupancy values aggregated over cylindrical plots containing thousands of points. Such ground truth is easier to produce than pixel-wise or point-wise annotations. Our method outperforms handcrafted and deep learning baselines in terms of precision by up to 30%, while simultaneously providing visual and interpretable predictions. We provide an open-source implementation along with a dataset of 199 agricultural plots to train and evaluate weakly supervised occupancy regression algorithms. Numéro de notice : A2022-578 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Thématique : IMAGERIE/INFORMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.jag.2022.102863 Date de publication en ligne : 19/07/2022 En ligne : https://doi.org/10.1016/j.jag.2022.102863 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99425 [article]

Documents numériques

peut être téléchargé Predicting vegetation stratum ... - pdf auteur Adobe Acrobat PDF

Classification of vegetation classes by using time series of Sentinel-2 images for large scale mapping in Cameroon / Hermann Tagne in ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol V-3-2022 (2022 edition)  

Public [article]  inISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences > vol V-3-2022 (2022 edition) . - pp 673 - 680 Titre : Classification of vegetation classes by using time series of Sentinel-2 images for large scale mapping in Cameroon Type de document : Article/Communication Auteurs : Hermann Tagne, Auteur ; Arnaud Le Bris , Auteur ; David Monkam, Auteur ; Clément Mallet , Auteur Année de publication : 2022 Projets : TOSCA Parcelle / Le Bris, Arnaud Article en page(s) : pp 673 - 680 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] Cameroun [Termes IGN] carte de la végétation [Termes IGN] classification par forêts d'arbres décisionnels [Termes IGN] fusion d'images [Termes IGN] image Sentinel-MSI [Termes IGN] occupation du sol [Termes IGN] série temporelle Résumé : (auteur) Sentinel-2 satellites provide dense image time series exhibiting high spectral, spatial and temporal resolutions. These images are in particular of utter interest for Land-Cover (LC) mapping at large scales. LC maps can now be computed on a yearly basis at the scale of a country with efficient supervised classifiers, assuming suitable training data are available. However, the efficient exploitation of large amount of Sentinel-2 imagery still remain challenging on unexplored areas where state-of-the-art classifiers are prone to fail. This paper focuses on Land-Cover mapping over Cameroon for the purpose of updating the Very High Resolution national topographic geodatabase. The ι2 framework is adopted and tested for the specificity of the country. Here, experiments focus on generic vegetation classes (five) which enables providing robust focusing masks for higher resolution classifications. Two strategies are compared: (i) a LC map is calculated out of a year long time series and (ii) monthly LC maps are generated and merged into a single yearly map. Satisfactory accuracy scores are obtained (>94% in Overall Accuracy), allowing to provide a first step towards finer-grained map retrieval. Numéro de notice : A2022-426 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.5194/isprs-annals-V-3-2022-673-2022 Date de publication en ligne : 18/05/2022 En ligne : https://doi.org/10.5194/isprs-annals-V-3-2022-673-2022 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100731 [article]

Preface: The 2022 edition of the XXIVth ISPRS congress / Loïc Landrieu in ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol V-3-2022 (2022 edition)  

Public [article]  inISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences > vol V-3-2022 (2022 edition) . - 5 p. Titre : Preface: The 2022 edition of the XXIVth ISPRS congress Type de document : Article/Communication Auteurs : Loïc Landrieu , Auteur ; Ewelina Rupnik , Auteur ; Sander J. Oude Elberink, Auteur ; Clément Mallet , Auteur ; Nicolas Paparoditis , Auteur Année de publication : 2022 Projets : 1-Pas de projet / Le Bris, Arnaud Article en page(s) : 5 p. Langues : Anglais (eng) Résumé : (auteur) [introduction] We report key elements and figures related to the proceedings of the 2022 edition of the XXIVth ISPRS Congress. Despite the uncertainty and turmoil caused by the COVID-19 pandemic, the 2022 edition of the Congress is going to take place in person in Nice (France, 6-11 June 2022) and online, with a significant expected turnout: 1,600 participants have registered including 300 online participation as of April 25. The dynamic and unpredictable global health situation makes it difficult to predict participation. Numéro de notice : A2022-339 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.5194/isprs-annals-V-3-2022-1-2022 Date de publication en ligne : 17/05/2022 En ligne : https://doi.org/10.5194/isprs-annals-V-3-2022-1-2022 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100721 [article]

Editorial : Changes in the team of associate editors / Qihao Weng in ISPRS Journal of photogrammetry and remote sensing, vol 185 (March 2022)  

Permalink

BuyTheDips : PathLoss for improved topology-preserving deep learning-based image segmentation / Minh On Vu Ngoc (2022)  

Permalink

Deep-learning based multiple land-cover map translation / Luc Baudoux (2022)  

Permalink

Learning spatio-temporal representations of satellite time series for large-scale crop mapping / Vivien Sainte Fare Garnot (2022)  

Permalink

Monitoring grassland dynamics by exploiting multi-modal satellite image time series / Anatol Garioud (2022)  

Permalink

Multi-layer modeling of dense vegetation from aerial LiDAR scans / Ekaterina Kalinicheva (2022)  

Permalink

Preface: the 2022 edition of the XXIVth ISPRS congress / Loïc Landrieu (2022)  

Permalink

SenRVM: A multi-modal deep learning regression methodology for continuous vegetation monitoring with dense temporal NDVI time series / Anatol Garioud (2022) 

Permalink

The use of volunteer geographic information for producing and maintaining authoritative land use and land cover data / Ana-Maria Olteanu-Raimond (2022)  

Permalink

Fast estimation for robust supervised classification with mixture models / Erwan Giry Fouquet in Pattern recognition letters, vol 152 (December 2021)  

Permalink