Détail de l'auteur
Auteur Ho Tong Minh Dinh |
Documents disponibles écrits par cet auteur (4)
Ajouter le résultat dans votre panier Affiner la recherche Interroger des sources externesCombining Sentinel-1 and Sentinel-2 Satellite image time series for land cover mapping via a multi-source deep learning architecture / Dino Lenco in ISPRS Journal of photogrammetry and remote sensing, Vol 158 (December 2019)
Titre : Combining Sentinel-1 and Sentinel-2 Satellite image time series for land cover mapping via a multi-source deep learning architecture Type de document : Article/Communication Auteurs : Dino Lenco, Auteur ; Roberto Interdonato, Auteur ; Raffaele Gaetano, Auteur ; Ho Tong Minh Dinh, Auteur Année de publication : 2019 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image mixte
[Termes IGN] apprentissage profond
[Termes IGN] Burkina Faso
[Termes IGN] carte de la végétation
[Termes IGN] classification par forêts d'arbres décisionnels
[Termes IGN] classification par réseau neuronal convolutif
[Termes IGN] fusion d'images
[Termes IGN] image à haute résolution
[Termes IGN] image multibande
[Termes IGN] image radar moirée
[Termes IGN] image Sentinel-MSI
[Termes IGN] image Sentinel-SAR
[Termes IGN] occupation du sol
[Termes IGN] Réunion, île de la
[Termes IGN] série temporelle
[Termes IGN] utilisation du solRésumé : (auteur) The huge amount of data currently produced by modern Earth Observation (EO) missions has allowed for the design of advanced machine learning techniques able to support complex Land Use/Land Cover (LULC) mapping tasks. The Copernicus programme developed by the European Space Agency provides, with missions such as Sentinel-1 (S1) and Sentinel-2 (S2), radar and optical (multi-spectral) imagery, respectively, at 10 m spatial resolution with revisit time around 5 days. Such high temporal resolution allows to collect Satellite Image Time Series (SITS) that support a plethora of Earth surface monitoring tasks. How to effectively combine the complementary information provided by such sensors remains an open problem in the remote sensing field. In this work, we propose a deep learning architecture to combine information coming from S1 and S2 time series, namely TWINNS (TWIn Neural Networks for Sentinel data), able to discover spatial and temporal dependencies in both types of SITS. The proposed architecture is devised to boost the land cover classification task by leveraging two levels of complementarity, i.e., the interplay between radar and optical SITS as well as the synergy between spatial and temporal dependencies. Experiments carried out on two study sites characterized by different land cover characteristics (i.e., the Koumbia site in Burkina Faso and Reunion Island, a overseas department of France in the Indian Ocean), demonstrate the significance of our proposal. Numéro de notice : A2019-544 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2019.09.016 Date de publication en ligne : 27/09/2019 En ligne : https://doi.org/10.1016/j.isprsjprs.2019.09.016 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=94186
in ISPRS Journal of photogrammetry and remote sensing > Vol 158 (December 2019)[article]Exemplaires(3)
Code-barres Cote Support Localisation Section Disponibilité 081-2019121 RAB Revue Centre de documentation En réserve L003 Disponible 081-2019123 DEP-RECP Revue LASTIG Dépôt en unité Exclu du prêt 081-2019122 DEP-RECF Revue Nancy Dépôt en unité Exclu du prêt
Titre : Aboveground biomass mapping in French Guiana by combining remote sensing, forest inventories and environmental data Type de document : Article/Communication Auteurs : Ibrahim Fayad, Auteur ; Nicolas Baghdadi, Auteur ; Stéphane Guitet Année de publication : 2016 Article en page(s) : pp 502 - 514 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection
[Termes IGN] biomasse aérienne
[Termes IGN] données environnementales
[Termes IGN] données ICEsat
[Termes IGN] forêt tropicale
[Termes IGN] Guyane (département français)
[Termes IGN] inventaire forestier national (données France)
[Termes IGN] régressionRésumé : (auteur) Mapping forest aboveground biomass (AGB) has become an important task, particularly for the reporting of carbon stocks and changes. AGB can be mapped using synthetic aperture radar data (SAR) or passive optical data. However, these data are insensitive to high AGB levels (>150 Mg/ha, and >300 Mg/ha for P-band), which are commonly found in tropical forests. Studies have mapped the rough variations in AGB by combining optical and environmental data at regional and global scales. Nevertheless, these maps cannot represent local variations in AGB in tropical forests. In this paper, we hypothesize that the problem of misrepresenting local variations in AGB and AGB estimation with good precision occurs because of both methodological limits (signal saturation or dilution bias) and a lack of adequate calibration data in this range of AGB values. We test this hypothesis by developing a calibrated regression model to predict variations in high AGB values (mean >300 Mg/ha) in French Guiana by a methodological approach for spatial extrapolation with data from the optical geoscience laser altimeter system (GLAS), forest inventories, radar, optics, and environmental variables for spatial inter- and extrapolation. Given their higher point count, GLAS data allow a wider coverage of AGB values. We find that the metrics from GLAS footprints are correlated with field AGB estimations (R2 = 0.54, RMSE = 48.3 Mg/ha) with no bias for high values. First, predictive models, including remote-sensing, environmental variables and spatial correlation functions, allow us to obtain “wall-to-wall” AGB maps over French Guiana with an RMSE for the in situ AGB estimates of ∼50 Mg/ha and R2 = 0.66 at a 1-km grid size. We conclude that a calibrated regression model based on GLAS with dependent environmental data can produce good AGB predictions even for high AGB values if the calibration data fit the AGB range. We also demonstrate that small temporal and spatial mismatches between field data and GLAS footprints are not a problem for regional and global calibrated regression models because field data aim to predict large and deep tendencies in AGB variations from environmental gradients and do not aim to represent high but stochastic and temporally limited variations from forest dynamics. Thus, we advocate including a greater variety of data, even if less precise and shifted, to better represent high AGB values in global models and to improve the fitting of these models for high values. Numéro de notice : A2016--202 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article DOI : 10.1016/j.jag.2016.07.015 Date de publication en ligne : 01/08/2016 En ligne : https://doi.org/10.1016/j.jag.2016.07.015 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96037
in International journal of applied Earth observation and geoinformation > vol 52 (October 2016) . - pp 502 - 514[article]
Titre : Capabilities of BIOMASS tomography for investigating tropical forests Type de document : Article/Communication Auteurs : Ho Tong Minh Dinh, Auteur ; Stefano Tebaldini, Auteur ; Fabio Rocca, Auteur ; et al., Auteur Année de publication : 2015 Article en page(s) : pp 965 - 975 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection
[Termes IGN] bande P
[Termes IGN] Biomass
[Termes IGN] biomasse (combustible)
[Termes IGN] dégradation d'image
[Termes IGN] forêt équatoriale
[Termes IGN] Guyane française
[Termes IGN] hauteur des arbres
[Termes IGN] image aérienne
[Termes IGN] image radar moirée
[Termes IGN] tomographie radarRésumé : (Auteur) The objective of this paper is to provide a better understanding of the capabilities of the BIOMASS tomography concerning the retrieval of forest biomass and height in tropical areas. The analysis presented in this paper is carried out on airborne data acquired by Office National d'Etudes et de Recherches Aérospatiales (ONERA) over the site of Paracou, French Guiana, during the European Space Agency campaign TropiSAR. This high-resolution data set (125-MHz bandwidth) was reprocessed in order to generate a new data stack consistent with BIOMASS as for the bandwidth (6 MHz) and the azimuth resolution (about 12 m). To do this, two different processing approaches have been considered. One approach consisted of degrading the resolution of the airborne data through the linear filtering of raw data, followed by standard SAR processing. The other approach consisted of recovering the 3-D distribution of the scatterers at a high resolution, which was then reprojected onto the BIOMASS geometry. The latter procedure allows us to obtain a data stack that is the most realistic emulation of BIOMASS imaging capabilities. In both approaches, neither ionospheric disturbances nor temporal decorrelation has been considered. The connection to the forest biomass has been examined in both cases by investigating the correlation between the backscatter at different forest heights and the above-ground biomass (AGB) values from in situ data. As expected, the reduction of the system bandwidth to 6 MHz resulted in significant vertical resolution losses compared with the original airborne data. Nevertheless, it was possible to retrieve the forest height to within an accuracy of better than 4 m, whereas the backscattered power at the volume height (30 m above the ground) exhibited a correlation higher than 0.8 with the in situ data and no bias phenomena over the AGB values ranging from 250 to 450 t/ha. Numéro de notice : A2015-108 Affiliation des auteurs : non IGN Thématique : FORET/GEOMATIQUE/IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2014.2331142 En ligne : https://doi.org/10.1109/TGRS.2014.2331142 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=75626
in IEEE Transactions on geoscience and remote sensing > vol 53 n° 2 (February 2015) . - pp 965 - 975[article]Exemplaires(1)
Code-barres Cote Support Localisation Section Disponibilité 065-2015021 RAB Revue Centre de documentation En réserve L003 Disponible
Titre : Ground-based array for tomographic imaging of the tropical forest in P-band Type de document : Article/Communication Auteurs : Ho Tong Minh Dinh, Auteur ; Stefano Tebaldini, Auteur ; Fabio Rocca, Auteur ; et al., Auteur Année de publication : 2013 Article en page(s) : pp 4460 - 4472 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image radar et applications
[Termes IGN] bande P
[Termes IGN] flore locale
[Termes IGN] forêt tropicale
[Termes IGN] Guyane (département français)
[Termes IGN] polarisation
[Termes IGN] tomographieRésumé : (Auteur) In this paper, we discuss the design concepts and preliminary results relating to the European Space Agency's ground-based campaign TropiScat, whose main goal is to evaluate temporal coherence at P-band in a tropical forest in quad-polarization, considering temporal lags ranging from hours to months and at different heights within the vegetation layer. The experiment has been successfully set up and operated since October 2011 at the Paracou field station, French Guiana, where the equipment was installed on top of the 55-m high Guyaflux Tower to illuminate the forest below. The system consists of a vector network analyzer connected to 20 antennas through a switchbox, which allows the use of any of them either as a transmitter or as a receiver. Vertical imaging and fully polarimetric capabilities are achieved by operating the 20 antennas in a multistatic fashion, resulting in an equivalent monostatic array consisting of 15 phase centers displaced along the vertical direction in each polarization. Such a design allows unambiguous imaging of the vegetation while yielding a minimum distance between nearby antennas on the order of 0.8 m, so as to minimize coupling effects. The equipment allows the gathering of signals with the tomographic array within a few minutes, resulting in the possibility to produce a tomographic image of the forest with a temporal sampling of 15 min. System calibration and validation was performed by employing a 2-m trihedral reflector and a rotating dihedral reflector. This allowed the evaluation of the system pulse response in all polarizations and also assessment of the extent of tower motions. As a result, tomographic images have been generated from 500 (P-band) to 900 MHz in all polarizations. Results from real data acquired in Fall 2011 confirm the feasibility of carrying out reliable coherence measurements for the whole duration of the campaign. Numéro de notice : A2013-418 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2013.2246795 En ligne : https://doi.org/10.1109/TGRS.2013.2246795 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=32556
in IEEE Transactions on geoscience and remote sensing > vol 51 n° 8 (August 2013) . - pp 4460 - 4472[article]Exemplaires(1)
Code-barres Cote Support Localisation Section Disponibilité 065-2013081 RAB Revue Centre de documentation En réserve L003 Disponible
