Centre dedocumentation
scientifique

Development of the GLASS 250-m leaf area index product (version 6) from MODIS data using the bidirectional LSTM deep learning model / Han Ma in Remote sensing of environment, vol 273 (May 2022)  

Public [article]  inRemote sensing of environment > vol 273 (May 2022) . - n° 112985 Titre : Development of the GLASS 250-m leaf area index product (version 6) from MODIS data using the bidirectional LSTM deep learning model Type de document : Article/Communication Auteurs : Han Ma, Auteur ; Shunlin Liang, Auteur Année de publication : 2022 Article en page(s) : n° 112985 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] apprentissage profond [Termes IGN] cohérence temporelle [Termes IGN] image Terra-MODIS [Termes IGN] Leaf Area Index [Termes IGN] réflectance de surface [Termes IGN] régression [Termes IGN] série temporelle [Termes IGN] surveillance de la végétation Résumé : (auteur) Leaf area index (LAI) is a terrestrial essential climate variable that is required in a variety of ecosystem and climate models. The Global LAnd Surface Satellite (GLASS) LAI product has been widely used, but its current version (V5) from Moderate Resolution Imaging Spectroradiometer (MODIS) data has several limitations, such as frequent temporal fluctuation, large data gaps, high dependence on the quality of surface reflectance, and low computational efficiency. To address these issues, this paper presents a deep learning model to generate a new version of the LAI product (V6) at 250-m resolution from MODIS data from 2000 onward. Unlike most existing algorithms that estimate one LAI value at one time for each pixel, this model estimates LAI for 2 years simultaneously. Three widely used LAI products (MODIS C6, GLASS V5, and PROBA-V V1) are used to generate global representative time-series LAI training samples using K-means clustering analysis and least difference criteria. We explore four machine learning models, the general regression neural network (GRNN), long short-term memory (LSTM), gated recurrent unit (GRU), and Bidirectional LSTM (Bi-LSTM), and identify Bi-LSTM as the best model for product generation. This new product is directly validated using 79 high-resolution LAI reference maps from three in situ observation networks. The results show that GLASS V6 LAI achieves higher accuracy, with a root mean square (RMSE) of 0.92 at 250 m and 0.86 at 500 m, while the RMSE is 0.98 for PROBA-V at 300 m, 1.08 for GLASS V5, and 0.95 for MODIS C6 both at 500 m. Spatial and temporal consistency analyses also demonstrate that the GLASS V6 LAI product is more spatiotemporally continuous and has higher quality in terms of presenting more realistic temporal LAI dynamics when the surface reflectance is absent for a long period owing to persistent cloud/aerosol contaminations. The results indicate that the new Bi-LSTM deep learning model runs significantly faster than the GLASS V5 algorithm, avoids the reconstruction of surface reflectance data, and is resistant to the noises (cloud and snow contamination) or missing values contained in surface reflectance than other methods, as the Bi-LSTM can effectively extract information across the entire time series of surface reflectance rather than a single time point. To our knowledge, this is the first global time-series LAI product at the 250-m spatial resolution that is freely available to the public (www.geodata.cn and www.glass.umd.edu). Numéro de notice : A2022-284 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article DOI : 10.1016/j.rse.2022.112985 Date de publication en ligne : 10/03/2022 En ligne : https://doi.org/10.1016/j.rse.2022.112985 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100303 [article]

Spatiotemporal fusion modelling using STARFM: Examples of Landsat 8 and Sentinel-2 NDVI in Bavaria / Maninder Singh Dhillon in Remote sensing, vol 14 n° 3 (February-1 2022)  

Public [article]  inRemote sensing > vol 14 n° 3 (February-1 2022) . - n° 677 Titre : Spatiotemporal fusion modelling using STARFM: Examples of Landsat 8 and Sentinel-2 NDVI in Bavaria Type de document : Article/Communication Auteurs : Maninder Singh Dhillon, Auteur ; Thorsten Dahms, Auteur ; Carina Kübert-Flock, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : n° 677 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] Bavière (Allemagne) [Termes IGN] carte d'occupation du sol [Termes IGN] fusion de données [Termes IGN] image Landsat-8 [Termes IGN] image Sentinel-MSI [Termes IGN] image Terra-MODIS [Termes IGN] Normalized Difference Vegetation Index [Termes IGN] pouvoir de résolution géométrique [Termes IGN] réflectance [Termes IGN] surveillance de la végétation [Termes IGN] utilisation du sol Résumé : (auteur) The increasing availability and variety of global satellite products provide a new level of data with different spatial, temporal, and spectral resolutions; however, identifying the most suited resolution for a specific application consumes increasingly more time and computation effort. The region’s cloud coverage additionally influences the choice of the best trade-off between spatial and temporal resolution, and different pixel sizes of remote sensing (RS) data may hinder the accurate monitoring of different land cover (LC) classes such as agriculture, forest, grassland, water, urban, and natural-seminatural. To investigate the importance of RS data for these LC classes, the present study fuses NDVIs of two high spatial resolution data (high pair) (Landsat (30 m, 16 days; L) and Sentinel-2 (10 m, 5–6 days; S), with four low spatial resolution data (low pair) (MOD13Q1 (250 m, 16 days), MCD43A4 (500 m, one day), MOD09GQ (250 m, one-day), and MOD09Q1 (250 m, eight day)) using the spatial and temporal adaptive reflectance fusion model (STARFM), which fills regions’ cloud or shadow gaps without losing spatial information. These eight synthetic NDVI STARFM products (2: high pair multiply 4: low pair) offer a spatial resolution of 10 or 30 m and temporal resolution of 1, 8, or 16 days for the entire state of Bavaria (Germany) in 2019. Due to their higher revisit frequency and more cloud and shadow-free scenes (S = 13, L = 9), Sentinel-2 (overall R2 = 0.71, and RMSE = 0.11) synthetic NDVI products provide more accurate results than Landsat (overall R2 = 0.61, and RMSE = 0.13). Likewise, for the agriculture class, synthetic products obtained using Sentinel-2 resulted in higher accuracy than Landsat except for L-MOD13Q1 (R2 = 0.62, RMSE = 0.11), resulting in similar accuracy preciseness as S-MOD13Q1 (R2 = 0.68, RMSE = 0.13). Similarly, comparing L-MOD13Q1 (R2 = 0.60, RMSE = 0.05) and S-MOD13Q1 (R2 = 0.52, RMSE = 0.09) for the forest class, the former resulted in higher accuracy and precision than the latter. Conclusively, both L-MOD13Q1 and S-MOD13Q1 are suitable for agricultural and forest monitoring; however, the spatial resolution of 30 m and low storage capacity makes L-MOD13Q1 more prominent and faster than that of S-MOD13Q1 with the 10-m spatial resolution. Numéro de notice : A2022-124 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.3390/rs14030677 Date de publication en ligne : 31/01/2022 En ligne : https://doi.org/10.3390/rs14030677 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99687 [article]

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

Public Titre : SenRVM: A multi-modal deep learning regression methodology for continuous vegetation monitoring with dense temporal NDVI time series Type de document : Article/Communication Auteurs : Anatol Garioud , Auteur ; Silvia Valero, Auteur ; Clément Mallet , Auteur Editeur : Saint-Mandé : Institut national de l'information géographique et forestière - IGN Année de publication : 2022 Conférence : LPS 2022, ESA Living Planet Symposium 22/05/2022 27/05/2022 Bonn Allemagne programme sans actes Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] analyse d'image orientée objet [Termes IGN] dynamique de la végétation [Termes IGN] image Sentinel-MSI [Termes IGN] Normalized Difference Vegetation Index [Termes IGN] phénologie [Termes IGN] régression [Termes IGN] série temporelle [Termes IGN] surveillance de la végétation Résumé : (auteur) The Earth's biosphere and the phenology of vegetation are at the heart of climatic, economic and social concerns. Human activities have led to a significant degradation of ecosystem services (e.g. carbon sequestration, biodiversity, water quality, flood, and erosion regulation) provided by various extensive ecosystems such as forests, grasslands or crops. A key parameter for relevant climate modeling, public policy implementations or commercial applications is the temporal resolution at which vegetation is observed. As a tool providing synoptic and regular coverage of Earth’s surfaces, satellite Earth Observation has been increasingly adopted, among others, for estimating biomass, yields, modeling different fluxes or detecting changes. Optical images have been historically used for vegetation monitoring, considering their efficient discrimination of phenomena related to photosynthetic activity. To deal with missing data due to clouds, many interpolation strategies integrating one or more optical sensors have been developed. Most of these strategies are based on trend modelling that does not reflect the real evolution of the vegetation cover in many cases (sudden climatic impact, man-made effects). As a result, data that may be weeks or months apart are often interpolated on areas suffering from high cloud cover. Copernicus Sentinels provide new opportunities and unprecedented observations for the monitoring of vegetation’s dynamics. In particular, concordant optical and SAR data sets provided by the Sentinel-1 and 2 satellites open the door to new multi-sensor methodologies aiming at the reconstruction of missing information. Taking into account the still numerous non-cloudy observations provided by the Sentinel-2 satellites, a deep learning regression methodology, namely the Sentinels Regression for Vegetation Monitoring (SenRVM), has been developed. Its goal is the translation of SAR features acquired regardless of the climatic conditions into NDVI. The developed architecture integrates several deep learning architectures such as Multilayer Perceptron and Recurrent Neural Networks. The SenRVM regression strategy proposes the integration of auxiliary data such as climatic and topographic features. This allows accurate NDVI time series to be predicted by minimizing effects exogenous to the vegetation’s phenology through SAR acquisitions contextualization. Object-oriented analysis of the results is carried out on large scale areas for various vegetation types with distinct phenologies (grasslands, crops and forests). The results are analyzed by taking into account spatial and temporal aspects or with an ablation study of the Network’s inputs. The proposed approach is further compared with traditional interpolation methods exploiting monomodal (Whittaker smoothing, linear weighted interpolation) or multimodal (Random Forest, Gaussian Regression Processes, single Multilayer Perceptron) features. The potential of high-temporal NDVI time series obtained by the SenRVM method for several vegetation-related applications is subsequently illustrated. In particular, the interest of the obtained time series to observe the phenology and its associated parameters of the three main vegetation classes is presented. Numéro de notice : C2022-011 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Thématique : IMAGERIE/INFORMATIQUE Nature : Poster nature-HAL : Poster-avec-CL DOI : sans Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100786

Documents numériques

peut être téléchargé SenRVM - poster Adobe Acrobat PDF

Multi-model estimation of forest canopy closure by using red edge bands based on Sentinel-2 images / Yiying Hua in Forests, vol 12 n° 12 (December 2021)  

Public [article]  inForests > vol 12 n° 12 (December 2021) . - n° 1768 Titre : Multi-model estimation of forest canopy closure by using red edge bands based on Sentinel-2 images Type de document : Article/Communication Auteurs : Yiying Hua, Auteur ; Xuesheng Zhao, Auteur Année de publication : 2021 Article en page(s) : n° 1768 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] apprentissage automatique [Termes IGN] bande infrarouge [Termes IGN] coefficient de corrélation [Termes IGN] couvert forestier [Termes IGN] détection de contours [Termes IGN] image multibande [Termes IGN] image Sentinel-MSI [Termes IGN] indice de végétation [Termes IGN] modèle numérique de surface de la canopée [Termes IGN] modèle statistique [Termes IGN] Mongolie intérieure (Chine) [Termes IGN] régression [Termes IGN] surveillance de la végétation Résumé : (auteur) In remote sensing, red edge bands are important indicators for monitoring vegetation growth. To examine the application potential of red edge bands in forest canopy closure estimation, three types of commonly used models—empirical statistical models (multiple stepwise regression (MSR)), machine learning models (back propagation neural network (BPNN)) and physical models (Li–Strahler geometric-optical (Li–Strahler GO) models)—were constructed and verified based on Sentinel-2 data, DEM data and measured data. In addition, we set up a comparative experiment without red edge bands. The relative error (ER) values of the BPNN model, MSR model, and Li–Strahler GO model with red edge bands were 16.97%, 20.76% and 24.83%, respectively. The validation accuracy measures of these models were higher than those of comparison models. For comparative experiments, the ER values of the MSR, Li–Strahler GO and BPNN models were increased by 13.07%, 4% and 1.22%, respectively. The experimental results demonstrate that red edge bands can effectively improve the accuracy of forest canopy closure estimation models to varying degrees. These findings provide a reference for modeling and estimating forest canopy closure using red edge bands based on Sentinel-2 images. Numéro de notice : A2021-125 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article DOI : 10.3390/f12121768 Date de publication en ligne : 14/12/2021 En ligne : https://doi.org/10.3390/f12121768 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99318 [article]

Recurrent-based regression of Sentinel time series for continuous vegetation monitoring / Anatol Garioud in Remote sensing of environment, vol 263 (15 September 2021)  

Public [article]  inRemote sensing of environment > vol 263 (15 September 2021) . - n° 112419 Titre : Recurrent-based regression of Sentinel time series for continuous vegetation monitoring Type de document : Article/Communication Auteurs : Anatol Garioud , Auteur ; Silvia Valero, Auteur ; Sébastien Giordano , Auteur ; Clément Mallet , Auteur Année de publication : 2021 Projets : 3-projet - voir note / Article en page(s) : n° 112419 Note générale : bibliographie This work is funded by the Agence de la transition écologique (ADEME) and the Centre National d'Études Spatiales (CNES). Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] classification par réseau neuronal [Termes IGN] classification par réseau neuronal récurrent [Termes IGN] fusion d'images [Termes IGN] image Sentinel-MSI [Termes IGN] image Sentinel-SAR [Termes IGN] Normalized Difference Vegetation Index [Termes IGN] régression [Termes IGN] série temporelle [Termes IGN] surveillance de la végétation Résumé : (auteur) Dense time series of optical satellite imagery describing vegetation activity provide essential information for the efficient and regular monitoring of vegetation. Nevertheless, the temporal resolution of optical sensors is strongly affected by cloud cover, resulting in significant missing information. The use of complementary acquisitions, such as Synthetic Aperture Radar (SAR) data, opens the door to the development of new multi-sensor methodologies aiming at the reconstruction of missing information. However, the joint exploitation of new radar and optical missions, such as the Sentinel, raises new challenges given the different nature and response of the two data sources. In this work, the SenRVM methodology is proposed as a new multi-sensor approach to regress SAR time series towards Normalized Difference Vegetation Index (NDVI). A deep Recurrent Neural Network architecture which integrates SAR acquisitions and ancillary data is adopted. The regression task permits a continuous optical temporal resolution of 6 days. Multiple experiments are carried out to assess the SenRVM framework by studying two large-scale areas in France. Through an extensive interpretation of the results, SenRVM is evaluated on three main vegetation types (grasslands, crops, and forests). High accurate results (R2 > 0.83 and MAE Numéro de notice : A2021-499 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Thématique : FORET/IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.rse.2021.112419 Date de publication en ligne : 25/06/2021 En ligne : https://doi.org/10.1016/j.rse.2021.112419 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=98004 [article]

Evaluating the performance of hyperspectral leaf reflectance to detect water stress and estimation of photosynthetic capacities / Jingjing Zhou in Remote sensing, vol 13 n° 11 (June-1 2021)  

Permalink

Rapid ecosystem change at the southern limit of the Canadian Arctic, Torngat Mountains National Park / Emma L. Davis in Remote sensing, vol 13 n° 11 (June-1 2021)  

Permalink

Mask R-CNN and OBIA fusion improves the segmentation of scattered vegetation in very high-resolution optical sensors / Emilio Guirado in Sensors, vol 21 n° 1 (January 2021)  

Permalink

Suivi des vignes par télédétection de proximité : le deep learning au service de l’agriculture de précision / Sami Beniaouf (2021) 

Permalink

Near-real time forecasting and change detection for an open ecosystem with complex natural dynamics / Jasper A. Slingsby in ISPRS Journal of photogrammetry and remote sensing, vol 166 (August 2020)  

Permalink

Towards a semi-automated mapping of Australia native invasive alien Acacia trees using Sentinel-2 and radiative transfer models in South Africa / Cecilia Masemola in ISPRS Journal of photogrammetry and remote sensing, vol 166 (August 2020)  

Permalink

Evaluating techniques for mapping island vegetation from unmanned aerial vehicle (UAV) images: Pixel classification, visual interpretation and machine learning approaches / S.M. Hamylton in International journal of applied Earth observation and geoinformation, vol 89 (July 2020)  

Permalink

Mapping the condition of macadamia tree crops using multi-spectral UAV and WorldView-3 imagery / Kasper Johansen in ISPRS Journal of photogrammetry and remote sensing, vol 165 (July 2020)  

Permalink

Wheat leaf area index retrieval using RISAT-1 hybrid polarized SAR data / Thota Sivasankar in Geocarto international, Vol 35 n° 8 ([01/06/2020])  

Permalink

Conterminous United States land cover change patterns 2001–2016 from the 2016 National Land Cover Database / Collin Homer in ISPRS Journal of photogrammetry and remote sensing, vol 162 (April 2020)  

Permalink