Centre dedocumentation
scientifique

Automatic detection and classification of low-level orographic precipitation processes from space-borne radars using machine learning / Malarvizhi Arulraj in Remote sensing of environment, vol 257 (May 2021)  

Public [article]  inRemote sensing of environment > vol 257 (May 2021) . - n° 112355 Titre : Automatic detection and classification of low-level orographic precipitation processes from space-borne radars using machine learning Type de document : Article/Communication Auteurs : Malarvizhi Arulraj, Auteur ; Ana P. Baros, Auteur Année de publication : 2021 Article en page(s) : n° 112355 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image radar et applications [Termes descripteurs IGN] Appalaches [Termes descripteurs IGN] apprentissage automatique [Termes descripteurs IGN] bande S [Termes descripteurs IGN] classification automatique [Termes descripteurs IGN] classification barycentrique [Termes descripteurs IGN] classification par forêts aléatoires [Termes descripteurs IGN] classification par réseau neuronal convolutif [Termes descripteurs IGN] image GPM [Termes descripteurs IGN] orographie [Termes descripteurs IGN] précipitation Résumé : (auteur) Ground-clutter is a significant cause of missed-detection and underestimation of precipitation in complex terrain from space-based radars such as the Global Precipitation Measurement Mission (GPM) Dual-frequency Precipitation Radar (DPR). This research proposes an Artificial Intelligence (AI) framework consisting of a precipitation detection model (PDM) and a precipitation regime classification model (PCM) to improve orographic precipitation retrievals from GPM-DPR using machine learning. The PDM is a Random Forest Classifier using GPM Microwave Imager (GMI) calibrated brightness temperatures (Tbs) and low-level precipitation mixing ratios from the High-Resolution Rapid Refresh (HRRR) analysis as inputs. The PCM is a Convolutional Neural Network that predicts the precipitation regime class, defined independently based on quantitative features of ground-based radar reflectivity profiles, using GPM DPR Ku-band (Ku-PR) reflectivity profiles and GMI Tbs. The AI framework is demonstrated for warm-season precipitation in the Southern Appalachian Mountains over. Numéro de notice : A2021-279 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1016/j.rse.2021.112355 date de publication en ligne : 19/02/2021 En ligne : https://doi.org/10.1016/j.rse.2021.112355 Format de la ressource électronique : url article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97372 [article]

Evaluating P-Band TomoSAR for biomass retrieval in boreal forest / Erik Blomberg in IEEE Transactions on geoscience and remote sensing, vol 59 n° 5 (May 2021)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 59 n° 5 (May 2021) . - pp 3793 - 3804 Titre : Evaluating P-Band TomoSAR for biomass retrieval in boreal forest Type de document : Article/Communication Auteurs : Erik Blomberg, Auteur ; Lars M.H. Ulander, Auteur ; Stefano Tebaldini, Auteur ; Laurent Ferro-Famil, Auteur Année de publication : 2021 Article en page(s) : pp 3793 - 3804 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes descripteurs IGN] bande P [Termes descripteurs IGN] biomasse forestière [Termes descripteurs IGN] données TomoSAR [Termes descripteurs IGN] forêt boréale Résumé : (Auteur) P-band synthetic aperture radar (SAR) is sensitive to above-ground biomass (AGB) but retrieval accuracy has been shown to deteriorate in topographic areas. In boreal forest, the signal penetrates through the canopy to interact with the ground producing variations in backscatter depending on ground topography, forest structure, and soil moisture. Tomographic processing of multiple SAR images Tomographic SAR (TomoSAR) provides information about the vertical backscatter distribution. This article evaluates the use of P-band TomoSAR data to improve AGB retrievals from backscattered intensity by suppressing the backscattered signal from the ground. This approach can be used even when the tomographic resolution is insufficient to resolve the vertical backscatter profile. The analysis is based on P-band data from two campaigns: BioSAR-1 (2007) in Remingstorp, southern Sweden, and BioSAR-2 (2008) in Krycklan (KR), northern Sweden. BioSAR airborne data were also processed to correspond as closely as possible to future BIOMASS TomoSAR acquisitions, with BioSAR-2-based results shown. A power law AGB model using volumetric HV polarized backscatter performs best in KR, with training residual root mean-squared error (RMSE) of 30%–36% (27–33 t/ha) for airborne data and 38%–39% for simulated BIOMASS data. Airborne TomoSAR data suggest that both vertical and horizontal tomographic resolution are of importance and that it is possible to greatly reduce AGB retrieval bias when compared with airborne P-band SAR backscatter intensity-based retrievals. A lack of significant ground slopes in Remningstorp reduces the benefit of using TomoSAR data which performs similar to retrievals based solely on P-band SAR backscatter intensity. Numéro de notice : A2021-339 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2020.3020775 date de publication en ligne : 22/09/2020 En ligne : https://doi.org/10.1109/TGRS.2020.3020775 Format de la ressource électronique : URL Article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97570 [article]

Forest height retrieval using P-band airborne multi-baseline SAR data: A novel phase compensation method / Hongliang Lu in ISPRS Journal of photogrammetry and remote sensing, vol 175 (May 2021)  

Public [article]  inISPRS Journal of photogrammetry and remote sensing > vol 175 (May 2021) . - pp 99 - 118 Titre : Forest height retrieval using P-band airborne multi-baseline SAR data: A novel phase compensation method Type de document : Article/Communication Auteurs : Hongliang Lu, Auteur ; Heng Zhang, Auteur ; Huaitao Fan, Auteur ; et al., Auteur Année de publication : 2021 Article en page(s) : pp 99 - 118 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image radar et applications [Termes descripteurs IGN] bande P [Termes descripteurs IGN] Chine [Termes descripteurs IGN] compensation [Termes descripteurs IGN] données TomoSAR [Termes descripteurs IGN] erreur de mesure [Termes descripteurs IGN] erreur de phase [Termes descripteurs IGN] Guyane (département français) [Termes descripteurs IGN] hauteur des arbres [Termes descripteurs IGN] image radar moirée [Termes descripteurs IGN] ligne de base [Termes descripteurs IGN] polarisation [Termes descripteurs IGN] tomographie radar [Termes descripteurs IGN] triangulation de Delaunay Résumé : (auteur) Synthetic aperture radar (SAR) tomography (TomoSAR) has been well-established for three-dimensional (3-D) information extraction of forests using the multi-baseline SAR data stacks. The multi-baseline SAR data stacks can be acquired by spaceborne and airborne SAR systems, but for forest scenarios, the data stacks acquired by the airborne SAR system are mostly used. Such a data stack has the advantages of short revisiting time and weak temporal decorrelation. However, due to the baseline errors (caused by the residual platform motion and the measurement errors of the navigation instruments), phase errors (PEs) will occur. PEs are independent of one track to the other, resulting in spreading and defocusing in tomographic imaging. In this paper, we proposed a novel phase compensation method named NC-PGA, which combines the methods of network construction (NC) and phase gradient autofocus (PGA) to estimate and compensate the PEs. The NC method uses the Delaunay triangulation network and beamforming to obtain an accurate elevation estimate of the selected permanent scatterers, which can be used as the prior information for subsequent processing to overcome the shortcomings of the PGA method in PEs estimation. The PGA method uses the spatial invariance of PEs in a limited area to compensate for the PE of each track. The applicability of the NC-PGA method is demonstrated using simulated data and real data. The real data contains two data stacks. The one is acquired by a full-polarization P-band airborne SAR system (developed independently by our project research team) over the study area in Saihanba Forest Farm in Hebei, China. The other one is acquired by ONERA SETHI airborne system over Paracou, French Guiana, in the frame of the European Space Agency’s campaign TropiSAR. We select a test area in the study area and successfully retrieve the height of the forest, and use LiDAR data for results validation and evaluation. Numéro de notice : A2021-271 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2021.02.022 date de publication en ligne : 14/03/2021 En ligne : https://doi.org/10.1016/j.isprsjprs.2021.02.022 Format de la ressource électronique : url article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97329 [article]

Graph convolutional networks by architecture search for PolSAR image classification / Hongying Liu in Remote sensing, vol 13 n° 7 (April 2021)  

Public [article]  inRemote sensing > vol 13 n° 7 (April 2021) . - n° 1404 Titre : Graph convolutional networks by architecture search for PolSAR image classification Type de document : Article/Communication Auteurs : Hongying Liu, Auteur ; Derong Xu, Auteur ; Tianwen Zhu, Auteur ; et al., Auteur Année de publication : 2021 Article en page(s) : n° 1404 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image radar et applications [Termes descripteurs IGN] apprentissage profond [Termes descripteurs IGN] bande L [Termes descripteurs IGN] classification par nuées dynamiques [Termes descripteurs IGN] classification par réseau neuronal convolutif [Termes descripteurs IGN] classification par séparateurs à vaste marge [Termes descripteurs IGN] classification semi-dirigée [Termes descripteurs IGN] échantillon [Termes descripteurs IGN] graphe [Termes descripteurs IGN] image AIRSAR [Termes descripteurs IGN] image radar moirée [Termes descripteurs IGN] noeud [Termes descripteurs IGN] polarimétrie radar [Termes descripteurs IGN] réseau neuronal de graphes Résumé : (auteur) Classification of polarimetric synthetic aperture radar (PolSAR) images has achieved good results due to the excellent fitting ability of neural networks with a large number of training samples. However, the performance of most convolutional neural networks (CNNs) degrades dramatically when only a few labeled training samples are available. As one well-known class of semi-supervised learning methods, graph convolutional networks (GCNs) have gained much attention recently to address the classification problem with only a few labeled samples. As the number of layers grows in the network, the parameters dramatically increase. It is challenging to determine an optimal architecture manually. In this paper, we propose a neural architecture search method based GCN (ASGCN) for the classification of PolSAR images. We construct a novel graph whose nodes combines both the physical features and spatial relations between pixels or samples to represent the image. Then we build a new searching space whose components are empirically selected from some graph neural networks for architecture search and develop the differentiable architecture search method to construction our ASGCN. Moreover, to address the training of large-scale images, we present a new weighted mini-batch algorithm to reduce the computing memory consumption and ensure the balance of sample distribution, and also analyze and compare with other similar training strategies. Experiments on several real-world PolSAR datasets show that our method has improved the overall accuracy as much as 3.76% than state-of-the-art methods. Numéro de notice : A2021-350 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.3390/rs13071404 date de publication en ligne : 06/04/2021 En ligne : https://doi.org/10.3390/rs13071404 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97600 [article]

Early detection of forest stress from European spruce bark beetle attack, and a new vegetation index: Normalized distance red & SWIR (NDRS) / Langning Huo in Remote sensing of environment, Vol 255 (March 2021)  

Public [article]  inRemote sensing of environment > Vol 255 (March 2021) . - n° 112240 Titre : Early detection of forest stress from European spruce bark beetle attack, and a new vegetation index: Normalized distance red & SWIR (NDRS) Type de document : Article/Communication Auteurs : Langning Huo, Auteur ; Henrik J. Persson, Auteur ; Eva Lindberg, Auteur Année de publication : 2021 Article en page(s) : n° 112240 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes descripteurs IGN] bande infrarouge [Termes descripteurs IGN] écho radar [Termes descripteurs IGN] houppier [Termes descripteurs IGN] image Sentinel-MSI [Termes descripteurs IGN] image Sentinel-SAR [Termes descripteurs IGN] indice de stress [Termes descripteurs IGN] indice de végétation [Termes descripteurs IGN] insecte nuisible [Termes descripteurs IGN] maladie parasitaire [Termes descripteurs IGN] Picea mariana [Termes descripteurs IGN] Scolytinae [Termes descripteurs IGN] signature spectrale [Termes descripteurs IGN] Suède [Termes descripteurs IGN] vulnérabilité Résumé : (auteur) The European spruce bark beetle (Ips typographus [L.]) is one of the most damaging pest insects of European spruce forests. A crucial measure in pest control is the removal of infested trees before the beetles leave the bark, which generally happens before the end of June. However, stressed tree crowns do not show any significant color changes in the visible spectrum at this early-stage of infestation, making early detection difficult. In order to detect the related forest stress at an early stage, we investigated the differences in radar and spectral signals of healthy and stressed trees. How the characteristics of stressed trees changed over time was analyzed for the whole vegetation season, which covered the period before attacks (April), early-stage infestation (‘green-attacks’, May to July), and middle to late-stage infestation (August to October). The results show that spectral differences already existed at the beginning of the vegetation season, before the attacks. The spectral separability between the healthy and infested samples did not change significantly during the ‘green-attack’ stage. The results indicate that the trees were stressed before the attacks and had spectral signatures that differed from healthy ones. These stress-induced spectral changes could be more efficient indicators of early infestations than the ‘green-attack’ symptoms. In this study we used Sentinel-1 and 2 images of a test site in southern Sweden from April to October in 2018 and 2019. The red and SWIR bands from Sentinel-2 showed the highest separability of healthy and stressed samples. The backscatter from Sentinel-1 and additional bands from Sentinel-2 contributed only slightly in the Random Forest classification models. We therefore propose the Normalized Distance Red & SWIR (NDRS) index as a new index based on our observations and the linear relationship between the red and SWIR bands. This index identified stressed forest with accuracies from 0.80 to 0.88 before the attacks, from 0.80 to 0.82 in the early-stage infestation, and from 0.81 to 0.91 in middle- and late-stage infestations. These accuracies are higher than those attained by established vegetation indices aimed at ‘green-attack’ detection, such as the Normalized Difference Water Index, Ratio Drought Index, and Disease Stress Water Index. By using the proposed method, we highlight the potential of using NDRS with Sentinel-2 images to estimate forest vulnerability to European spruce bark beetle attacks early in the vegetation season. Numéro de notice : A2021-190 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article DOI : 10.1016/j.rse.2020.112240 date de publication en ligne : 20/01/2021 En ligne : https://doi.org/10.1016/j.rse.2020.112240 Format de la ressource électronique : url article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97111 [article]

A soil texture categorization mapping from empirical and semi-empirical modelling of target parameters of synthetic aperture radar / Shoba Periasamy in Geocarto international, vol 36 n° 5 ([15/03/2021])  

Permalink

Cluster-based empirical tropospheric corrections applied to InSAR time series analysis / Kyle Dennis Murray in IEEE Transactions on geoscience and remote sensing, Vol 59 n° 3 (March 2021)  

Permalink

Saline-soil deformation extraction based on an improved time-series InSAR approach / Wei Xiang in ISPRS International journal of geo-information, vol 10 n° 3 (March 2021)  

Permalink

Forest height estimation using a single-pass airborne L-band polarimetric and interferometric SAR system and tomographic techniques / Yue Huang in Remote sensing, Vol 13 n° 3 (February 2021)  

Permalink

Optimizing flood mapping using multi-synthetic aperture radar images for regions of the lower mekong basin in Vietnam / Vu Anh Tuan in European journal of remote sensing, vol 54 n° 1 (2021)  

Permalink

Tropical forest canopy height estimation from combined polarimetric SAR and LiDAR using machine-learning / Maryam Pourshamsi in ISPRS Journal of photogrammetry and remote sensing, vol 172 (February 2021)  

Permalink

Super-resolution of VIIRS-measured ocean color products using deep convolutional neural network / Xiaoming Liu in IEEE Transactions on geoscience and remote sensing, vol 59 n° 1 (January 2021)  

Permalink

Unmixing-based Sentinel-2 downscaling for urban land cover mapping / Fei Xu in ISPRS Journal of photogrammetry and remote sensing, vol 171 (January 2021)  

Permalink

Monitoring of wheat crops using the backscattering coefficient and the interferometric coherence derived from Sentinel-1 in semi-arid areas / Nadia Ouaadi in Remote sensing of environment, Vol 251 (15 December 2020)  

Permalink

Integrated Kalman filter of accurate ranging and tracking with wideband radar / Shaopeng Wei in IEEE Transactions on geoscience and remote sensing, Vol 58 n° 12 (December 2020)  

Permalink