Centre dedocumentation
scientifique

CNN semantic segmentation to retrieve past land cover out of historical orthoimages and DSM: first experiments / Arnaud Le Bris in ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, V-2 (August 2020)  

Public [article]  inISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences > V-2 (August 2020) . - pp 1013 - 1019 Titre : CNN semantic segmentation to retrieve past land cover out of historical orthoimages and DSM: first experiments Type de document : Article/Communication Auteurs : Arnaud Le Bris , Auteur ; Sébastien Giordano , Auteur ; Clément Mallet , Auteur Année de publication : 2020 Projets : HIATUS / Giordano, Sébastien Conférence : ISPRS 2020, Commission 2, virtual Congress, Imaging today foreseeing tomorrow 31/08/2020 02/09/2020 Nice (en ligne) France Annals Commission 2 Article en page(s) : pp 1013 - 1019 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes descripteurs IGN] base de données historiques [Termes descripteurs IGN] classification par réseau neuronal convolutif [Termes descripteurs IGN] image aérienne [Termes descripteurs IGN] modèle numérique de surface [Termes descripteurs IGN] occupation du sol [Termes descripteurs IGN] orthoimage [Termes descripteurs IGN] segmentation sémantique Résumé : (auteur) Images from archival aerial photogrammetric surveys are a unique and relatively unexplored means to chronicle 3D land-cover changes occurred since the mid 20th century. They provide a relatively dense temporal sampling of the territories with a very high spatial resolution. Thus, they offer time series data which can answer a large variety of long-term environmental monitoring studies. Besides, they are generally stereoscopic surveys, making it possible to derive 3D information (Digital Surface Models). In recent years, they have often been digitized, making them more suitable to be considered in automatic analyses processes. Some photogrammetric softwares make it possible to retrieve their geometry (pose and camera calibration) and to generate corresponding DSM and orthophotomosaic. Thus, archival aerial photogrammetric surveys appear as being a powerful remote sensing data source to study land use/cover evolution over the last century. However, several difficulties have to be faced to be able to use them in automatic analysis processes. Indeed, surveys available on a study area can exhibit very different characteristics: survey pattern, focal, spatial resolution, modality (panchromatic, colour, infrared…). Planimetric and altimetric accuracies of derived products strongly depend on these characteristics. Thus, analysis processes have to cope with these uncertainties. Another important gap states in the lack of training data. Deep learning methods and especially Convolutional Neural Networks (CNN) are at present the most efficient semantic segmentation methods as long as a sufficient training dataset is available. However, temporal gaps can be very important between existing available databases and archival data. In this study, two custom variants of simple yet effective U-net - Deconv-Net inspired DL architectures are developed to process ortho-image and DSM based information. They are then trained out of a groundtruth derived out of a recent database to process archival datasets. Numéro de notice : A2020-469 Affiliation des auteurs : LaSTIG (2020- ) Thématique : IMAGERIE/INFORMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.5194/isprs-annals-V-2-2020-1013-2020 date de publication en ligne : 03/08/2020 En ligne : https://doi.org/10.5194/isprs-annals-V-2-2020-1013-2020 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=95637 [article]

Correction of systematic radiometric inhomogeneity in scanned aerial campaigns using principal component analysis / Lâmân Lelégard in ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, V-2 (August 2020)  

Public [article]  inISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences > V-2 (August 2020) . - pp 871 - 876 Titre : Correction of systematic radiometric inhomogeneity in scanned aerial campaigns using principal component analysis Type de document : Article/Communication Auteurs : Lâmân Lelégard , Auteur ; Arnaud Le Bris , Auteur ; Sébastien Giordano , Auteur Année de publication : 2020 Projets : HIATUS / Giordano, Sébastien Conférence : ISPRS 2020, Commission 2, virtual Congress, Imaging today foreseeing tomorrow 31/08/2020 02/09/2020 Nice (en ligne) France Annals Commission 2 Article en page(s) : pp 871 - 876 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes descripteurs IGN] algorithme d'homogénéisation [Termes descripteurs IGN] analyse en composantes principales [Termes descripteurs IGN] correction radiométrique [Termes descripteurs IGN] image numérisée Résumé : (auteur) Orthophotomosaic is defined as a single image that can be layered on a map. The term “mosaic” implies that it is produced from a set of images, usually aerial images. Even if these images are taken during cloudless period, they are impaired by radiometric inhomogeneity mostly due to atmospheric phenomena, like hotspot, haze or high altitude clouds shadows as well as imaging device systematisms, like lens vignetting. These create some unsightly radiometric inhomogeneity in the orthophotomosaic that could be corrected by using a Wallis filter. Yet this solution leads to a significant loss of contrast at small scales. This work introduces an alternative to Wallis filter by considering some systematic radiometric behaviours in the images through a principal component analysis process. Numéro de notice : A2020-501 Affiliation des auteurs : LaSTIG (2020- ) Thématique : IMAGERIE/INFORMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.5194/isprs-annals-V-2-2020-871-2020 date de publication en ligne : 03/08/2020 En ligne : https://doi.org/10.5194/isprs-annals-V-2-2020-871-2020 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=95642 [article]