Centre de documentation
scientifique

Benchmarking of convolutional neural network approaches for vegetation land cover mapping / Benjamin Carpentier (2021)  

Public Titre : Benchmarking of convolutional neural network approaches for vegetation land cover mapping Type de document : Article/Communication Auteurs : Benjamin Carpentier, Auteur ; Antoine Masse , Auteur ; Emeric Lavergne, Auteur ; C. Sannier, Auteur Editeur : International Society for Photogrammetry and Remote Sensing ISPRS Année de publication : 2021 Collection : International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences, ISSN 1682-1750 num. 43-B2-2021 Conférence : ISPRS 2021, Commission 2, XXIV ISPRS Congress, Imaging today foreseeing tomorrow 05/07/2021 09/07/2021 Nice Virtuel France OA Archives Commission 2 Importance : pp 915 - 922 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] analyse comparative [Termes IGN] carte d'occupation du sol [Termes IGN] carte de la végétation [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] image Sentinel-MSI [Termes IGN] série temporelle Résumé : (auteur) Satellite Image Time Series (SITS) are becoming available at high spatial, spectral and temporal resolutions across the globe by the latest remote sensing sensors. These series of images can be highly valuable when exploited by classification systems to produce frequently updated and accurate land cover maps. The richness of spectral, spatial and temporal features in SITS is a promising source of data for developing better classification algorithms. However, machine learning methods such as Random Forests (RF), despite their fruitful application to SITS to produce land cover maps, are structurally unable to properly handle intertwined spatial, spectral and temporal dynamics without breaking the structure of the data. Therefore, the present work proposes a comparative study of various deep learning algorithms from the Convolutional Neural Network (CNN) family and evaluate their performance on SITS classification. They are compared to the processing chain coined iota2, developed by the CESBIO and based on a RF model. Experiments are carried out in an operational context using with sparse annotations from 290 labeled polygons. Less than 80 000 pixel time series belonging to 8 land cover classes from a year of Sentinel-2 monthly syntheses are used. Results show on a test set of 131 polygons that CNNs using 3D convolutions in space and time are more accurate than 1D temporal, stacked 2D and RF approaches. Best-performing models are CNNs using spatio-temporal features, namely 3D-CNN, 2D-CNN and SpatioTempCNN, a two-stream model using both 1D and 3D convolutions. Numéro de notice : C2021-017 Affiliation des auteurs : non IGN Thématique : IMAGERIE/INFORMATIQUE Nature : Communication DOI : 10.5194/isprs-archives-XLIII-B2-2021-915-2021 Date de publication en ligne : 28/06/2021 En ligne : https://doi.org/10.5194/isprs-archives-XLIII-B2-2021-915-2021 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=98069