Centre de documentation
scientifique

Evaluation of a neural network with uncertainty for detection of ice and water in SAR imagery / Nazanin Asadi in IEEE Transactions on geoscience and remote sensing, vol 59 n° 1 (January 2021)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 59 n° 1 (January 2021) . - pp 247 - 259 Titre : Evaluation of a neural network with uncertainty for detection of ice and water in SAR imagery Type de document : Article/Communication Auteurs : Nazanin Asadi, Auteur ; K. Andrea Scott, Auteur ; Alexander S. Komarov, Auteur ; et al., Auteur Année de publication : 2021 Article en page(s) : pp 247 - 259 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image radar et applications [Termes IGN] apprentissage profond [Termes IGN] assimilation des données [Termes IGN] classification pixellaire [Termes IGN] glace de mer [Termes IGN] image radar moirée [Termes IGN] incertitude des données [Termes IGN] modèle d'incertitude [Termes IGN] Perceptron multicouche [Termes IGN] pondération [Termes IGN] précision de la classification [Termes IGN] régression logistique [Termes IGN] réseau neuronal artificiel Résumé : (auteur) Synthetic aperture radar (SAR) sea ice imagery is a promising source of data for sea ice data assimilation. Classification of SAR sea ice imagery into ice and water is of particular relevance due to its relationship with ice concentration, a key variable in sea ice data assimilation systems. With increasing volumes of SAR data, automated methods to carry out these classifications are of particular importance. Although several automated approaches have been proposed, none look at the impact of including an estimate of uncertainty of the model parameters and input features on the classification output. This article uses an established database of SAR image features to train a multilayer perceptron (MLP) neural network to classify pixel locations as either ice, water, or unknown. The classification accuracies are benchmarked using a recently developed logistic regression approach for the same database. The two methods are found to be comparable. The MLP approach is then enhanced to allow uncertainty to be estimated at each pixel location. Following methods proposed in the deep learning community, two kinds of uncertainty are considered. The first, epistemic uncertainty, is that due to uncertainty in the MLP weights. The second kind of uncertainty, aleatoric uncertainty, is that which cannot be explained by the model, and is therefore associated with the input data. It is found that including these uncertainties in the MLP models reduces their accuracies slightly, but also reduces misclassification rates. This is of particular importance for data assimilation applications, where misclassifications could severely degrade the analysis. Numéro de notice : A2021-033 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2020.2992454 Date de publication en ligne : 09/06/2020 En ligne : https://doi.org/10.1109/TGRS.2020.2992454 Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96735 [article]