Centre de documentation
scientifique

Domain adaptive transfer attack-based segmentation networks for building extraction from aerial images / Younghwan Na in IEEE Transactions on geoscience and remote sensing, vol 59 n° 6 (June 2021)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 59 n° 6 (June 2021) . - pp 5171 - 5182 Titre : Domain adaptive transfer attack-based segmentation networks for building extraction from aerial images Type de document : Article/Communication Auteurs : Younghwan Na, Auteur ; Jun Hee Kim, Auteur ; Kyungsu Lee, Auteur ; et al., Auteur Année de publication : 2021 Article en page(s) : pp 5171 - 5182 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] détection du bâti [Termes IGN] entropie [Termes IGN] image aérienne [Termes IGN] segmentation sémantique Résumé : (auteur) Semantic segmentation models based on convolutional neural networks (CNNs) have gained much attention in relation to remote sensing and have achieved remarkable performance for the extraction of buildings from high-resolution aerial images. However, the issue of limited generalization for unseen images remains. When there is a domain gap between the training and test data sets, the CNN-based segmentation models trained by a training data set fail to segment buildings for the test data set. In this article, we propose segmentation networks based on a domain adaptive transfer attack (DATA) scheme for building extraction from aerial images. The proposed system combines the domain transfer and the adversarial attack concepts. Based on the DATA scheme, the distribution of the input images can be shifted to that of the target images while turning images into adversarial examples against a target network. Defending adversarial examples adapted to the target domain can overcome the performance degradation due to the domain gap and increase the robustness of the segmentation model. Cross-data set experiments and ablation study are conducted for three different data sets: the Inria aerial image labeling data set, the Massachusetts building data set, and the WHU East Asia data set. Compared with the performance of the segmentation network without the DATA scheme, the proposed method shows improvements in the overall intersection over union (IoU). Moreover, it is verified that the proposed method outperforms even when compared with feature adaptation (FA) and output space adaptation (OSA). Numéro de notice : A2021-427 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2020.3010055 Date de publication en ligne : 30/07/2020 En ligne : https://doi.org/10.1109/TGRS.2020.3010055 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97783 [article]