Centre de documentation
scientifique

Saliency-guided deep neural networks for SAR image change detection / Jie Geng in IEEE Transactions on geoscience and remote sensing, Vol 57 n° 10 (October 2019)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > Vol 57 n° 10 (October 2019) . - pp 7365 - 7377 Titre : Saliency-guided deep neural networks for SAR image change detection Type de document : Article/Communication Auteurs : Jie Geng, Auteur ; Xiaorui Ma, Auteur ; Xiaojun Zhou, Auteur ; et al., Auteur Année de publication : 2019 Article en page(s) : pp 7365 - 7377 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image radar et applications [Termes IGN] analyse d'image orientée objet [Termes IGN] classification floue [Termes IGN] classification non dirigée [Termes IGN] classification par réseau neuronal [Termes IGN] détection de changement [Termes IGN] échantillonnage d'image [Termes IGN] extraction de traits caractéristiques [Termes IGN] filtre de déchatoiement [Termes IGN] image radar moirée [Termes IGN] logique floue [Termes IGN] occupation du sol [Termes IGN] saillance [Termes IGN] télédétection en hyperfréquence Mots-clés libres : hierarchical fuzzy C-means clustering (HFCM) Résumé : (auteur) Change detection is an important task to identify land-cover changes between the acquisitions at different times. For synthetic aperture radar (SAR) images, inherent speckle noise of the images can lead to false changed points, which affects the change detection performance. Besides, the supervised classifier in change detection framework requires numerous training samples, which are generally obtained by manual labeling. In this paper, a novel unsupervised method named saliency-guided deep neural networks (SGDNNs) is proposed for SAR image change detection. In the proposed method, to weaken the influence of speckle noise, a salient region that probably belongs to the changed object is extracted from the difference image. To obtain pseudotraining samples automatically, hierarchical fuzzy C-means (HFCM) clustering is developed to select samples with higher probabilities to be changed and unchanged. Moreover, to enhance the discrimination of sample features, DNNs based on the nonnegative- and Fisher-constrained autoencoder are applied for final detection. Experimental results on five real SAR data sets demonstrate the effectiveness of the proposed approach. Numéro de notice : A2019-536 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2019.2913095 Date de publication en ligne : 19/05/2019 En ligne : http://doi.org/10.1109/TGRS.2019.2913095 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=94154 [article]

Deep supervised and contractive neural network for SAR image classification / Jie Geng in IEEE Transactions on geoscience and remote sensing, vol 55 n° 4 (April 2017)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 55 n° 4 (April 2017) . - pp 2442 - 2459 Titre : Deep supervised and contractive neural network for SAR image classification Type de document : Article/Communication Auteurs : Jie Geng, Auteur ; Hongyu Wang, Auteur ; Jianchao Fan, Auteur ; Xiaorui Ma, Auteur Année de publication : 2017 Article en page(s) : pp 2442 - 2459 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image radar et applications [Termes IGN] algorithme Graph-Cut [Termes IGN] analyse discriminante [Termes IGN] apprentissage profond [Termes IGN] classification par réseau neuronal [Termes IGN] extraction de traits caractéristiques [Termes IGN] filtre de déchatoiement [Termes IGN] filtre de Gabor [Termes IGN] image radar moirée [Termes IGN] matrice de co-occurrence [Termes IGN] niveau de gris (image) Résumé : (Auteur) The classification of a synthetic aperture radar (SAR) image is a significant yet challenging task, due to the presence of speckle noises and the absence of effective feature representation. Inspired by deep learning technology, a novel deep supervised and contractive neural network (DSCNN) for SAR image classification is proposed to overcome these problems. In order to extract spatial features, a multiscale patch-based feature extraction model that consists of gray level-gradient co-occurrence matrix, Gabor, and histogram of oriented gradient descriptors is developed to obtain primitive features from the SAR image. Then, to get discriminative representation of initial features, the DSCNN network that comprises four layers of supervised and contractive autoencoders is proposed to optimize features for classification. The supervised penalty of the DSCNN can capture the relevant information between features and labels, and the contractive restriction aims to enhance the locally invariant and robustness of the encoding representation. Consequently, the DSCNN is able to produce effective representation of sample features and provide superb predictions of the class labels. Moreover, to restrain the influence of speckle noises, a graph-cut-based spatial regularization is adopted after classification to suppress misclassified pixels and smooth the results. Experiments on three SAR data sets demonstrate that the proposed method is able to yield superior classification performance compared with some related approaches. Numéro de notice : A2017-176 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2016.2645226 En ligne : http://dx.doi.org/10.1109/TGRS.2016.2645226 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=84748 [article]

Semisupervised classification for hyperspectral image based on multi-decision labeling and deep feature learning / Xiaorui Ma in ISPRS Journal of photogrammetry and remote sensing, vol 120 (october 2016)  

Public [article]  inISPRS Journal of photogrammetry and remote sensing > vol 120 (october 2016) . - pp 99 - 107 Titre : Semisupervised classification for hyperspectral image based on multi-decision labeling and deep feature learning Type de document : Article/Communication Auteurs : Xiaorui Ma, Auteur ; Hongyu Wang, Auteur ; Jie Wang, Auteur Année de publication : 2016 Article en page(s) : pp 99 - 107 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] apprentissage profond [Termes IGN] apprentissage semi-dirigé [Termes IGN] classification par réseau neuronal [Termes IGN] classification semi-dirigée [Termes IGN] image hyperspectrale [Termes IGN] pondération Résumé : (Auteur) Semisupervised learning is widely used in hyperspectral image classification to deal with the limited training samples, however, some more information of hyperspectral image should be further explored. In this paper, a novel semisupervised classification based on multi-decision labeling and deep feature learning is presented to exploit and utilize as much information as possible to realize the classification task. First, the proposed method takes two decisions to pre-label each unlabeled sample: local decision based on weighted neighborhood information is made by the surrounding samples, and global decision based on deep learning is performed by the most similar training samples. Then, some unlabeled ones with high confidence are selected to extent the training set. Finally, self decision, which depends on the self features exploited by deep learning, is employed on the updated training set to extract spectral-spatial features and produce classification map. Experimental results with real data indicate that it is an effective and promising semisupervised classification method for hyperspectral image. Numéro de notice : A2016-797 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2016.09.001 En ligne : https://doi.org/10.1016/j.isprsjprs.2016.09.001 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=82532 [article]