Centre de documentation
scientifique

Unsupervised-restricted deconvolutional neural network for very high resolution remote-sensing image classification / Yiting Tao in IEEE Transactions on geoscience and remote sensing, vol 55 n° 12 (December 2017)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 55 n° 12 (December 2017) . - pp 6805 - 6823 Titre : Unsupervised-restricted deconvolutional neural network for very high resolution remote-sensing image classification Type de document : Article/Communication Auteurs : Yiting Tao, Auteur ; Miaozhong Xu, Auteur ; Fan Zhang, Auteur ; Bo Du, Auteur ; Liangpei Zhang, Auteur Année de publication : 2017 Article en page(s) : pp 6805 - 6823 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] apprentissage non-dirigé [Termes IGN] classification pixellaire [Termes IGN] déconvolution [Termes IGN] image Geoeye [Termes IGN] image Quickbird [Termes IGN] méthode fondée sur le noyau [Termes IGN] réseau neuronal convolutif Résumé : (Auteur) As the acquisition of very high resolution (VHR) satellite images becomes easier owing to technological advancements, ever more stringent requirements are being imposed on automatic image interpretation. Moreover, per-pixel classification has become the focus of research interests in this regard. However, the efficient and effective processing and the interpretation of VHR satellite images remain a critical task. Convolutional neural networks (CNNs) have recently been applied to VHR satellite images with considerable success. However, the prevalent CNN models accept input data of fixed sizes and train the classifier using features extracted directly from the convolutional stages or the fully connected layers, which cannot yield pixel-to-pixel classifications. Moreover, training a CNN model requires large amounts of labeled reference data. These are challenging to obtain because per-pixel labeled VHR satellite images are not open access. In this paper, we propose a framework called the unsupervised-restricted deconvolutional neural network (URDNN). It can solve these problems by learning an end-to-end and pixel-to-pixel classification and handling a VHR classification using a fully convolutional network and a small number of labeled pixels. In URDNN, supervised learning is always under the restriction of unsupervised learning, which serves to constrain and aid supervised training in learning more generalized and abstract feature. To some degree, it will try to reduce the problems of overfitting and undertraining, which arise from the scarcity of labeled training data, and to gain better classification results using fewer training samples. It improves the generality of the classification model. We tested the proposed URDNN on images from the Geoeye and Quickbird sensors and obtained satisfactory results with the highest overall accuracy (OA) achieved as 0.977 and 0.989, respectively. Experiments showed that the combined effects of additional kernels and stages may have produced better results, and two-stage URDNN consistently produced a more stable result. We compared URDNN with four methods and found that with a small ratio of selected labeled data items, it yielded the highest and most stable results, whereas the accuracy values of the other methods quickly decreased. For some categories with fewer training pixels, accuracy for categories from other methods was considerably worse than that in URDNN, with the largest difference reaching almost 10%. Hence, the proposed URDNN can successfully handle the VHR image classification using a small number of labeled pixels. Furthermore, it is more effective than state-of-the-art methods. Numéro de notice : A2017-766 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2017.2734697 En ligne : https://doi.org/10.1109/TGRS.2017.2734697 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=88803 [article]

Extrapolated georeferencing of high-resolution satellite imagery based on the strip constraint / Jinshan Cao in Photogrammetric Engineering & Remote Sensing, PERS, vol 83 n° 7 (July 2017)  

Public [article]  inPhotogrammetric Engineering & Remote Sensing, PERS > vol 83 n° 7 (July 2017) . - pp 493 - 499 Titre : Extrapolated georeferencing of high-resolution satellite imagery based on the strip constraint Type de document : Article/Communication Auteurs : Jinshan Cao, Auteur ; Xiuxiao Yuan, Auteur ; Jianya Gong, Auteur ; Miaozhong Xu, Auteur Année de publication : 2017 Article en page(s) : pp 493 - 499 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Photogrammétrie numérique [Termes IGN] compensation [Termes IGN] géoréférencement indirect [Termes IGN] image à haute résolution [Termes IGN] image satellite [Termes IGN] image ZiYuan-3 [Termes IGN] modèle relationnel [Termes IGN] point d'appui Résumé : (auteur) Ground control points (GCPs) are necessary in order to achieve precise georeferencing of high-resolution satellite (HRS) imagery. However, measuring GCPs is costly, laborious, and time consuming. In some remote areas, we cannot even obtain well-defined GCPs. In this study, a strip constraint model is established. Based on the bias-compensated rational function model and the strip constraint model, a feasible extrapolated georeferencing approach for HRS imagery is presented. The presented approach remains effective even when the intermediate images in the strip are unavailable. Experimental results of the two ZiYuan-3 (ZY-3) nadir datasets show that the direct georeferencing accuracy of the ZY-3 nadir images reaches only 9 to 12 pixels. With four GCPs in the first image, the georeferencing accuracy of the other images in the strip is improved to better than 2 pixels through extrapolated georeferencing. Numéro de notice : A2017-433 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.14358/PERS.83.7.493 En ligne : https://doi.org/10.14358/PERS.83.7.493 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=86337 [article]

Unsupervised segmentation of high-resolution remote sensing images based on classical models of the visual receptive field / Miaozhong Xu in Geocarto international, vol 30 n° 9 - 10 (October - November 2015)  

Public [article]  inGeocarto international > vol 30 n° 9 - 10 (October - November 2015) . - pp 997 - 1015 Titre : Unsupervised segmentation of high-resolution remote sensing images based on classical models of the visual receptive field Type de document : Article/Communication Auteurs : Miaozhong Xu, Auteur ; Ming Cong, Auteur ; Tianpeng Xie, Auteur Année de publication : 2015 Article en page(s) : pp 997 - 1015 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] champ aléatoire de Markov [Termes IGN] filtre de Gabor [Termes IGN] segmentation d'image [Termes IGN] transformation en ondelettes Résumé : (Auteur) Here, we describe an unsupervised segmentation method incorporating log-Gabor (LG) filters and a Markov random field (MRF) model for high-resolution (HR) remote sensing (RS) images, based on classical models of the visual receptive field. LG filters were utilised to model the receptive fields of the simple cells in the primary visual cortex and extract detailed features from HR–RS images followed by construction of image pyramid through wavelet decomposition to simulate the hierarchical structure of the visual sensing system. Finally, based on the original HR–RS images, their detailed features and the image pyramid, the MRF image segmentation model was applied to obtain the final segmentation result. Real HR–RS images were used as experimental data to validate the proposed method, both qualitatively (visually) and numerically (with the overall accuracy and Kappa index).The experimental results indicate that the proposed method is effective, feasible and robust to noise. Numéro de notice : A2015-627 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/10106049.2015.1006529 Date de publication en ligne : 26/02/2015 En ligne : http://www.tandfonline.com/doi/full/10.1080/10106049.2015.1006529 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=78110 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
059-2015051	RAB	Revue	Centre de documentation	En réserve L003	Disponible