Centre dedocumentation
scientifique

Dépouillements

Hyperspectral image classification with squeeze multibias network / Leyuan Fang in IEEE Transactions on geoscience and remote sensing, vol 57 n° 3 (March 2019)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 57 n° 3 (March 2019) . - pp 1291 - 1301 Titre : Hyperspectral image classification with squeeze multibias network Type de document : Article/Communication Auteurs : Leyuan Fang, Auteur ; Guangyun Liu, Auteur ; Shutao Li, Auteur ; Pedram Ghamisi, Auteur ; Jon Atli Benediktsson, Auteur Année de publication : 2019 Article en page(s) : pp 1291 - 1301 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] erreur systématique [Termes IGN] image hyperspectrale Résumé : (Auteur) A convolutional neural network (CNN) has recently demonstrated its outstanding capability for the classification of hyperspectral images (HSIs). Typical CNN-based methods usually adopt image patches as inputs to the network. However, a fixed-size image patch in HSI with complex spatial contexts may contain multiple ground objects of different classes, which will deteriorate the classification performance of the CNN. In addition, traditional convolutional layers adopted in the CNN have a huge amount of parameters needed to be tuned, which will cause high computational cost. To address the above-mentioned issues, a novel squeeze multibias network (SMBN) is proposed for HSI classification. Specifically, the proposed SMBN first introduces the multibias module (MBM), which incorporates multibias into the rectified linear unit layers. The MBM can decouple the feature maps of input patches into multiple response maps (corresponding to different ground objects) and adaptively select the meaningful maps for classification. Furthermore, the proposed SMBN replaces the traditional convolutional layer with a squeeze convolution module, which can greatly reduce the number of parameters in the network, thus saving the running time, while still maintaining high classification accuracy. Experimental results on three real HSIs demonstrate the superiority of the proposed SMBN method over several state-of-the-art classification approaches. Numéro de notice : A2019-113 Affiliation des auteurs : non IGN Thématique : IMAGERIE/INFORMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2018.2865953 Date de publication en ligne : 13/09/2018 En ligne : https://doi.org/10.1109/TGRS.2018.2865953 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=92453 [article]

Stem-leaf segmentation and phenotypic trait extraction of individual maize using terrestrial LiDAR data / Shichao Jin in IEEE Transactions on geoscience and remote sensing, vol 57 n° 3 (March 2019)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 57 n° 3 (March 2019) . - pp 1336 - 1346 Titre : Stem-leaf segmentation and phenotypic trait extraction of individual maize using terrestrial LiDAR data Type de document : Article/Communication Auteurs : Shichao Jin, Auteur ; Yanjun Su, Auteur ; Fangfang Wu, Auteur ; et al., Auteur Année de publication : 2019 Article en page(s) : pp 1336 - 1346 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] données lidar [Termes IGN] données localisées 3D [Termes IGN] maïs (céréale) [Termes IGN] phénologie [Termes IGN] segmentation en régions Résumé : (Auteur) Accurate and high throughput extraction of crop phenotypic traits, as a crucial step of molecular breeding, is of great importance for yield increasing. However, automatic stem-leaf segmentation as a prerequisite of many precise phenotypic trait extractions is still a big challenge. Current works focus on the study of the 2-D image-based segmentation, which are sensitive to illumination and occlusion. Light detection and ranging (LiDAR) can obtain accurate 3-D information with its active laser scanning and strong penetration ability, which breaks through phenotyping from 2-D to 3-D. However, few researches have addressed the problem of the LiDAR-based stem-leaf segmentation. In this paper, we proposed a median normalized-vector growth (MNVG) algorithm, which can segment stem and leaf with four steps, i.e., preprocessing, stem growth, leaf growth, and postprocessing. The MNVG method was tested by 30 maize samples with different heights, compactness, leaf numbers, and densities from three growing stages. Moreover, phenotypic traits at leaf, stem, and individual levels were extracted with the truly segmented instances. The mean accuracy of segmentation at point level in terms of the recall, precision, F-score, and overall accuracy were 0.92, 0.93, 0.92, and 0.93, respectively. The accuracy of phenotypic trait extraction in leaf, stem, and individual levels ranged from 0.81 to 0.95, 0.64 to 0.97, and 0.96 to 1, respectively. To our knowledge, this paper proposed the first LiDAR-based stem-leaf segmentation and phenotypic trait extraction method in agriculture field, which may contribute to the study of LiDAR-based plant phonemics and precise agriculture. Numéro de notice : A2019-114 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2018.2866056 Date de publication en ligne : 19/09/2018 En ligne : https://doi.org/10.1109/TGRS.2018.2866056 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=92454 [article]

Calibration of the normalized radar cross section for sentinel-1 wave mode / Huimin Li in IEEE Transactions on geoscience and remote sensing, vol 57 n° 3 (March 2019)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 57 n° 3 (March 2019) . - pp 1514 - 1522 Titre : Calibration of the normalized radar cross section for sentinel-1 wave mode Type de document : Article/Communication Auteurs : Huimin Li, Auteur ; Alexis Mouche, Auteur ; Justin E. Stopa, Auteur ; Bertrand Chapron, Auteur Année de publication : 2019 Article en page(s) : pp 1514 - 1522 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image radar et applications [Termes IGN] Amazonie [Termes IGN] étalonnage radiométrique [Termes IGN] forêt équatoriale [Termes IGN] image Sentinel-SAR [Termes IGN] résidu Résumé : (Auteur) Sentinel-1 (S-1) is a two-satellite constellation for continuity of operational synthetic aperture radar (SAR) observations. Wave mode (WV) is the default mode over open ocean for S-1 to monitor global ocean waves and wind field. Therefore, proper radiometric calibration is essential to accurately infer these geophysical quantities. Based on the global data set acquired by S-1A WV, assessment of normalized radar cross section (NRCS) is carried out through comparison with CMOD5.N predictions over open ocean. The calibration accuracy quantified by NRCS residuals between SAR measurements and CMOD5.N demonstrates distinct features for two incidence angles (23.8° and 36.8°). Particularly, NRCS at 23.8° is overall consistent with CMOD5.N, while NRCS at 36.8° displays great deviation. Two recalibration methods are then implemented by examining the backscattering profile over Amazon rain forest and ocean calibration. Both methods show the necessity for recalibration and obtain comparable correction factors for WV1 and WV2, respectively. The NRCS residuals by applying both methods are significantly reduced toward zero. By comparison, ocean calibration is more efficient and practical to implement. Numéro de notice : A2019-128 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2018.2867035 Date de publication en ligne : 14/09/2018 En ligne : https://doi.org/10.1109/TGRS.2018.2867035 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=92457 [article]

A novel sharpening approach for superresolving multiresolution optical images / Claudia Paris in IEEE Transactions on geoscience and remote sensing, vol 57 n° 3 (March 2019)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 57 n° 3 (March 2019) . - pp 1545 - 1560 Titre : A novel sharpening approach for superresolving multiresolution optical images Type de document : Article/Communication Auteurs : Claudia Paris, Auteur ; José Bioucas-Dias, Auteur ; Lorenzo Bruzzone, Auteur Année de publication : 2019 Article en page(s) : pp 1545 - 1560 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] filtrage du bruit [Termes IGN] image multibande [Termes IGN] image Sentinel-MSI [Termes IGN] problème inverse Résumé : (Auteur) This paper aims to provide a compact superresolution formulation specific for multispectral (MS) multiresolution optical data, i.e., images characterized by different scales across different spectral bands. The proposed method, named multiresolution sharpening approach (MuSA), relies on the solution of an optimization problem tailored to the properties of those images. The superresolution problem is formulated as the minimization of an objective function containing a data-fitting term that models the blurs and downsamplings of the different bands and a patch-based regularizer that promotes image self-similarity guided by the geometric details provided by the high-resolution bands. By exploiting the approximately low-rank property of the MS data, the ill-posedness of the inverse problem in hand is strongly reduced, thus sharply improving its conditioning. The state-of-the-art color block-matching and 3D filtering (C-BM3D) image denoiser is used as a patch-based regularizer by leveraging the “plug-and-play” framework: the denoiser is plugged into the iterations of the alternating direction method of multipliers. The main novelties of the proposed method are: 1) the introduction of an observation model tailored to the specific properties of (MS) multiresolution images and 2) the exploitation of the high-spatial-resolution bands to guide the grouping step in the color block-matching and 3D filtering (C-BM3D) denoiser, which constitutes a form of regularization learned from the high-resolution channels. The results obtained on the real and synthetic Sentinel 2 data sets give an evidence of the effectiveness of the proposed approach. Numéro de notice : A2019-129 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2018.2867284 Date de publication en ligne : 26/09/2018 En ligne : https://doi.org/10.1109/TGRS.2018.2867284 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=92458 [article]

Developing a subswath-based wind speed retrieval model for sentinel-1 VH-Polarized SAR data over the ocean surface / Kangyu Zhang in IEEE Transactions on geoscience and remote sensing, vol 57 n° 3 (March 2019)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 57 n° 3 (March 2019) . - pp 1561 - 1572 Titre : Developing a subswath-based wind speed retrieval model for sentinel-1 VH-Polarized SAR data over the ocean surface Type de document : Article/Communication Auteurs : Kangyu Zhang, Auteur ; Jingfeng Huang, Auteur ; Lamin R. Mansaray, Auteur ; Qiaoying Guo, Auteur ; et al., Auteur Année de publication : 2019 Article en page(s) : pp 1561 - 1572 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image radar et applications [Termes IGN] bande C [Termes IGN] données polarimétriques [Termes IGN] image Sentinel-SAR [Termes IGN] polarimétrie radar [Termes IGN] surface de la mer [Termes IGN] vent [Termes IGN] vitesse Résumé : (Auteur) This paper evaluates the capability of Sentinel-1 VH-polarized synthetic aperture radar signals, involving 738 scenes in the interferometric wide swath (IW) mode, for ocean surface wind speed retrieval using a novel subswath-based C-band cross-polarized ocean model. When compared with in situ measurements, it is observed that wind speed retrieval accuracy varies progressively along swath, with the most accurate wind speed retrievals being derived from subswath 3 [root-mean-square error (RMSE) of 1.82 m · s -1 ], followed by subswath 2 (RMSE of 1.92 m · s -1 ), while subswath 1 showed the lowest retrieval accuracy (RMSE of 2.37 m · s -1 ). The average RMSE of wind speeds retrieved from all the three subswaths is 2.08 m · s -1 under low-to-high wind speed regimes (wind speeds Numéro de notice : A2019-130 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2018.2867438 Date de publication en ligne : 20/09/2018 En ligne : https://doi.org/10.1109/TGRS.2018.2867438 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=92459 [article]

Conditional random field and deep feature learning for hyperspectral image classification / Fahim Irfan Alam in IEEE Transactions on geoscience and remote sensing, vol 57 n° 3 (March 2019)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 57 n° 3 (March 2019) . - pp 1612 - 1628 Titre : Conditional random field and deep feature learning for hyperspectral image classification Type de document : Article/Communication Auteurs : Fahim Irfan Alam, Auteur ; Jun Zhou, Auteur ; Alan Wee-Chung Liew, Auteur ; Xiuping Jia, Auteur ; et al., Auteur Année de publication : 2019 Article en page(s) : pp 1612 - 1628 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] analyse multibande [Termes IGN] champ aléatoire conditionnel [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] déconvolution [Termes IGN] données localisées 3D [Termes IGN] image hyperspectrale [Termes IGN] voxel Résumé : (Auteur) Image classification is considered to be one of the critical tasks in hyperspectral remote sensing image processing. Recently, a convolutional neural network (CNN) has established itself as a powerful model in classification by demonstrating excellent performances. The use of a graphical model such as a conditional random field (CRF) contributes further in capturing contextual information and thus improving the classification performance. In this paper, we propose a method to classify hyperspectral images by considering both spectral and spatial information via a combined framework consisting of CNN and CRF. We use multiple spectral band groups to learn deep features using CNN, and then formulate deep CRF with CNN-based unary and pairwise potential functions to effectively extract the semantic correlations between patches consisting of 3-D data cubes. Furthermore, we introduce a deep deconvolution network that improves the final classification performance. We also introduced a new data set and experimented our proposed method on it along with several widely adopted benchmark data sets to evaluate the effectiveness of our method. By comparing our results with those from several state-of-the-art models, we show the promising potential of our method. Numéro de notice : A2019-131 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2018.2867679 Date de publication en ligne : 20/09/2018 En ligne : https://doi.org/10.1109/TGRS.2018.2867679 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=92461 [article]