Centre dedocumentation
scientifique

Self-attention and generative adversarial networks for algae monitoring / Nhut Hai Huynh in European journal of remote sensing, vol 55 n° 1 (January 2022)  

Public [article]  inEuropean journal of remote sensing > vol 55 n° 1 (January 2022) . - pp 10 - 22 Titre : Self-attention and generative adversarial networks for algae monitoring Type de document : Article/Communication Auteurs : Nhut Hai Huynh, Auteur ; Gordon Boër, Auteur ; Hauke Schramm, Auteur Année de publication : 2022 Article en page(s) : pp 10 - 22 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] algue [Termes IGN] analyse en composantes principales [Termes IGN] apprentissage profond [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] image hyperspectrale [Termes IGN] plancton [Termes IGN] réseau antagoniste génératif [Termes IGN] réseau neuronal artificiel Résumé : (auteur) Water is important for the natural environment and human health. Monitoring algae concentrations yield information on the water quality. Compared with in situ measurements of water quality parameters, which are often complex and expensive, remote sensing techniques, using hyperspectral data analysis, are fast and cost-effective. The objectives of this study are (1) to estimate the algae concentrations from hyperspectral data using deep learning techniques, (2) to investigate the applicability of attention mechanisms in the analysis of hyperspectral data, and (3) to augment the training data using generative adversarial networks (GANs). The results show that the accuracy of deep learning techniques is 7.6% higher than that of simpler artificial neural networks. Compared to noise injection and principal component analysis-based data augmentation, the use of a GAN-based data augmentation method significantly improves the accuracy of algae concentration estimates (>5%). In addition, models with added attention mechanisms yield an on average 3.13% higher accuracy than those without attention techniques. This result demonstrates the improvement of spectral features of artificial hyperspectral data based on the self-attention approach, revealing the potential of attention techniques in hyperspectral remote sensing. Numéro de notice : A2022-097 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : https://doi.org/10.1080/22797254.2021.2010605 Date de publication en ligne : 02/01/2022 En ligne : https://doi.org/10.1080/22797254.2021.2010605 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99547 [article]

Detection and biomass estimation of phaeocystis globosa blooms off Southern China from UAV-based hyperspectral measurements / Xue Li in IEEE Transactions on geoscience and remote sensing, vol 60 n° 1 (January 2022)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 60 n° 1 (January 2022) . - n° 4200513 Titre : Detection and biomass estimation of phaeocystis globosa blooms off Southern China from UAV-based hyperspectral measurements Type de document : Article/Communication Auteurs : Xue Li, Auteur ; Shaoling Shang, Auteur ; Zhongping Lee, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : n° 4200513 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] algue [Termes IGN] biomasse [Termes IGN] cartographie thématique [Termes IGN] Chine [Termes IGN] chlorophylle [Termes IGN] couleur de l'océan [Termes IGN] espèce exotique envahissante [Termes IGN] image captée par drone [Termes IGN] image hyperspectrale [Termes IGN] plancton [Termes IGN] réflectance Résumé : (auteur) Phaeocystis globosa (P. globosa) is a unique causative species of harmful algal blooms, which can form gelatinous colonies. We, for the first time, used unmanned aerial vehicle (UAV) measurements to identify P. globosa blooms and to quantify the biomass. Based on in situ measured remote sensing reflectance ( Rrs ), it is found that, for P. globosa blooms, the maximum of the second-derivative ( dλ2Rrs ) of Rrs(λ) in the 460–480-nm domain is beyond 466 nm. An analysis of the absorption properties from algal cultures suggested that this feature comes from the absorption of chlorophyll c3 (Chl −/c3 ) around 466 nm, a prominent feature of P. globosa. This position of dλ2Rrs maximum was, thus, selected as the criterion for P. globosa identification. The spatial extent of P. globosa blooms in two bays off southern China was then mapped by applying the criterion to UAV-measured Rrs . Twelve out of 16 UAV and in situ match-up stations were consistently identified as dominated by P. globosa, indicating the accuracy of 75%. Furthermore, using localized empirical models, chlorophyll a (Chl −/a ) concentration and colony numbers of P. globosa were estimated from UAV-derived Rrs , where P. globosa colonies were found in a range of ~3–37 gel matrix/L, indicating the occurrence of weak to moderate P. globosa blooms during the surveys. The promising results suggest a high potential for detection and quantification of P. globosa blooms in near-shore bays or harbors using UAV-based hyperspectral remote sensing, where conventional ocean color satellite remote sensing runs into difficulties. Numéro de notice : A2022-025 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2021.3051466 Date de publication en ligne : 26/01/2021 En ligne : https://doi.org/10.1109/TGRS.2021.3051466 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99254 [article]

A novel unmixing-based hypersharpening method via convolutional neural network / Xiaochen Lu in IEEE Transactions on geoscience and remote sensing, vol 60 n° 1 (January 2022)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 60 n° 1 (January 2022) . - n° 5503614 Titre : A novel unmixing-based hypersharpening method via convolutional neural network Type de document : Article/Communication Auteurs : Xiaochen Lu, Auteur ; Tong Li, Auteur ; Junping Zhang, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : n° 5503614 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] analyse des mélanges spectraux [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] fusion d'images [Termes IGN] image hyperspectrale [Termes IGN] image multibande [Termes IGN] pansharpening (fusion d'images) [Termes IGN] pouvoir de résolution géométrique [Termes IGN] pouvoir de résolution spectrale Résumé : (auteur) Hypersharpening (namely, hyperspectral (HS) and multispectral (MS) image fusion) aims at enhancing the spatial resolution of HS image via an auxiliary higher resolution MS image. Currently, numerous hypersharpening methods are proposed successively, among which the unmixing-based approaches have been widely researched and demonstrated their effectiveness in the spectral fidelity aspect. However, existing unmixing-based fusion methods substantially employ mathematical techniques to solve the spectral mixture model, without taking full advantage of the collaborative spatial–spectral information that is usually helpful for abundance estimation improvement. To overcome this drawback, in this article, a novel unmixing-based HS and MS image fusion method, via a convolutional neural network (CNN), is proposed to promote spectral fidelity. The main idea of this work is to use CNN to fully explore the spatial information and the spectral information of both HS and MS images simultaneously, thereby enhancing the accuracy of estimating the abundance maps. Experiments on four simulated and real remote sensing data sets demonstrate that the proposed method is beneficial to the spectral fidelity of the fused images compared with some state-of-the-art algorithms. Meanwhile, it is also easy to implement and has a certain advantage in running time. Numéro de notice : A2022-028 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2021.3063105 Date de publication en ligne : 22/03/2021 En ligne : https://doi.org/10.1109/TGRS.2021.3063105 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99264 [article]

Early detection of spruce vitality loss with hyperspectral data: Results of an experimental study in Bavaria, Germany / Kathrin Einzmann in Remote sensing of environment, vol 266 (December 2021)  

Public [article]  inRemote sensing of environment > vol 266 (December 2021) . - n° 112676 Titre : Early detection of spruce vitality loss with hyperspectral data: Results of an experimental study in Bavaria, Germany Type de document : Article/Communication Auteurs : Kathrin Einzmann, Auteur ; Clement Atzberger, Auteur ; Nicole Pinnel, Auteur ; et al., Auteur Année de publication : 2021 Article en page(s) : n° 112676 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] Bavière (Allemagne) [Termes IGN] changement climatique [Termes IGN] classification par forêts aléatoires [Termes IGN] dépérissement [Termes IGN] détection de changement [Termes IGN] houppier [Termes IGN] image hyperspectrale [Termes IGN] indice de végétation [Termes IGN] insecte nuisible [Termes IGN] phénomène climatique extrême [Termes IGN] Picea abies [Termes IGN] réflectance spectrale [Termes IGN] série temporelle [Termes IGN] stress hydrique Résumé : (auteur) Vitality loss of trees caused by extreme weather conditions, drought stress or insect infestations, are expected to increase with ongoing climate change. The detection of vitality loss at an early stage is thus of vital importance for forestry and forest management to minimize ecological and economical damage. Remote sensing instruments are able to detect changes over large areas down to the level of individual trees. The scope of our study is to investigate whether it is possible to detect stress-related spectral changes at an early stage using hyperspectral sensors. For this purpose, two Norway spruce (Picea abies) forest stands, both different in age and maintenance, were monitored in the field over two vegetation periods. In parallel, time series of airborne hyperspectral remote sensing data were acquired. For each stand 70 trees were artificially stressed (ring-barked) and 70 trees were used as control trees. The data collected in south-eastern Germany consists of measurements at multiple times and at different scales: (1) crown conditions were visually assessed in the field (2) needle reflectance spectra were acquired in the laboratory using a FieldSpec spectrometer, and (3) hyperspectral airborne data (HySpex) were flown at 0.5 m spatial resolution. We aimed for a simultaneous data acquisition at the three levels. This unique data set was investigated whether any feature can be discriminated to detect vitality loss in trees at an early stage. Several spectral transformations were applied to the needle and tree crown spectra, such as spectral derivatives, vegetation indices and angle indices. All features were examined for their separability (ring-barked vs. control trees) with the Random Forest (RF) classification algorithm. As result, the younger, well maintained forest stand only showed minor changes over the 2-year period, whereas changes in the older forest stand were observable both in the needle and in the hyperspectral tree crown spectra, respectively. These changes could even be detected before changes were visible by field observations. The tree spectral reactions to ring-barking were first noticeable 11 months after ring-barking and 6 weeks before they were visible by field inspection. The most discriminative features for separating the two groups were the reflectance spectra and the spectral derivatives, over the VIs or angle indices. The tree crown spectra of the two groups could be separated by the RF classifier with a 79% overall accuracy at the beginning of the second vegetation period and 1 month later with 92% overall accuracy with high kappa index. The results clearly demonstrate the great potential of hyperspectral remote sensing in detecting early vitality changes of stressed trees. Numéro de notice : A2021-921 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article DOI : 10.1016/j.rse.2021.112676 Date de publication en ligne : 21/09/2021 En ligne : https://doi.org/10.1016/j.rse.2021.112676 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99274 [article]

Flexible Gabor-based superpixel-level unsupervised LDA for hyperspectral image classification / Sen Jia in IEEE Transactions on geoscience and remote sensing, vol 59 n° 12 (December 2021)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 59 n° 12 (December 2021) . - pp 10394 - 10409 Titre : Flexible Gabor-based superpixel-level unsupervised LDA for hyperspectral image classification Type de document : Article/Communication Auteurs : Sen Jia, Auteur ; Qingqing Zhao, Auteur ; Jiayue Zhuang, Auteur ; et al., Auteur Année de publication : 2021 Article en page(s) : pp 10394 - 10409 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] analyse discriminante [Termes IGN] classification non dirigée [Termes IGN] classification par séparateurs à vaste marge [Termes IGN] extraction de traits caractéristiques [Termes IGN] filtre de Gabor [Termes IGN] image hyperspectrale [Termes IGN] ondelette de Gabor [Termes IGN] segmentation d'image [Termes IGN] superpixel Résumé : (auteur) Hyperspectral images encompass abundant information and provide unique characteristics for material classification. However, the labeling of training samples can be challenging in hyperspectral image classification. To address this problem, this study proposes a framework named flexible Gabor-based superpixel-level unsupervised linear discriminant analysis (FG- Su ULDA) to extract the most informative and discriminating features for classification. First, a number of 3-D flexible Gabor filters are rigorously designed using an asymmetric sinusoidal wave to sufficiently characterize the spatial–spectral structure in hyperspectral images. Then, an unsupervised linear discriminant analysis strategy guided by the entropy rate superpixel (ERS) segmentation algorithm, called Su ULDA, is skillfully introduced to reduce the extracted large amount of FG features. The Su ULDA method not only boosts the classification capability but also increases the peculiarity of features, with the aid of superpixel information. Finally, the achieved features are imported to the popular support vector machine classifier. The proposed FG- Su ULDA framework is applied to four real hyperspectral image data sets, and the experiments constantly prove that our FG- Su ULDA is superior to several state-of-the-art methods in both classification performance and computational efficiency, especially with scarce training samples. The codes of this work are available at http://jiasen.tech/papers/ for the sake of reproducibility. Numéro de notice : A2021-872 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2020.3048994 Date de publication en ligne : 20/01/2021 En ligne : https://doi.org/10.1109/TGRS.2020.3048994 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99131 [article]

Bagging and boosting ensemble classifiers for classification of multispectral, hyperspectral and PolSAR data: A comparative evaluation / Hamid Jafarzadeh in Remote sensing, vol 13 n° 21 (November-1 2021)  

Permalink

Adaptive edge preserving maps in Markov random fields for hyperspectral image classification / Chao Pan in IEEE Transactions on geoscience and remote sensing, vol 59 n° 10 (October 2021)  

Permalink

Endmember bundle extraction based on multiobjective optimization / Rong Liu in IEEE Transactions on geoscience and remote sensing, vol 59 n° 10 (October 2021)  

Permalink

Hyperspectral image fusion and multitemporal image fusion by joint sparsity / Han Pan in IEEE Transactions on geoscience and remote sensing, Vol 59 n° 9 (September 2021)  

Permalink

Large-area inventory of species composition using airborne laser scanning and hyperspectral data / Hans Ole Ørka in Silva fennica, vol 55 n° 4 (September 2021)  

Permalink

Multi-task fully convolutional network for tree species mapping in dense forests using small training hyperspectral data / Laura Elena Cué La Rosa in ISPRS Journal of photogrammetry and remote sensing, vol 179 (September 2021)  

Permalink

Two hidden layer neural network-based rotation forest ensemble for hyperspectral image classification / Laxmi Narayana Eeti in Geocarto international, vol 36 n° 16 ([17/08/2021])  

Permalink

Unsupervised band selection of hyperspectral data based on mutual information derived from weighted cluster entropy for snow classification / Divyesh Varade in Geocarto international, vol 36 n° 15 ([10/08/2021])  

Permalink

Target-constrained interference-minimized band selection for hyperspectral target detection / Xiaodi Shang in IEEE Transactions on geoscience and remote sensing, Vol 59 n° 7 (July 2021)  

Permalink

An incremental isomap method for hyperspectral dimensionality reduction and classification / Yi Ma in Photogrammetric Engineering & Remote Sensing, PERS, vol 87 n° 6 (June 2021)  

Permalink