Centre dedocumentation
scientifique

Deep learning in denoising of micro-computed tomography images of rock samples / Mikhail Sidorenko in Computers & geosciences, vol 151 (June 2021)  

Public [article]  inComputers & geosciences > vol 151 (June 2021) . - n° 104716 Titre : Deep learning in denoising of micro-computed tomography images of rock samples Type de document : Article/Communication Auteurs : Mikhail Sidorenko, Auteur ; Denis Orlov, Auteur ; Mohammad Ebadi, Auteur ; et al., Auteur Année de publication : 2021 Article en page(s) : n° 104716 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image [Termes IGN] accentuation d'image [Termes IGN] apprentissage profond [Termes IGN] classification dirigée [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] filtrage du bruit [Termes IGN] filtre passe-bande [Termes IGN] roche [Termes IGN] tomographie Résumé : (auteur) Nowadays, the advantages of Digital Rock Physics (DRP) are well known and widely applied in comprehensive core analysis. It is also known that the quality of the 3D pore scale model drastically influences the results of rock properties simulation, which makes the preprocessing stage of DRP very important. In this work, we consider the application of Deep Convolutional Neural Networks (CNNs) for the preprocessing of CT images, specifically for denoising, in two setups - conventional fully-supervised learning and the self-supervised learning, when the only available data is the noisy images. To train CNNs in a supervised setup, we use images processed by a combination of bilateral and bandpass filters. We trained CNNs of the same architecture with different loss functions to find out how the choice of a loss function influences the model's performance. Some of the obtained CNNs yielded the highest quality in terms of full-reference and no-reference metrics and significant histogram effect (bimodal intensity distribution). Images denoised with these models were qualitatively and quantitatively better than the reference “ground truth” images used for training. We use the Deep Image Prior algorithm to train denoising models in a self-supervised setup. The obtained models are much better than ones obtained in fully-supervised setup, but are too slow, as they are optimization-based rather than feed-forward. Such an algorithm can be used in the dataset generation for feed-forward meta-models. These results could help to develop an AI-based instrument to build high-quality 3D segmented models of rocks for DRP applications. Numéro de notice : A2021-389 Affiliation des auteurs : non IGN Thématique : IMAGERIE/INFORMATIQUE Nature : Article DOI : 10.1016/j.cageo.2021.104716 Date de publication en ligne : 02/03/2021 En ligne : https://doi.org/10.1016/j.cageo.2021.104716 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97672 [article]

Analyzing and improving Graph Neural Networks / Guillaume Renton (2021)  

Public Titre : Analyzing and improving Graph Neural Networks Type de document : Thèse/HDR Auteurs : Guillaume Renton, Auteur ; Sébastien Adam, Auteur Editeur : Université de Rouen Année de publication : 2021 Importance : 130 p. Format : 21 x 30 cm Note générale : bibliographie Thèse présentée pour obtenir le diplôme de Doctorat de l'Université de Rouen Normandie, spécialité Informatique Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Intelligence artificielle [Termes IGN] apprentissage profond [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] filtre passe-bande [Termes IGN] filtre spectral [Termes IGN] noeud [Termes IGN] réseau neuronal de graphes [Termes IGN] théorie des graphes [Termes IGN] transformation de Laplace [Termes IGN] transformation inverse Index. décimale : THESE Thèses et HDR Résumé : (auteur) Although theorised about fifteen years ago, the scientific community’s interest for graph neural networks has only really taken off recently. Those models aim to transpose the representation learning capacity inherent in deep neural network onto graph data, via the learning of hidden states associated with the graph nodes. These hidden states are computed and updated according to the information contained in the neighborhoud of each node.This recent interest for graph neural networks (GNNs) has led to a "jungle" of models and frameworks, making this field of research sometimes confusing. Historically, two main strategies have been explored : the spatial GNNs on one side and the spectral GNNs on the other side. Spatial GNNs, sometimes also called Message Passing Neural Network, are based on the computation of a message which agregates the information contained in the neighborhoud of each node. On the other side, spectral GNNs are based on the spectral graph theory and thus on the graph Laplacian. The eigendecomposition of the graph Laplacian allows to define a graph Fourier transform and its inverse. From these transforms, different filters can be applied on the graph, leading to similar result than filtering on images or signals. In this thesis, we begin by introducing a third category, called spectral rooted spatial convolution. Indeed, some recent methods are taking root in the spectral domain while avoiding to compute the eigendecomposition of the graph Laplacian. This third category leads to question about the fundamental difference between spectral and spatial GNNs. We answer this question by proposing a general model unifying both strategies, showing notably that spectral GNNs are a particular case of spatial GNNs. This unified model also allowed us to propose a spectral analysis of some popular GNNs in the scientificcommunitic, namely GCN, GIN, GAT, ChebNet and CayleyNet. This analysis shows that spatial models are limited to low-pass and high-pass filtering, while spectral models can produce any kind of filters. Those results are then found with the presentation of a toy problem, showing in the first instance the limitation of spatial models to define pass-band filters, and the importance of designing such filters. Those results have led us to propose a method allowing any kind of filter, while limiting the network’s number of parameters. Indeed, even though spectral models are able to design any kind of filtering, each new filter require the add of a new weight matrix in the neural network. In order to reduce the number of parameters, we propose to adapt Depthwise Separable Convolution to graphs through a method called Depthwise Separable Graph Convolution Network. This method is evaluated on both transductive and inductive learning, outperforming state-of-the-arts results.Finally, we propose a method defined in the spatial domain in order to take into account edge attributes. Indeed, this issue has been little studied by the scientific community, and the number of methods allowing to include edge attributes is very small. Our proposal, called Edge Embedding Graph Neural Network, consists in embedding edge attributes into a new space through a first neural network, before using the extracted features in a GNN. This method is evaluated on a particular problem of symbol detection in a graph. Note de contenu : 1- Introduction 2- Background 3- What is a Graph Neural Network ? 4- Graph Neural Networks: Are they Spectral or Spatial ? 5- Depthwise Separable Graph Convolution Network 6- Edge Embedding Graph Neural Network 7- Conclusion Numéro de notice : 15259 Affiliation des auteurs : non IGN Thématique : INFORMATIQUE/MATHEMATIQUE Nature : Thèse française Note de thèse : Thèse de Doctorat : Informatique : Rouen : 2021 Organisme de stage : Laboratoire LITIS DOI : sans En ligne : https://tel.hal.science/tel-03346018/ Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100612

Low-frequency desert noise intelligent suppression in seismic data based on multiscale geometric analysis convolutional neural network / Yuxing Zhao in IEEE Transactions on geoscience and remote sensing, vol 58 n° 1 (January 2020)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 58 n° 1 (January 2020) . - pp 650 - 665 Titre : Low-frequency desert noise intelligent suppression in seismic data based on multiscale geometric analysis convolutional neural network Type de document : Article/Communication Auteurs : Yuxing Zhao, Auteur ; Yue Li, Auteur ; Baojun Yang, Auteur Année de publication : 2020 Article en page(s) : pp 650 - 665 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement du signal [Termes IGN] algorithme de filtrage [Termes IGN] analyse multiéchelle [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] désert [Termes IGN] enregistrement de données [Termes IGN] filtrage du bruit [Termes IGN] filtre passe-bande [Termes IGN] interruption du signal [Termes IGN] lutte contre le bruit [Termes IGN] rapport signal sur bruit [Termes IGN] reconstruction du signal [Termes IGN] séisme Résumé : (auteur) Existing denoising algorithms often need to meet some premise assumptions and applicable conditions, such as the signal-to-noise ratio (SNR) cannot be too low, and the noise needs to obey a specific distribution (such as Gaussian distribution) and to satisfy some properties (such as stationarity). For the desert noise that shares the same frequency band with the effective signal and has complex characteristics (nonlinear, nonstationary, and non-Gaussian), it is difficult to find a universally applicable method. In response to this problem, a multiscale geometric analysis (MGA) convolutional neural network (CNN) is proposed in this article. One of the most important features of the CNN is that it can extract data-rich intrinsic information from the training set without relying on a priori assumption. By introducing the CNN into the MGA, a new kind of denoising method can be created, which can achieve good results even under a low SNR. This article takes the nonsubsampled contourlet transform as an example to create a denoising network named NC-CNN for high-efficiency and intelligent denoising of desert seismic data. The processing results of synthetic seismic records and field seismic records prove that NC-CNN can effectively suppress the low-frequency noise (random noise and surface wave), and the effective signal almost has no energy loss. In addition, the reconstruction ability of the missing signals is also an advantage of this method. Numéro de notice : A2020-076 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2019.2938836 Date de publication en ligne : 24/09/2019 En ligne : https://doi.org/10.1109/TGRS.2019.2938836 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=94608 [article]

FFT swept filtering: a bias-free method for processing fringe signals in absolute gravimeters / Petr Křen in Journal of geodesy, vol 93 n° 2 (February 2019)  

Public [article]  inJournal of geodesy > vol 93 n° 2 (February 2019) . - pp 219 - 227 Titre : FFT swept filtering: a bias-free method for processing fringe signals in absolute gravimeters Type de document : Article/Communication Auteurs : Petr Křen, Auteur ; Vojtech Pálinkáš, Auteur ; Pavel Mašika, Auteur ; Miloš Val’ko, Auteur Année de publication : 2019 Article en page(s) : pp 219 - 227 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géodésie physique [Termes IGN] distorsion du signal [Termes IGN] erreur systématique [Termes IGN] filtre passe-bande [Termes IGN] gravimètre absolu [Termes IGN] pesanteur terrestre [Termes IGN] régression linéaire [Termes IGN] transformation rapide de Fourier Résumé : (auteur) Absolute gravimeters, based on laser interferometry, are widely used for many applications in geoscience and metrology. Although currently the most accurate FG5 and FG5X gravimeters declare standard uncertainties at the level of 2−3μGal, their inherent systematic errors affect the gravity reference determined by international key comparisons based predominately on the use of FG5-type instruments. The measurement results for FG5-215 and FG5X-251 clearly showed that the measured g-values depend on the size of the fringe signal and that this effect might be approximated by a linear regression with a slope of up to 0.030μGal/mV. However, these empirical results do not enable one to identify the source of the effect or to determine a reasonable reference fringe level for correcting g-values in an absolute sense. Therefore, both gravimeters were equipped with new measuring systems (according to Křen et al. in Metrologia 53:27–40, 2016. https://doi.org/10.1088/0026-1394/53/1/27 applied for FG5), running in parallel with the original systems. The new systems use an analogue-to-digital converter HS5 to digitize the fringe signal and a new method of fringe signal analysis based on FFT swept bandpass filtering. We demonstrate that the source of the fringe size effect is connected to a distortion of the fringe signal due to the electronic components used in the FG5(X) gravimeters. To obtain a bias-free g-value, the FFT swept method should be applied for the determination of zero-crossings. A comparison of g-values obtained from the new and the original systems clearly shows that the original system might be biased by approximately 3−5μGal due to improperly distorted fringe signal processing. Numéro de notice : A2019-079 Affiliation des auteurs : non IGN Thématique : POSITIONNEMENT Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1007/s00190-018-1154-y Date de publication en ligne : 19/05/2018 En ligne : https://doi.org/10.1007/s00190-018-1154-y Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=92197 [article]

UAV sensors for environmental monitoring / Felipe Gonzalez Toro (2018)  

Public Titre : UAV sensors for environmental monitoring Type de document : Monographie Auteurs : Felipe Gonzalez Toro, Éditeur scientifique ; Antonios Tsourdos, Éditeur scientifique Editeur : Bâle [Suisse] : Multidisciplinary Digital Publishing Institute MDPI Année de publication : 2018 Importance : 660 p. Format : 17 x 25 cm ISBN/ISSN/EAN : 978-3-03842-753-7 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] accident de la route [Termes IGN] biodiversité [Termes IGN] capteur aérien [Termes IGN] capteur terrestre [Termes IGN] détection de cible [Termes IGN] filtre passe-bande [Termes IGN] image captée par drone [Termes IGN] image hyperspectrale [Termes IGN] image optique [Termes IGN] image radar moirée [Termes IGN] instrument embarqué [Termes IGN] photogrammétrie aérienne [Termes IGN] prévention des risques [Termes IGN] surveillance écologique [Termes IGN] surveillance hydrologique Résumé : (éditeur) The rapid development and growth of UAVs as a remote sensing platform, as well as advances in the miniaturization of instrumentation and data systems, are catalyzing a renaissance in remote sensing in a variety of fields and disciplines from precision agriculture to ecology, atmospheric research, and disaster response. This Special Issue was open for submissions that highlight advances in the development and use of sensors deployed on UAVs. Topics include, but were not limited, to: - Optical, multi-spectral, hyperspectral, laser, and optical SAR technologies - Gas analyzers and sensors - Artificial intelligence and data mining based strategies from UAVs - UAV onboard data storage, transmission, and retrieval - Collaborative strategies and mechanisms to control multiple UAVs and sensor networks - UAV sensor applications: precision agriculture; pest detection, forestry, mammal species tracking search and rescue; target tracking, the monitoring of the atmosphere; chemical, biological, and natural disaster phenomena; fire prevention, flood prevention; volcanic monitoring, pollution monitoring, micro-climates and land use Note de contenu : Preface 1- UAV-based photogrammetry and integrated technologies for architectural applications—methodological strategies for the after-quake survey of vertical structures in Mantua (Italy) 2- Towards the development of a low cost airborne sensing system to monitor dust particles after blasting at open-pit mine sites 3- Multi-UAV routing for area coverage and remote sensing with minimum time 4- UAV deployment exercise for mapping purposes: evaluation of emergency response applications 5- Automated identification of river hydromorphological features using UAV high resolution aerial imagery 6- Autonomous aerial refueling ground test demonstration—a sensor-in-the-loop, non-tracking method 7- A new calibration method using low cost MEM IMUs to verify the performance of UAV-borne MMS payloads 8- Adaptive environmental source localization and tracking with unknown permittivity and pathloss coefficients 9- Vision-based detection and distance estimation of micro unmanned aerial vehicles 10- Unmanned aerial vehicles (UAVs) and artificial intelligence revolutionizing wildlife monitoring and conservation 11- UAVs task and motion planning in the presence of obstacles and prioritized targets 12- Towards the development of a smart flying sensor: Illustration in the field of precision agriculture 13- Flight test result for the ground-based radio navigation system sensor with an unmanned air vehicle 12- Multisensor super resolution using directionally-adaptive regularization for UAV images 13- UAV control on the basis of 3D landmark bearing-only observations 14- Cooperative surveillance and pursuit using unmanned aerial vehicles and unattended ground sensors 15- A multispectral image creating method for a new airborne four-camera system with different bandpass filters 16- Vision and control for UAVs: A survey of general methods and of inexpensive platforms for infrastructure inspection 17- Feasibility of using synthetic aperture radar to aid UAV navigation 18- Towards an autonomous vision-based unmanned aerial system against wildlife poachers 19- Formation flight of multiple UAVs via onboard sensor information sharing 20- Mini-UAV based sensory system for measuring environmental variables in greenhouses 21- Mini-UAV based sensory system for measuring environmental variables in greenhouses 22- Dual-stack single-radio communication architecture for UAV acting as a mobile node to collect data in WSNs 23- Development and evaluation of a UAV-photogrammetry system for precise 3D environmental modeling 24- Prototyping a GNSS-based passiveRadar for UAVs: An instrument to classify the waterContent feature of lands 25- Enabling UAV navigation with sensor and environmental uncertainty in cluttered and GPS-denied environments 26- UAV-based estimation of carbon exports from heterogeneous soil landscapes—a case study from the carboZALF experimental area 27- Wavelength-adaptive dehazing using histogram merging-based classification for UAV images 28- A Space-Time Network-Based Modeling Framework for Dynamic Unmanned Aerial Vehicle Routing in Traffic Incident Monitoring Applications Numéro de notice : 25930 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Monographie En ligne : https://doi.org/10.3390/books978-3-03842-754-4 Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96199

Filter design for GOCE gravity gradients / Zs. Polgár Polgár in Geocarto international, vol 28 n° 1-2 (February - May 2013)  

Permalink

Interference suppression algorithm for SAR based on time-frequency transform / S. Zhang in IEEE Transactions on geoscience and remote sensing, vol 49 n° 10 Tome 1 (October 2011)  

Permalink

An empirical investigation of cross-sensor relationships of NDVI and red/near-infrared reflectance using EO-1 Hyperion data / T. Miura in Remote sensing of environment, vol 100 n° 2 (30 January 2006)  

Permalink

Microstrip filters for RF/microwave applications / J.S. Hong (2001)

Permalink

Modex / Stéphane Mallat (1998)

Permalink