Centre de documentation
scientifique

Machine learning and geodesy: A survey / Jemil Butt in Journal of applied geodesy, vol 15 n° 2 (April 2021)  

Public [article]  inJournal of applied geodesy > vol 15 n° 2 (April 2021) . - pp 117 - 133 Titre : Machine learning and geodesy: A survey Type de document : Article/Communication Auteurs : Jemil Butt, Auteur ; Andreas Wieser, Auteur ; Zan Gojcic, Auteur ; Caifa Zhou, Auteur Année de publication : 2021 Article en page(s) : pp 117 - 133 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Intelligence artificielle [Termes IGN] analyse de données [Termes IGN] apprentissage automatique [Termes IGN] données géodésiques [Termes IGN] espace de Hilbert [Termes IGN] méthode fondée sur le noyau Résumé : (Auteur) The goal of classical geodetic data analysis is often to estimate distributional parameters like expected values and variances based on measurements that are subject to uncertainty due to unpredictable environmental effects and instrument specific noise. Its traditional roots and focus on analytical solutions at times require strong prior assumptions regarding problem specification and underlying probability distributions that preclude successful application in practical cases for which the goal is not regression in presence of Gaussian noise. Machine learning methods are more flexible with respect to assumed regularity of the input and the form of the desired outputs and allow for nonparametric stochastic models at the cost of substituting easily analyzable closed form solutions by numerical schemes. This article aims at examining common grounds of geodetic data analysis and machine learning and showcases applications of algorithms for supervised and unsupervised learning to tasks concerned with optimal estimation, signal separation, danger assessment and design of measurement strategies that occur frequently and naturally in geodesy. Numéro de notice : A2021-321 Affiliation des auteurs : non IGN Thématique : INFORMATIQUE/POSITIONNEMENT Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1515/jag-2020-0043 Date de publication en ligne : 20/02/2021 En ligne : https://doi.org/10.1515/jag-2020-0043 Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97478 [article]

Fusion of ground penetrating radar and laser scanning for infrastructure mapping / Dominik Merkle in Journal of applied geodesy, vol 15 n° 1 (January 2021)  

Public [article]  inJournal of applied geodesy > vol 15 n° 1 (January 2021) . - pp 31 - 45 Titre : Fusion of ground penetrating radar and laser scanning for infrastructure mapping Type de document : Article/Communication Auteurs : Dominik Merkle, Auteur ; Carsten Frey, Auteur ; Alexander Reiterer, Auteur Année de publication : 2021 Article en page(s) : pp 31 - 45 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications photogrammétriques [Termes IGN] base de données localisées 3D [Termes IGN] données lidar [Termes IGN] espace de Hilbert [Termes IGN] lasergrammétrie [Termes IGN] lever souterrain [Termes IGN] radar pénétrant GPR [Termes IGN] radargrammétrie [Termes IGN] réseau technique souterrain [Termes IGN] semis de points [Termes IGN] sous-sol [Termes IGN] surface du sol [Termes IGN] système de numérisation mobile Résumé : (auteur) Mobile mapping vehicles, equipped with cameras, laser scanners (in this paper referred to as light detection and ranging, LiDAR), and positioning systems are limited to acquiring surface data. However, in this paper, a method to fuse both LiDAR and 3D ground penetrating radar (GPR) data into consistent georeferenced point clouds is presented, allowing imaging both the surface and subsurface. Objects such as pipes, cables, and wall structures are made visible as point clouds by thresholding the GPR signal’s Hilbert envelope. The results are verified with existing utility maps. Varying soil conditions, clutter, and noise complicate a fully automatized approach. Topographic correction of the GPR data, by using the LiDAR data, ensures a consistent ground height. Moreover, this work shows that the LiDAR point cloud, as a reference, increases the interpretability of GPR data and allows measuring distances between above ground and subsurface structures. Numéro de notice : A2021-044 Affiliation des auteurs : non IGN Thématique : IMAGERIE/POSITIONNEMENT Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1515/jag-2020-0004 Date de publication en ligne : 06/11/2020 En ligne : https://doi.org/10.1515/jag-2020-0004 Format de la ressource électronique : url article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96771 [article]

Spheroidal spline interpolation and its application in geodesy / Mostafa Kiani in Geodesy and cartography, vol 46 n° 3 (October 2020)  

Public [article]  inGeodesy and cartography > vol 46 n° 3 (October 2020) . - pp 123 - 135 Titre : Spheroidal spline interpolation and its application in geodesy Type de document : Article/Communication Auteurs : Mostafa Kiani, Auteur ; Nabi Chegini, Auteur ; Abdolreza Safari, Auteur Année de publication : 2020 Article en page(s) : pp 123 - 135 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géodésie physique [Termes IGN] analyse harmonique [Termes IGN] espace de Hilbert [Termes IGN] fonction de Green [Termes IGN] fonction spline d'interpolation [Termes IGN] force de gravitation [Termes IGN] optimisation (mathématiques) [Termes IGN] sphèroïde Résumé : (auteur) The aim of this paper is to study the spline interpolation problem in spheroidal geometry. We follow the minimization of the norm of the iterated Beltrami-Laplace and consecutive iterated Helmholtz operators for all functions belong-ing to an appropriate Hilbert space defined on the spheroid. By exploiting surface Green’s functions, reproducing kernels for discrete Dirichlet and Neumann conditions are constructed in the spheroidal geometry. According to a complete system of surface spheroidal harmonics, generalized Green’s functions are also defined. Based on the minimization problem and corresponding reproducing kernel, spline interpolant which minimizes the desired norm and satisfies the given discrete conditions is defined on the spheroidal surface. The application of the results in Geodesy is explained in the gravity data interpolation over the globe. Numéro de notice : A2020-783 Affiliation des auteurs : non IGN Thématique : MATHEMATIQUE/POSITIONNEMENT Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.3846/gac.2020.11316 En ligne : https://doi.org/10.3846/gac.2020.11316 Format de la ressource électronique : url article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96479 [article]

Foundations of deep convolutional models through kernel methods / Alberto Bietti (2019)  

Public Titre : Foundations of deep convolutional models through kernel methods Type de document : Thèse/HDR Auteurs : Alberto Bietti, Auteur ; Julien Mairal, Directeur de thèse Editeur : Grenoble : Université de Grenoble Année de publication : 2019 Importance : 194 p. Format : 21 x 30 cm Note générale : bibliographie Thèse pour obtenir le grade de Docteur de la Communauté Université Grenoble Alpes, Spécialité : Mathématiques Appliquées Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Intelligence artificielle [Termes IGN] apprentissage automatique [Termes IGN] apprentissage profond [Termes IGN] approche hiérarchique [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] espace de Hilbert [Termes IGN] état de l'art [Termes IGN] invariance [Termes IGN] jeu de données [Termes IGN] méthode fondée sur le noyau [Termes IGN] optimisation (mathématiques) [Termes IGN] Perceptron multicouche Index. décimale : THESE Thèses et HDR Résumé : (auteur) The increased availability of large amounts of data, from images in social networks, speech waveforms from mobile devices, and large text corpuses, to genomic and medical data, has led to a surge of machine learning techniques. Such methods exploit statistical patterns in these large datasets for making accurate predictions on new data. In recent years, deep learning systems have emerged as a remarkably successful class of machine learning algorithms, which rely on gradient-based methods for training multi-layer models that process data in a hierarchical manner. These methods have been particularly successful in tasks where the data consists of natural signals such as images or audio; this includes visual recognition, object detection or segmentation, and speech recognition.For such tasks, deep learning methods often yield the best known empirical performance; yet, the high dimensionality of the data and large number of parameters of these models make them challenging to understand theoretically. Their success is often attributed in part to their ability to exploit useful structure in natural signals, such as local stationarity or invariance, for instance through choices of network architectures with convolution and pooling operations. However, such properties are still poorly understood from a theoretical standpoint, leading to a growing gap between the theory and practice of machine learning. This thesis is aimed towards bridging this gap, by studying spaces of functions which arise from given network architectures, with a focus on the convolutional case. Our study relies on kernel methods, by considering reproducing kernel Hilbert spaces (RKHSs) associated to certain kernels that are constructed hierarchically based on a given architecture. This allows us to precisely study smoothness, invariance, stability to deformations, and approximation properties of functions in the RKHS. These representation properties are also linked with optimization questions when training deep networks with gradient methods in some over-parameterized regimes where such kernels arise. They also suggest new practical regularization strategies for obtaining better generalization performance on small datasets, and state-of-the-art performance for adversarial robustness on image tasks. Note de contenu : 1- Introduction 2- Invariance, Stability to deformations, and complexity of deep convolutional representations 3- A kernel perspective on regularization and robustness of deep neural networks 4- Links with optimization: inductive bias of neural tangent kernels 5- Invariance and stability through regularization: a stochastic optimization algorithm for data augmentation 6- Conclusion and perspectives Numéro de notice : 25833 Affiliation des auteurs : non IGN Thématique : INFORMATIQUE/MATHEMATIQUE Nature : Thèse française Note de thèse : Thèse de Doctorat : Mathématiques Appliquées : Grenoble Alpes : 2019 nature-HAL : Thèse DOI : sans En ligne : https://hal.archives-ouvertes.fr/tel-02543073/ document Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=95171

3D Hilbert space filling curves in 3D city modeling for faster spatial queries / Uznir Ujang in International journal of 3-D information modeling, vol 3 n° 2 (April - June 2014)  

Public [article]  inInternational journal of 3-D information modeling > vol 3 n° 2 (April - June 2014) . - pp 1 - 18 Titre : 3D Hilbert space filling curves in 3D city modeling for faster spatial queries Type de document : Article/Communication Auteurs : Uznir Ujang, Auteur ; François Anton, Auteur ; Suhaibah Azri, Auteur ; et al., Auteur Année de publication : 2014 Article en page(s) : pp 1 - 18 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications photogrammétriques [Termes IGN] CityGML [Termes IGN] courbe de Hilbert [Termes IGN] données ouvertes [Termes IGN] espace de Hilbert [Termes IGN] modélisation 3D [Termes IGN] reconstruction 3D du bâti [Termes IGN] requête spatiale [Termes IGN] vitesse [Termes IGN] XML Résumé : (Auteur) The advantages of three dimensional (3D) city models can be seen in various applications including photogrammetry, urban and regional planning, computer games, etc. They expand the visualization and analysis capabilities of Geographic Information Systems on cities, and they can be developed using web standards. However, these 3D city models consume much more storage compared to two dimensional (2 D) spatial data. They involve extra geometrical and topological information together with semantic data. Without a proper spatial data clustering method and its corresponding spatial data access method, retrieving portions of and especially searching these 3D city models, will not be done optimally. Even though current developments are based on an open data model allotted by the Open Geospatial Consortium (OGC) called CityGML, its XML-based structure makes it challenging to cluster the 3D urban objects. In this research, the authors propose an opponent data constellation technique of space-filling curves (3D Hilbert curves) for 3D city model data representation. Unlike previous methods, that try to project 3D or n-dimensional data down to 2D or 3D using Principal Component Analysis (PCA) or Hilbert mappings, in this research, they extend the Hilbert space-filling curve to one higher dimension for 3D city model data implementations. The query performance was tested for single object, nearest neighbor and range search queries using a CityGML dataset of 1,000 building blocks and the results are presented in this paper. The advantages of implementing space-filling curves in 3D city modeling will improve data retrieval time by means of optimized 3D adjacency, nearest neighbor information and 3D indexing. The Hilbert mapping, which maps a sub-interval of the ([0,1]) interval to the corresponding portion of the d-dimensional Hilbert's curve, preserves the Lebesgue measure and is Lipschitz continuous. Depending on the applications, several alternatives are possible in order to cluster spatial data together in the third dimension compared to its clustering in 2 D. Numéro de notice : A2014-652 Affiliation des auteurs : non IGN Thématique : IMAGERIE/INFORMATIQUE Nature : Article DOI : 10.4018/ij3dim.2014040101 En ligne : http://dx.doi.org/10.4018/ij3dim.2014040101 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=75102 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
138-2014021	RAB	Revue	Centre de documentation	En réserve L003	Disponible

Dynamic positioning configuration and its first-order optimization / Shuqiang Xue in Journal of geodesy, vol 88 n° 2 (February 2014)  

Permalink

Blind evaluation of location based queries using space transformation to preserve location privacy / Ali Khshgozaran in Geoinformatica, vol 17 n° 4 (October 2013)  

Permalink

Spectral unmixing in multiple-kernel Hilbert space for hyperspectral imagery / Yanfeng Gu in IEEE Transactions on geoscience and remote sensing, vol 51 n° 7 Tome 1 (July 2013)  

Permalink

Mathématiques pour le traitement du signal / M. Bergounioux (2010)

Permalink

Regional gravity field modeling using airborne gravimetry data / Bas Alberts (2009)  

Permalink

Function bases on Manifolds / Bruno Vallet (2008)  

Permalink

Mathématiques tout-en-un PC PSI / Claude Deschamps (2005)

Permalink

A minimum energy condition for the inverse gravimetric problem / Olivier Jamet (2005)  

Permalink

Wavelets in geodesy and geodynamics / W. Keller (2004)

Permalink

Analysis of multi-dimensional space-filling curves / M.F. Mokbel in Geoinformatica, vol 7 n° 3 (September - November 2003)  

Permalink