Centre dedocumentation
scientifique

Machine learning models applied to a GNSS sensor network for automated bridge anomaly detection / Nicolas Manzini in Journal of structural engineering, Vol 148 n° 11 (November 2022)  

Public [article]  inJournal of structural engineering > Vol 148 n° 11 (November 2022) Titre : Machine learning models applied to a GNSS sensor network for automated bridge anomaly detection Type de document : Article/Communication Auteurs : Nicolas Manzini, Auteur ; André Orcesi, Auteur ; Christian Thom , Auteur ; Marc-Antoine Brossault, Auteur ; Serge Botton , Auteur ; Miguel Ortiz, Auteur ; John Dumoulin, Auteur Année de publication : 2022 Projets : 2-Pas d'info accessible - article non ouvert / Note générale : bibliographie EN ATTENTE DU DOCUMENT Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Topographie [Termes IGN] apprentissage automatique [Termes IGN] classification par réseau neuronal récurrent [Termes IGN] détection d'anomalie [Termes IGN] ouvrage d'art [Termes IGN] pont [Termes IGN] régression [Termes IGN] série temporelle [Termes IGN] surveillance d'ouvrage [Termes IGN] topométrie de précision Résumé : (auteur) Structural health monitoring (SHM) based on global navigation satellite systems (GNSS) is an interesting solution to provide absolute positions at different locations of a structure in a global reference frame. In particular, low-cost GNSS stations for large-scale bridge monitoring have gained increasing attention these last years because recent experiments showed the ability to achieve a subcentimeter accuracy for continuous monitoring with adequate combinations of antennas and receivers. Technical solutions now allow displacement monitoring of long bridges with a cost-effective deployment of GNSS sensing networks. In particular, the redundancy of observations within the GNSS network with various levels of correlations between the GNSS time series makes such monitoring solution a good candidate for anomaly detection based on machine learning models, using several predictive models for each sensor (based on environmental conditions, or other sensors as input data). This strategy is investigated in this paper based on GNSS time series, and an anomaly indicator is proposed to detect and locate anomalous structural behavior. The proposed concepts are applied to a cable-stayed bridge for illustration, and the comparison between multiple tools highlights recurrent neural networks (RNN) as an effective regression tool. Coupling this tool with the proposed anomaly detection strategy enables one to identify and localize both real and simulated anomalies in the considered data set. Numéro de notice : A2022-672 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Thématique : POSITIONNEMENT Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1061/(ASCE)ST.1943-541X.0003469 En ligne : https://doi.org/10.1061/(ASCE)ST.1943-541X.0003469 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101615 [article]

Attention-based vandalism detection in OpenStreetMap / Nicolas Tempelmeier (2022)  

Public Titre : Attention-based vandalism detection in OpenStreetMap Type de document : Article/Communication Auteurs : Nicolas Tempelmeier, Auteur ; Elena Demidova, Auteur Editeur : New York [Etats-Unis] : Association for computing machinery ACM Année de publication : 2022 Conférence : WWW 2022, ACM Web Conference 2022 25/04/2022 29/04/2022 Lyon online France Proceedings ACM Importance : pp 643 - 651 Note générale : bibliographie Langues : Français (fre) Descripteur : [Vedettes matières IGN] Bases de données localisées [Termes IGN] attention (apprentissage automatique) [Termes IGN] détection d'anomalie [Termes IGN] fiabilité des données [Termes IGN] historique des données [Termes IGN] OpenStreetMap Mots-clés libres : vandalisme Résumé : (auteur) OpenStreetMap (OSM), a collaborative, crowdsourced Web map, is a unique source of openly available worldwide map data, increasingly adopted in Web applications. Vandalism detection is a critical task to support trust and maintain OSM transparency. This task is remarkably challenging due to the large scale of the dataset, the sheer number of contributors, various vandalism forms, and the lack of annotated data. This paper presents Ovid - a novel attention-based method for vandalism detection in OSM. Ovid relies on a novel neural architecture that adopts a multi-head attention mechanism to summarize information indicating vandalism from OSM changesets effectively. To facilitate automated vandalism detection, we introduce a set of original features that capture changeset, user, and edit information. Furthermore, we extract a dataset of real-world vandalism incidents from the OSM edit history for the first time and provide this dataset as open data. Our evaluation conducted on real-world vandalism data demonstrates the effectiveness of Ovid. Numéro de notice : C2022-008 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Communication DOI : 10.1145/3485447.3512224 Date de publication en ligne : 25/04/2022 En ligne : https://doi.org/10.1145/3485447.3512224 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100493

Proceedings of the 1st Conference of the European Association on Quality Control of Bridges and Structures : EUROSTRUCT 2021. An automated machine learning-based approach for structural novelty detection based on SHM / Nicolas Manzini (2022)  

Public Titre de série : Proceedings of the 1st Conference of the European Association on Quality Control of Bridges and Structures : EUROSTRUCT 2021 Titre : An automated machine learning-based approach for structural novelty detection based on SHM Type de document : Article/Communication Auteurs : Nicolas Manzini, Auteur ; Ndeye Mar, Auteur ; Franziska Schmidt, Auteur ; Jean-François Bercher, Auteur ; André Orcesi, Auteur ; Pierre Marchand, Auteur ; Julien Gazeaux , Auteur ; Christian Thom , Auteur Editeur : Springer Nature Année de publication : 2022 Collection : Lecture Notes in Civil Engineering num. 200 Projets : 2-Pas d'info accessible - article non ouvert / Importance : pp 1180 - 1189 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de géodésie spatiale [Termes IGN] apprentissage automatique [Termes IGN] détection d'anomalie [Termes IGN] ouvrage d'art [Termes IGN] pont [Termes IGN] régression multiple [Termes IGN] réseau de capteurs [Termes IGN] résidu [Termes IGN] surveillance d'ouvrage Résumé : (auteur) One major goal of structural health monitoring (SHM) is to detect, and possibly locate, quantify or predict damage on structures. Without detailed knowledge of structural mechanical behavior, data analysis is a complex task and operational monitoring is often limited to the use of more or less arbitrary thresholds. Data-driven techniques, which rely on a statistical analysis of data, have encountered a growing interest over the past two decades. In parallel, SHM is now increasingly considered for several types of structures with the development of low-cost sensors and IoT. In this context, this paper proposes an approach based on multiple automated machine learning-based models for novelty detection and location in monitoring data. This study focuses on the monitoring of large structures with multiple sensors. For each sensor, multiple regression models (based on neural networks) are generated using the same training set, with various input data: internal temperature, environmental conditions, or data from other sensors deployed on the structure. Anomalies are then identified in the dataset based on residuals between model outputs and in situ data. For a given sensor, residuals of all models are then compiled to produce an anomaly indicator. This paper presents some of the results obtained on data acquired from the monitoring of a large concrete bridge. Some anomalies are simulated and added to the dataset to demonstrate the detection performance of the proposed approach. Numéro de notice : C2022-001 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Thématique : POSITIONNEMENT Nature : Communication nature-HAL : ComAvecCL&ActesPubliésIntl DOI : 10.1007/978-3-030-91877-4_134 Date de publication en ligne : 12/12/2021 En ligne : https://doi.org/10.1007/978-3-030-91877-4_134 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99378

An innovative and automated method for characterizing wood defects on trunk surfaces using high-density 3D terrestrial LiDAR data / Van-Tho Nguyen in Annals of Forest Science [en ligne], vol 78 n° 2 (June 2021)  

Public [article]  inAnnals of Forest Science [en ligne] > vol 78 n° 2 (June 2021) . - Article 32 Titre : An innovative and automated method for characterizing wood defects on trunk surfaces using high-density 3D terrestrial LiDAR data Type de document : Article/Communication Auteurs : Van-Tho Nguyen, Auteur ; Thiéry Constant, Auteur ; Francis Colin, Auteur Année de publication : 2021 Article en page(s) : Article 32 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] analyse comparative [Termes IGN] apprentissage automatique [Termes IGN] détection d'anomalie [Termes IGN] données de terrain [Termes IGN] données lidar [Termes IGN] données localisées 3D [Termes IGN] écorce [Termes IGN] Fagus sylvatica [Termes IGN] qualité du bois [Termes IGN] Quercus sessiliflora [Termes IGN] segmentation d'image [Termes IGN] télémétrie laser terrestre [Termes IGN] tronc Résumé : (Auteur) We designed a novel method allowing to automatically detect and measure defects on the surface of trunks including branches, branch scars, and epicormics from terrestrial LiDAR data by using only high-density 3D information. We could automatically detect and measure the defects with a diameter as small as 0.5 cm on either oak (Quercus petraea (Matt.) Liebl.) or beech (Fagus sylvatica L.) trees that display either rough or smooth bark. Context : Ground-based light detection and ranging (LiDAR) technology describes standing trees with a high level of detail. This provides an opportunity to assess standing tree quality and to use this information in forest inventory. Assuming the availability of a very high level of detail, we could extract information about the surface defects, mainly inherited from past ramification and having a strong impact on wood quality. Aims : Within the general framework of the development of a computing method able to detect, identify, and quantify the defects on the trunk surface described from 3D data produced by a terrestrial LiDAR, this study focuses on the relevance of the whole process for two tree species with contrasted bark roughness (Quercus petraea (Matt.) Liebl. and Fagus sylvatica L.) in terms of detection, identification of the defects, and comparison with measurements performed manually on the bark surface. Methods : First, a segmentation algorithm detected singularities on the trunk surface. Next, a Random Forests machine learning algorithm identified the most probable defect type and allowed the elimination of false detections. Finally, we estimated the position, horizontal, and vertical dimensions of each defect from 3D data, and we compared them to those observed directly on the trunk by an operator. Results : The defects were detected and classified with a high accuracy with an average F1 score (harmonic mean of precision and recall) of 0.74. There were differences in computed and observed defect areas, but a much closer agreement for the number of defects. Conclusion : The information about the defects present on the trunk surface measured from terrestrial LiDAR data can be used in an automated procedure for grading standing trees or roundwoods. Numéro de notice : A2021-326 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1007/s13595-020-01022-3 Date de publication en ligne : 01/04/2021 En ligne : https://doi.org/10.1007/s13595-020-01022-3 Format de la ressource électronique : URL Article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97484 [article]

Initialization methods of convolutional neural networks for detection of image manipulations / Ivan Castillo Camacho (2021)  

Public Titre : Initialization methods of convolutional neural networks for detection of image manipulations Titre original : Méthodes d'initialisation des réseaux de neurones convolutifs pour la détection des manipulations d'images Type de document : Thèse/HDR Auteurs : Ivan Castillo Camacho, Auteur ; Kai Wang, Directeur de thèse Editeur : Grenoble [France] : Université Grenoble Alpes Année de publication : 2021 Importance : 145 p. Format : 21 x 30 cm Note générale : bibliographie Thèse pour obtenir le grade de Docteur de l'Université Grenoble, spécialité : signal, image, paroles, télécoms Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] altération [Termes IGN] apprentissage profond [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] covariance [Termes IGN] détection d'anomalie [Termes IGN] estompage [Termes IGN] filtre passe-haut [Termes IGN] flux de données [Termes IGN] infraction [Termes IGN] manipulation de données [Termes IGN] qualité des données [Termes IGN] retouche [Termes IGN] variance Index. décimale : THESE Thèses et HDR Résumé : (auteur) Fake images and videos have engulfed mass communication media. This is not something recent, manipulations and forgeries have occurred since the advent of photography itself. These alterations can go from innocent retouches in an attempt to make an image visually attractive to the spread of misleading information or even the use of false media in legal instances. Accordingly, the creation of methods that can help us assure the authenticity of an image presented as non-modified is of paramount importance. In this thesis, we aim at detecting image manipulation operations using deep learning techniques. We present three methods showing the progression of our work under one common objective, i.e, the design and test of Convolutional Neural Network (CNN) initialization methods for image forensic problems with a variance stability focus for the output of a CNN layer.First, we carry out an extensive review of the state of the art in deep-learning-based methods for image forensics. From this review we can confirm that the first layer of a CNN has big impact on the final performance. Specifically, the initialization used on the first-layer filters plays an important role that should be in line with the image forensic task in hand.As our first attempt to address this research problem, we propose a low-complexity initialization method for CNNs. Taking advantage of previous methods designed for the computer vision field, we extend the popular Xavier method to design a filter that would provide variance stability after a convolution operation. This method generates a set of random high-pass filters for the initialization of a CNN's first layer. These filters allow us to better identify forensic traces which usually lie towards the high-frequency part of the image.This first approach constitutes a good staring point of our work. However, a wrong assumption, largely utilized in the research community, was made. This is corrected in our second method where we follow a different data-dependent approach and take into consideration the real statistical properties of natural images. Accordingly, we propose a scaling method for first-layer filters which can cope well with different CNN initialization algorithms. The objective remains in keeping the stability of the variance of data flow in a CNN. We also present theoretical and experimental studies on the output variance for convolutional filter, which are the basis of our proposed data-dependent scaling.Next we describe a revisited version of our first proposal now with a corrected assumption on the statistics of natural images. More precisely, we propose an improved random high-pass initialization method which does not explicitly compute the statistics of input data. We believe that such a ``data-independent'' approach has higher flexibility and broader application range than our second method in situations where the computation of input statistics is not possible.Our proposed methods are tested over several image forensic problems and different CNN architectures.Finally, during all this thesis work we took part in a challenge competition of image forgery detection organized by the French National Research Agency and the French Directorate General of Armaments. We explain in the Appendix the objectives of the challenge along with a brief description of our work conducted for the competition. Note de contenu : 1- Introduction 2- Background knowledge and state of the art 3- Random high-pass initialization 4- Data-dependent initialization 5- Revisiting the random high-pass initialization 6- Conclusions and perspectives Numéro de notice : 28437 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Thèse française Note de thèse : Thèse de Doctorat : signal, image, paroles, télécoms : Grenoble : 2021 DOI : sans En ligne : https://tel.archives-ouvertes.fr/tel-03346063/document Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=98833

Network-constrained bivariate clustering method for detecting urban black holes and volcanoes / Qiliang Liu in International journal of geographical information science IJGIS, vol 34 n° 10 (October 2020)  

Permalink

Un été brûlant sous l’oeil des satellites / Laurent Polidori in Géomètre, n° 2173 (octobre 2019)

Permalink

Sea level prediction in the Yellow Sea from satellite altimetry with a combined least squares-neural network approach / Jian Zhao in Marine geodesy, vol 42 n° 4 (July 2019)  

Permalink

The necessary yet complex evaluation of 3D city models: a semantic approach / Oussama Ennafii (2019)    

Permalink

Le vandalisme dans l’information géographique volontaire, détection de l’IG volontaire vandalisée : du concept à la détection non supervisée d’anomalie / Quy Thy Truong in Revue internationale de géomatique, vol 29 n° 1 (janvier - mars 2019)  

Permalink

Intra-annual phenology for detecting understory plant invasion in urban forests / Kunwar K. Singh in ISPRS Journal of photogrammetry and remote sensing, vol 142 (August 2018)  

Permalink

Towards vandalism detection in OpenStreetMap through a data driven approach [short paper] / Quy Thy Truong (2018)    

Permalink

Band subset selection for anomaly detection in hyperspectral imagery / Lin Wang in IEEE Transactions on geoscience and remote sensing, vol 55 n° 9 (September 2017)  

Permalink

Detection of inconsistencies in geospatial data with geostatistics / Adriana Maria Rocha Trancoso Santos in Boletim de Ciências Geodésicas, vol 23 n° 2 (abr - jun 2017)  

Permalink

A spatial anomaly points and regions detection method using multi-constrained graphs and local density / Yan Shi in Transactions in GIS, vol 21 n° 2 (April 2017)  

Permalink