Centre dedocumentation
scientifique

Combining deep learning and mathematical morphology for historical map segmentation / Yizi Chen (2021)  

Public contenu dans ArXiv / Université Cornell (Etats-Unis) (1991)   Titre : Combining deep learning and mathematical morphology for historical map segmentation Type de document : Article/Communication Auteurs : Yizi Chen, Auteur ; Edwin Carlinet, Auteur ; Joseph Chazalon, Auteur ; Clément Mallet , Auteur ; Bertrand Duménieu , Auteur ; Julien Perret , Auteur Editeur : Ithaca [New York - Etats-Unis] : ArXiv - Université Cornell Année de publication : 2021 Projets : SODUCO / Perret, Julien Note générale : bibliographie soumis à DGMM 2021 Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géomatique [Termes descripteurs IGN] analyse diachronique [Termes descripteurs IGN] apprentissage profond [Termes descripteurs IGN] carte ancienne [Termes descripteurs IGN] chaîne de traitement [Termes descripteurs IGN] classification par réseau neuronal convolutif [Termes descripteurs IGN] détection d'objet [Termes descripteurs IGN] données maillées [Termes descripteurs IGN] morphologie mathématique [Termes descripteurs IGN] vectorisation Résumé : (auteur) The digitization of historical maps enables the study of ancient, fragile, unique, and hardly accessible information sources. Main map features can be retrieved and tracked through the time for subsequent thematic analysis. The goal of this work is the vectorization step, i.e., the extraction of vector shapes of the objects of interest from raster images of maps. We are particularly interested in closed shape detection such as buildings, building blocks, gardens, rivers, etc. in order to monitor their temporal evolution. Historical map images present significant pattern recognition challenges. The extraction of closed shapes by using traditional Mathematical Morphology (MM) is highly challenging due to the overlapping of multiple map features and texts. Moreover, state-of-the-art Convolutional Neural Networks (CNN) are perfectly designed for content image filtering but provide no guarantee about closed shape detection. Also, the lack of textural and color information of historical maps makes it hard for CNN to detect shapes that are represented by only their boundaries. Our contribution is a pipeline that combines the strengths of CNN (efficient edge detection and filtering) and MM (guaranteed extraction of closed shapes) in order to achieve such a task. The evaluation of our approach on a public dataset shows its effectiveness for extracting the closed boundaries of objects in historical maps. Numéro de notice : P2021-001 Affiliation des auteurs : LaSTIG+Ext (2020- ) Autre URL associée : vers HAL Thématique : GEOMATIQUE Nature : Preprint nature-HAL : Préprint En ligne : https://arxiv.org/abs/2101.02144 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96739

Assessing historical maps for characterizing fluvial corridor changes at a regional network scale / Samuel Dunesme in Cartographica, vol 55 n° 4 (Winter 2020)  

Public [article]  inCartographica > vol 55 n° 4 (Winter 2020) . - pp 251 - 265 Titre : Assessing historical maps for characterizing fluvial corridor changes at a regional network scale Type de document : Article/Communication Auteurs : Samuel Dunesme , Auteur ; Hervé Piegay, Auteur ; Sébastien Mustière , Auteur Année de publication : 2020 Projets : EUR H20'Lyon / Article en page(s) : pp 251 - 265 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Bases de données localisées [Termes descripteurs IGN] analyse diachronique [Termes descripteurs IGN] base de données historiques [Termes descripteurs IGN] base de données topographiques [Termes descripteurs IGN] carte de base [Termes descripteurs IGN] corridor biologique [Termes descripteurs IGN] données hydrographiques [Termes descripteurs IGN] géomorphologie [Termes descripteurs IGN] rivière [Termes descripteurs IGN] trame bleue [Termes descripteurs IGN] vectorisation [Termes descripteurs IGN] vingtième siècle Résumé : (Auteur) Fluvial corridor quality assessment requires that historical data be collected at a regional scale. In this article, our goal is to assess potential map resources to explore riverscape changes at a regional network scale and to define key issues in using an automated vectorization protocol to characterize such changes on such a large scale. We consider IGN’s Nouvelle Carte de France a potentially good resource for our objective of two-date (oldest + actual vector database) comparisons on 1:20,000–1:25,000 scale maps, notably when applied at a regional scale. The French IGN corpus is a good example of topographic maps that were produced in the twentieth century in Europe with fairly homogeneous data over a whole national territory. Moreover, the digitization and georeferencing processes applied by IGN are very accurate. The evolution of conventional features is not as significant for the hydrographic theme and should not be a problem for automatic vectorization. The potential temporal coverage is from 1922 to 1993, but the complexity of the sheet divisions, partial revisions, and the heterogeneity of coverage over time prevent multidate analysis. Numéro de notice : A2020-775 Affiliation des auteurs : LaSTIG+Ext (2020- ) Autre URL associée : vers HAL Thématique : GEOMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.3138/cart-2019-0025 date de publication en ligne : 22/12/2020 En ligne : https://doi.org/10.3138/cart-2019-0025 Format de la ressource électronique : URL Article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96689 [article]

Can we characterize river corridor evolution at a continental scale from historical topographic maps? A first assessment from the comparison of four countries / J. Horacio Garcia in River Research and Applications, vol 36 n° 6 (July 2020)  

Public [article]  inRiver Research and Applications > vol 36 n° 6 (July 2020) . - pp 934 - 946 Titre : Can we characterize river corridor evolution at a continental scale from historical topographic maps? A first assessment from the comparison of four countries Type de document : Article/Communication Auteurs : J. Horacio Garcia, Auteur ; Samuel Dunesme , Auteur ; Hervé Piegay, Auteur Année de publication : 2020 Projets : EUR H20'Lyon / Article en page(s) : pp 934 - 946 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géomatique [Termes descripteurs IGN] Belgique [Termes descripteurs IGN] carte ancienne [Termes descripteurs IGN] corridor biologique [Termes descripteurs IGN] dix-neuvième siècle [Termes descripteurs IGN] forêt ripicole [Termes descripteurs IGN] géomorphologie locale [Termes descripteurs IGN] rivière [Termes descripteurs IGN] Suisse [Termes descripteurs IGN] vectorisation Résumé : (auteur) National historical map resources are assessed in four European countries to characterize river corridor features and associated channel changes, as well as identify issues limiting or promoting geomorphic assessment procedures at a continental scale. A geomorphic audit that launches potential data for diagnosis from reach to continental scales could offer a good resource for biology and ecology managers of river authorities or government agencies and engineers. The assessment compares the resources available by country in terms of period covered, spatial scale, history and chronology, and representation of the fluvial corridor features. We then applied the Historical Maps Vectorization Toolbox, initially developed for vectorizing river corridors from French maps, to detect and extract flow channels, unvegetated bars and riparian vegetation patches from historical topographical maps. We found that (a) it is difficult to apply an audit of channel changes to the whole continental scale because map legends differ between countries due to geographic and political specificity; (b) there exists an opportunity to get assessment information in all countries at reach or national scale where map resources are available; (c) the highest potential is observed in Switzerland and Belgium where there is high quality national map coverage from the 19th century; and (d) the algorithm Historical Maps Vectorization Toolbox applied to map resources works well with any of the countries, and its widespread application is encouraging. Numéro de notice : A2020-362 Affiliation des auteurs : ENSG+Ext (2012-2019) Thématique : FORET/GEOMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1002/rra.3582 date de publication en ligne : 30/12/2019 En ligne : https://doi.org/10.1002/rra.3582 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=95333 [article]

A discriminative tensor representation model for feature extraction and classification of multispectral LiDAR data / Qingwang Wang in IEEE Transactions on geoscience and remote sensing, vol 58 n° 3 (March 2020)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 58 n° 3 (March 2020) . - pp 1568 -1586 Titre : A discriminative tensor representation model for feature extraction and classification of multispectral LiDAR data Type de document : Article/Communication Auteurs : Qingwang Wang, Auteur ; Yanfeng Gu, Auteur Année de publication : 2020 Article en page(s) : pp 1568 -1586 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes descripteurs IGN] Amérique du nord [Termes descripteurs IGN] analyse discriminante [Termes descripteurs IGN] calcul tensoriel [Termes descripteurs IGN] carte d'occupation du sol [Termes descripteurs IGN] classification multibande [Termes descripteurs IGN] classification par séparateurs à vaste marge [Termes descripteurs IGN] données lidar [Termes descripteurs IGN] données localisées 3D [Termes descripteurs IGN] état de l'art [Termes descripteurs IGN] extraction de traits caractéristiques [Termes descripteurs IGN] image multibande [Termes descripteurs IGN] modèle géométrique [Termes descripteurs IGN] semis de points [Termes descripteurs IGN] tenseur [Termes descripteurs IGN] vectorisation [Termes descripteurs IGN] voisinage (topologie) Résumé : (Auteur) Multispectral light detection and ranging (MS-LiDAR) systems open the door to the possibility in the 3-D land cover classification at a finer scale using only point cloud data. This article proposes a model based on the tensor representation for multispectral point cloud classification. The proposed method combines the 3-D local spatial structure of each multispectral point by characterizing the point with a second-order tensor. The first mode of the tensor indicates the spatial location and spectral information of each point (i.e., the row of the second-order tensor) and the second mode denotes the neighborhood geometric and spectral structures (i.e., the column of the second-order tensor). Then we develop a novel tensor manifold discriminant embedding (TMDE) algorithm to extract the geometric–spectral features for multispectral point clouds classification. TMDE solves the mapping matrices of each mode by preserving the intraclass samples’ distribution further making it more compact and maximizing the distance of different classes. Finally, the support vector machine classifier with the extracted features as input is used to implement the classification of multispectral point clouds. Experiments are conducted on two real multispectral point cloud data sets. The experimental results demonstrate that the proposed method can achieve significant improvements in classification accuracies in comparison with several state-of-the-art algorithms. Numéro de notice : A2020-086 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2019.2947081 date de publication en ligne : 30/10/2019 En ligne : https://doi.org/10.1109/TGRS.2019.2947081 Format de la ressource électronique : URL Article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=94660 [article]

Learning and geometric approaches for automatic extraction of objects from remote sensing images / Nicolas Girard (2020)  

Public Titre : Learning and geometric approaches for automatic extraction of objects from remote sensing images Type de document : Thèse/HDR Auteurs : Nicolas Girard, Auteur Editeur : Nice : Université Côte d'Azur Année de publication : 2020 Importance : 169 p. Format : 21 x 30 cm Note générale : bibliographie Thèse de Doctorat Présentée en vue de l’obtention du grade de docteur en Automatique, Traitement du Signal et des Images de l'Université Côte d’Azur Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes descripteurs IGN] alignement [Termes descripteurs IGN] appariement de données localisées [Termes descripteurs IGN] apprentissage profond [Termes descripteurs IGN] automatisation des processus [Termes descripteurs IGN] classification par réseau neuronal convolutif [Termes descripteurs IGN] détection du bâti [Termes descripteurs IGN] erreur [Termes descripteurs IGN] figure géométrique [Termes descripteurs IGN] filtrage du bruit [Termes descripteurs IGN] jeu de données [Termes descripteurs IGN] polygonation [Termes descripteurs IGN] réalité de terrain [Termes descripteurs IGN] segmentation d'image [Termes descripteurs IGN] télédétection [Termes descripteurs IGN] vectorisation Résumé : (auteur) Creating a digital double of the Earth in the form of a map has many applications in e.g. autonomous driving, automated drone delivery, urban planning, telecommunications, and disaster management. Geographic Information Systems (GIS) are the frameworks used to integrate geolocalized data and represent maps. They represent shapes of objects in a vector representation so that it is as sparse as possible while representing shapes accurately, as well as making it easier to edit than raster data. With the increasing amount of satellite and aerial images being captured every day, automatic methods are being developed to transfer the information found in those remote sensing images into Geographic Information Systems. Deep learning methods for image segmentation are able to delineate the shapes of objects found in images, but they do so with a raster representation, in the form of a mask. Post-processing vectorization methods then convert that raster representation into a vector representation compatible with GIS. Another challenge in remote sensing is to deal with a certain type of noise in the data, which is the misalignment between different layers of geolocalized information (e.g. between images and building cadaster data). This type of noise is frequent due to various errors introduced during the processing of remote sensing data. This thesis develops combined learning and geometric approaches with the purpose to improve automatic GIS mapping from remote sensing images. We first propose a method for correcting misaligned maps over images, with the first motivation for them to match, but also with the motivation to create remote sensing datasets for image segmentation with alignment-corrected ground truth. Indeed training a model on misaligned ground truth would not lead to a nice segmentation, whereas aligned ground truth annotations will result in better segmentation models. During this work we also observed a denoising effect of our alignment model and use it to denoise a misaligned dataset in a self-supervised manner, meaning only the misaligned dataset was used for training. We then propose a simple approach to use a neural network to directly output shape information in the vector representation, in order to by-pass the post-processing vectorization step. Experimental results on a dataset of solar panels show that the proposed network succeeds in learning to regress polygon coordinates, yielding directly vectorial map outputs. Our simple method is limited to predicting polygons with a fixed number of vertices though. While more recent methods for learning directly in the vector representation are not limited to a fixed number of vertices, they still have other limitations in terms of the type of object shapes they can predict. More complex topological cases such as objects with holes or buildings touching each other (with a common wall which is very typical of European city centers) are not handled by these fully deep learning methods. We thus propose a hybrid approach alleviating those limitations by training a neural network to output a segmentation probability map as usual and also to output a frame field aligned with the contours of detected objects (buildings in our case). The frame field constitutes additional shape information learned by the network. We then propose our highly parallelizable polygonization method for leveraging that frame field information to vectorize the segmentation probability map efficiently. Because our polygonization method has access to additional information in the form of a frame field, it can be less complex than other advanced vectorization methods and is thus faster. Lastly, requiring an image segmentation network to also output a frame field only adds two convolutional layers and virtually does not increase inference time, making the use of a frame field only beneficial. Note de contenu : 1- Introduction 2- Building alignment 3- Building alignment from noisy ground truth 4- PolyCNN: learning polygons 5- Frame field learning 6- Polygonization by frame field 7- Conclusions and perspectives Numéro de notice : 28501 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Thèse française Note de thèse : Thèse de Doctorat : Traitement du Signal et des Images : Université Côte d’Azur : 2020 Organisme de stage : Inria Sophia-Antipolis DOI : sans En ligne : https://hal.inria.fr/tel-03111628/document Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96940

Very high resolution land cover mapping of urban areas at global scale with convolutional neural network / Thomas Tilak (2020)  

Permalink

Une nouvelle méthode de vectorisation du cadastre ancien / Antony Chalais in Géomatique expert, n° 129 (août - septembre 2019)

Permalink

An image-pyramid-based raster-to-vector conversion (IPBRTVC) framework for consecutive-scale cartography and synchronized generalization of classic objects / Chang Li in Photogrammetric Engineering & Remote Sensing, PERS, vol 85 n° 3 (March 2019)  

Permalink

Lever-nivelé de la place de Barcelone, 1823-1827 / Blanca Baella in Cartes & Géomatique, n° 238 (Décembre 2018)

Permalink

Une méthodologie nationale pour le géoréférencement et la vectorisation des cartes d’état-major, minutes au 1/40 000 / Thierry Lallemant in Revue forestière française [en ligne], vol 69 n° 4-5 (2017)  

Permalink

Towards a protocol for the collection of VGI vector data / Peter Mooney in ISPRS International journal of geo-information, vol 5 n° 11 (November 2016)    

Permalink

Multi-criteria, graph-based road centerline vectorization using ordered weighted averaging operators / Fateme Ameri in Photogrammetric Engineering & Remote Sensing, PERS, vol 82 n° 2 (February 2016)

Permalink

Use of GIS-supported comparative cartography and historical maps in long-term forest cover changes analysis in the Holy cross mountains (Poland) / Tadeusz Ciupa in Baltic forestry, vol 22 n° 1 ([01/02/2016])  

Permalink

Est-il possible de caractériser l’état du réseau fluvial à une échelle régionale à partir de cartes topographiques ? / Samuel Dunesme (2016) 

Permalink

Road vectorisation from high-resolution imagery based on dynamic clustering using particle swarm optimisation / Fateme Ameri in Photogrammetric record, vol 30 n° 152 (December 2015 - February 2016)  

Permalink