Centre de documentation
scientifique

Exploring the potential of deep learning segmentation for mountain roads generalisation / Azelle Courtial in ISPRS International journal of geo-information, vol 9 n° 5 (May 2020)  

Public [article]  inISPRS International journal of geo-information > vol 9 n° 5 (May 2020) . - n° 338 ; 21 p. Titre : Exploring the potential of deep learning segmentation for mountain roads generalisation Type de document : Article/Communication Auteurs : Azelle Courtial , Auteur ; Achraf El Ayedi, Auteur ; Guillaume Touya , Auteur ; Xiang Zhang, Auteur Année de publication : 2020 Projets : 1-Pas de projet / Article en page(s) : n° 338 ; 21 p. Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Termes IGN] 1:25.000 [Termes IGN] 1:250.000 [Termes IGN] Alpes (France) [Termes IGN] apprentissage profond [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] données routières [Termes IGN] données vectorielles [Termes IGN] généralisation automatique de données [Termes IGN] montagne [Termes IGN] route [Termes IGN] segmentation [Termes IGN] symbole graphique [Termes IGN] virage [Vedettes matières IGN] Généralisation Résumé : (auteur) Among cartographic generalisation problems, the generalisation of sinuous bends in mountain roads has always been a popular one due to its difficulty. Recent research showed the potential of deep learning techniques to overcome some remaining research problems regarding the automation of cartographic generalisation. This paper explores this potential on the popular mountain road generalisation problem, which requires smoothing the road, enlarging the bend summits, and schematising the bend series by removing some of the bends. We modelled the mountain road generalisation as a deep learning problem by generating an image from input vector road data, and tried to generate it as an output of the model a new image of the generalised roads. Similarly to previous studies on building generalisation, we used a U-Net architecture to generate the generalised image from the ungeneralised image. The deep learning model was trained and evaluated on a dataset composed of roads in the Alps extracted from IGN (the French national mapping agency) maps at 1:250,000 (output) and 1:25,000 (input) scale. The results are encouraging as the output image looks like a generalised version of the roads and the accuracy of pixel segmentation is around 65%. The model learns how to smooth the output roads, and that it needs to displace and enlarge symbols but does not always correctly achieve these operations. This article shows the ability of deep learning to understand and manage the geographic information for generalisation, but also highlights challenges to come. Numéro de notice : A2020-295 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Thématique : GEOMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.3390/ijgi9050338 Date de publication en ligne : 25/05/2020 En ligne : https://doi.org/10.3390/ijgi9050338 Format de la ressource électronique : url article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=95131 [article]

Constraint based evaluation of generalized images generated by deep learning / Azelle Courtial (2020)  

Public Titre : Constraint based evaluation of generalized images generated by deep learning Type de document : Article/Communication Auteurs : Azelle Courtial , Auteur ; Guillaume Touya , Auteur ; Xiang Zhang, Auteur Editeur : ICA Commission on Generalisation and Multiple Representation Année de publication : 2020 Projets : 1-Pas de projet / Conférence : ICA 2020, 23rd Workshop on Map Generalisation and Multiple Representation 05/11/2020 06/11/2020 Delft Pays-Bas Open Access Proceedings Importance : 3 p. Format : 21 x 30 cm Note générale : Bibliographie Langues : Français (fre) Descripteur : [Termes IGN] 1:25.000 [Termes IGN] 1:250.000 [Termes IGN] Alpes (France) [Termes IGN] apprentissage profond [Termes IGN] carte routière [Termes IGN] classification pixellaire [Termes IGN] données maillées [Termes IGN] généralisation automatique de données [Termes IGN] généralisation cartographique automatisée [Termes IGN] montagne [Termes IGN] précision cartographique [Termes IGN] programmation par contraintes [Termes IGN] réseau routier [Vedettes matières IGN] Généralisation Résumé : (Auteur) The use of deep learning techniques for map generalisation raises new problems regarding the evaluation of the results: (1) images are used as input/output instead of vector data; (2) the deep learning processes do not guarantee results that follow cartographic principles; (3) the deep learning models are black boxes that hide the causal mechanisms. Also, deep learning intern evaluation is mostly based on the realism of the images and the pixel classification accuracy, and none of these criteria is sufficient to evaluate a generalisation process. In this article, we propose an adaptation of the constraint-based evaluation to the images generated by deep learning. Six raster-based constraints are proposed for a mountain road generalisation use case. Numéro de notice : C2020-018 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Thématique : GEOMATIQUE Nature : Communication nature-HAL : ComAvecCL&ActesPubliésIntl DOI : sans Date de publication en ligne : 17/11/2020 En ligne : https://varioscale.bk.tudelft.nl/events/icagen2020/ICAgen2020/ICAgen2020_paper_2 [...] Format de la ressource électronique : URL Article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96380

Representing forested regions at small scales: automatic derivation from very large scale data / William A Mackaness in Cartographic journal (the), vol 45 n° 1 (February 2008)  

Public [article]  inCartographic journal (the) > vol 45 n° 1 (February 2008) . - pp 6 - 17 Titre : Representing forested regions at small scales: automatic derivation from very large scale data Type de document : Article/Communication Auteurs : William A Mackaness, Auteur ; S. Perikleous, Auteur ; Omair Chaudhry, Auteur Année de publication : 2008 Article en page(s) : pp 6 - 17 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Bases de données localisées [Termes IGN] 1:250.000 [Termes IGN] carte dérivée [Termes IGN] forêt tempérée [Termes IGN] généralisation automatique de données [Termes IGN] généralisation cartographique automatisée [Termes IGN] grande échelle [Termes IGN] Ordnance Survey (UK) [Termes IGN] petite échelle [Termes IGN] Royaume-Uni [Termes IGN] service web géographique Résumé : (Auteur) As with any class of feature, it is important to be able to view woodland or forest at multiple levels of detail. At the detailed level, a map can show clusters of trees, tree types, tracks and paths; at the small scale, say 1:250 000, we can discern broad patterns of forests and other land use, which can inform planners and act as input to land resource models. Rather than store such information in separate databases (requiring multiple points of maintenance), the vision is that the information has a single point of storage and maintenance, and that from this detailed level, various, more generalised forms can be automatically derived. This paper presents a methodology and algorithm for automatically deriving forest patches suitable for representation at 1:250 000 scale directly from a detailed dataset. In addition to evaluation of the output, the paper demonstrates how such algorithms can be shared and utilised via 'generalisation web services', arguing that the sharing of such algorithms can help accelerate developments in map generalisation, and increase the uptake of research solutions within commercial systems. Copyright British Cartographic Society Numéro de notice : A2008-130 Affiliation des auteurs : non IGN Thématique : FORET/GEOMATIQUE/INFORMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1179/000870408X276576 En ligne : https://doi.org/10.1179/000870408X276576 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=29125 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
030-08011	RAB	Revue	Centre de documentation	En réserve L003	Disponible

Texture analysis and classification of ERS SAR images for map updating of urban areas in The Netherlands / R. Dekker in IEEE Transactions on geoscience and remote sensing, vol 41 n° 9 (September 2003)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 41 n° 9 (September 2003) . - pp 1950 - 1958 Titre : Texture analysis and classification of ERS SAR images for map updating of urban areas in The Netherlands Type de document : Article/Communication Auteurs : R. Dekker, Auteur Année de publication : 2003 Article en page(s) : pp 1950 - 1958 Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image radar et applications [Termes IGN] 1:250.000 [Termes IGN] analyse texturale [Termes IGN] cartographie urbaine [Termes IGN] classification [Termes IGN] histogramme [Termes IGN] image ERS-SAR [Termes IGN] image radar [Termes IGN] milieu urbain [Termes IGN] mise à jour cartographique [Termes IGN] Pays-Bas [Termes IGN] Rotterdam (Pays-Bas) [Termes IGN] variogramme Résumé : (Auteur) In single-band and single-polarized synthetic aperture radar (SAR) image classification, texture holds useful information. In a study to assess the map-updating capabilities of such sensors in urban areas, some modern texture measures were investigated. Among them were histogram measures, wavelet energy, fractal dimension, lacunarity, and semivariograms. The latter were chosen as an alternative for the well-known gray-level cooccurrence family of features. The area that was studied using a European Remote Sensing Satellite 1(ERS1) SAR image was the conurbation around Rotterdam and The Hague in The Netherlands. The area can be characterized as a well-planned dispersed urban area with residential areas, industry, greenhouses, pasture, arable land, and some forest. The digital map to be updated was a 1: 250 000 Vector Map (VMapl). The study was done on the basis of non-parametric separability measures and classification techniques because most texture distributions were not normal. The conclusion is that texture improves the classification accuracy. The measures that performed best were mean intensity (actually no texture), variance, weighted-rank fill ratio, and semivariogram, but the accuracies vary for different classes. Despite the improvement, the overall classification accuracy indicates that the land-cover information content of ERS1 leaves something to be desired. Numéro de notice : A2003-250 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2003.814628 En ligne : https://doi.org/10.1109/TGRS.2003.814628 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=22545 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
065-03091	RAB	Revue	Centre de documentation	En réserve L003	Disponible

Machine learning techniques for determining parameters of cartographic generalisation algorithms / Lagrange (Enseigne de Vaisseau) (2000)  

Public Titre : Machine learning techniques for determining parameters of cartographic generalisation algorithms Type de document : Article/Communication Auteurs : Lagrange (Enseigne de Vaisseau), Auteur ; Landras (Enseigne de Vaisseau), Auteur ; Sébastien Mustière , Auteur Editeur : International Society for Photogrammetry and Remote Sensing ISPRS Année de publication : 2000 Collection : International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences, ISSN 1682-1750 num. 33-B4 Conférence : ISPRS 2000, 19th ISPRS Congress Technical Commission 4, Mapping and Geographic Information Systems 16/07/2000 23/07/2000 Amsterdam Pays-Bas OA Proceedings archives Importance : pp 718 - 725 Format : 21 x 30 cm Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Termes IGN] 1:250.000 [Termes IGN] apprentissage automatique [Termes IGN] BD Carto [Termes IGN] carte routière [Termes IGN] lissage de données [Vedettes matières IGN] Généralisation Résumé : (auteur) This paper reports on research performed in the field of automated map generalization. We address the issue of determining how to set parameters of transformation algorithms. Empirical and theoretical studies have shown that, even given fixed map scale and purpose, these parameter values vary from one object to the other according to different characteristics such as the shape, size or environment of the object. Because of the complexity of cartographic rules and generalisation algorithms, we address this problem with techniques developed in the field of Machine Learning from examples. Specifically, we automatically learn, with neural networks, how to determine an algorithm parameters according to a set of measures describing the object to be transformed. We present the main issues to be addressed to use neural networks. We show that our approach is useful and that its main limit stands in the lack of good measures to describe an object. As a case study, this paper presents results of learning how to set the strength of a smoothing algorithm on a line from the French BDCarto database to represent a road on a 1/250,000 scale road map. Numéro de notice : C2000-029 Affiliation des auteurs : COGIT+Ext (1988-2011) Thématique : GEOMATIQUE Nature : Communication nature-HAL : ComAvecCL&ActesPubliésIntl DOI : sans En ligne : https://www.isprs.org/proceedings/XXXIII/congress/part4/718_XXXIII-part4.pdf Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=103272

Comparison of (7.5 minute) and (1 degree) digital elevation models / D.L. Isaacson in Photogrammetric Engineering & Remote Sensing, PERS, vol 56 n° 11 (november 1990)  

Permalink

Visual versus digital analysis for vegetation mapping: some examples on central Spain / E. Chuvieco in Geocarto international, vol 5 n° 3 (September - November 1990)  

Permalink

Cartographie agro-écologique des llanos occidentaux du Venezuela à petite et moyenne échelle : paysages et évolutions 1979, 1982, 1987 (Landsat et Spot) / M. Pouyllau (25/10/1989)

Permalink

Evaluation des conditions d'utilisation des images Spot pour la révision des cartes topographiques au 1:100000 et 1:250000 / Pierre Planques in Bulletin d'information de l'Institut géographique national, n° 56 (décembre 1988)

Permalink

Toute la France sur vos genoux : les cartes routières cartographiées par l'IGN au 1:250000 / Anonyme in XYZ, n° 30 (mars - mai 1987)

Permalink

Traitements d'images à l'IGN / Jean-Claude Lummaux in Bulletin d'information de l'Institut géographique national, n° 44 (mars 1982)

Permalink

The quarter inch to one mile map of Great Britain / D. Griffith in Annuaire international de cartographie, n° 5 (1965)

Permalink

Cartes orohydrographiques et éditions sans teinte verte / IGN in Bulletin d'information de l'Institut géographique national, n° 1 (novembre 1964)

Permalink

Studie über ein amtliches Kartenwerk im Maßstab 1:200.000 oder 1:250.000 / H. Knorr (1956)

Permalink