Descripteur
Documents disponibles dans cette catégorie (501)
Ajouter le résultat dans votre panier
Visionner les documents numériques
Affiner la recherche Interroger des sources externes
Etendre la recherche sur niveau(x) vers le bas
A stixel approach for enhancing semantic image segmentation using prior map information / Sylvain Jonchery (2018)
Titre : A stixel approach for enhancing semantic image segmentation using prior map information Type de document : Article/Communication Auteurs : Sylvain Jonchery, Auteur ; Guillaume Bresson, Auteur ; Bruno Vallet , Auteur ; Rafal Żbikowski, Auteur Editeur : New York : Institute of Electrical and Electronics Engineers IEEE Année de publication : 2018 Projets : 2-Pas d'info accessible - article non ouvert / Conférence : ICARCV 2018, 15th International Conference on Control, Automation, Robotics and Vision 10/11/2018 21/11/2018 Singapour Singapour Proceedings IEEE Importance : pp 1715 - 1720 Format : 21 x 30 cm Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] classification par réseau neuronal convolutif
[Termes IGN] précision de la classification
[Termes IGN] scène urbaine
[Termes IGN] segmentation sémantiqueRésumé : (auteur) A key problem for autonomous car navigation is the understanding, at an object level, of the current driving situation. Addressing this issue requires the extraction of meaningful information from on-board stereo imagery by classifying the fundamental elements of urban scenes into semantic categories that can more easily be interpreted and be reflected upon (streets, buildings, pedestrians, vehicles, signs, etc.). A probabilistic method is proposed to fuse a coarse prior 3D map data with stereo imagery classification. A novel fusion architecture based on the Stixel framework is presented for combining semantic pixel-wise segmentation from a convolutional neural network (CNN) with depth information obtained from stereo imagery while integrating coarse prior depth and label information. The proposed approach was tested on a manually labeled data set in urban environments. The results show that the classification accuracy of the fundamental elements composing the urban scene was significantly enhanced by this method compared to what is obtained from the semantic pixel-wise segmentation of a CNN alone. Numéro de notice : C2018-094 Affiliation des auteurs : LASTIG MATIS+Ext (2012-2019) Thématique : IMAGERIE Nature : Communication nature-HAL : ComAvecCL&ActesPubliésIntl DOI : 10.1109/ICARCV.2018.8581150 Date de publication en ligne : 20/12/2018 En ligne : https://doi.org/10.1109/ICARCV.2018.8581150 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=94247 SuperPoint Graph : segmentation sémantique de nuages de points LiDAR à grande échelle / Loïc Landrieu (2018)
contenu dans 27èmes Journées de la Recherche de l'IGN / Journées Recherche de l'IGN 2018, 27es Journées (22 - 23 mars 2018; Cité Descartes, Champs-sur-Marne, France) (2018)
Titre : SuperPoint Graph : segmentation sémantique de nuages de points LiDAR à grande échelle Type de document : Article/Communication Auteurs : Loïc Landrieu , Auteur ; Martin Simonovsky, Auteur Editeur : Saint-Mandé : Institut national de l'information géographique et forestière - IGN (2012-) Année de publication : 2018 Conférence : Journées Recherche de l'IGN 2018, 27es Journées 22/03/2018 23/03/2018 Champs-sur-Marne France programme sans actes Langues : Français (fre) Descripteur : [Vedettes matières IGN] Traitement d'image
[Termes IGN] données lidar
[Termes IGN] données localisées 3D
[Termes IGN] méthode de réduction d'énergie
[Termes IGN] réseau neuronal artificiel
[Termes IGN] segmentation sémantique
[Termes IGN] semis de pointsRésumé : (Auteur) SuperPoint Graph est un nouvel algorithme permettant la sémantisation précise de très grands volumes de nuages de points acquis par LiDAR. Il repose sur une partition du nuage en formes simples à l'aide d'un modèle d'énergie globale, qui permet de réduire considérablement la taille et la complexité des entrées. Une représentation profonde de chaque forme est obtenue grâce à un réseau de neurones spécialisé dans le traitement de petits nuages de points. Enfin, un réseau de réseaux de neurones récurrents spatialement structuré permet d'exploiter les relations contextuelles entre formes. La précision des résultats obtenus a permis à SuperPoint Graph de se hisser à la tête de plusieurs benchmarks internationaux. Numéro de notice : C2018-088 Affiliation des auteurs : LASTIG MATIS+Ext (2012-2019) Thématique : IMAGERIE Nature : Communication nature-HAL : ComSansActesPubliés-Unpublished DOI : sans Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=91540 Documents numériques
en open access
SuperPoint Graph - diaporama de présentationAdobe Acrobat PDF Open land cover from OpenStreetMap and remote sensing / Michael Schultz in International journal of applied Earth observation and geoinformation, vol 63 (December 2017)
[article]
Titre : Open land cover from OpenStreetMap and remote sensing Type de document : Article/Communication Auteurs : Michael Schultz, Auteur ; Janek Voss, Auteur ; Michael Auer, Auteur ; Sarah Carter, Auteur ; Alexander Zipf, Auteur Année de publication : 2017 Article en page(s) : pp 206 - 213 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Bases de données localisées
[Termes IGN] analyse diachronique
[Termes IGN] Bade-Wurtemberg (Allemagne)
[Termes IGN] classification par forêts d'arbres décisionnels
[Termes IGN] image Landsat-8
[Termes IGN] occupation du sol
[Termes IGN] OpenStreetMap
[Termes IGN] segmentation sémantiqueRésumé : (auteur) OpenStreetMap (OSM) tags were used to produce a global Open Land Cover (OLC) product with fractional data gaps available at osmlanduse.org. Data gaps in the global OLC map were filled for a case study in Heidelberg, Germany using free remote sensing data, which resulted in a land cover (LC) prototype with complete coverage in this area. Sixty tags in the OSM were used to allocate a Corine Land Cover (CLC) level 2 land use classification to 91.8% of the study area, and the remaining gaps were filled with remote sensing data. For this case study, complete are coverage OLC overall accuracy was estimated 87%, which performed better than the CLC product (81% overall accuracy) of 2012. Spatial thematic overlap for the two products was 84%. OLC was in large parts found to be more detailed than CLC, particularly when LC patterns were heterogeneous, and outperformed CLC in the classification of 12 of the 14 classes. Our OLC product represented data created in different periods; 53% of the area was 2011–2016, and 46% of the area was representative of 2016–2017. Numéro de notice : A2017-638 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.jag.2017.07.014 En ligne : https://doi.org/10.1016/j.jag.2017.07.014 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=86989
in International journal of applied Earth observation and geoinformation > vol 63 (December 2017) . - pp 206 - 213[article]Cut Pursuit: Fast algorithms to learn piecewise constant functions on general weighted graphs / Loïc Landrieu in SIAM Journal on Imaging Sciences, vol 10 n° 4 (November 2017)
[article]
Titre : Cut Pursuit: Fast algorithms to learn piecewise constant functions on general weighted graphs Type de document : Article/Communication Auteurs : Loïc Landrieu , Auteur ; Guillaume Obozinski, Auteur Année de publication : 2017 Projets : 2-Pas d'info accessible - article non ouvert / Article en page(s) : pp 1724 - 1766 Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image
[Termes IGN] algorithme Cut Pursuit
[Termes IGN] graphe
[Termes IGN] pondérationRésumé : (auteur) We propose working set/greedy algorithms to efficiently solve problems penalized, respectively, by the total variation on a general weighted graph and its $\ell_0$ counterpart the total level-set boundary size when the piecewise constant solutions have a small number of distinct level sets; this is typically the case when the total level-set boundary size is small, which is encouraged by these two forms of penalization. Our algorithms exploit this structure by recursively splitting the level sets of a piecewise constant candidate solution using graph cuts. We obtain significant speedups over state-of-the-art algorithms for images that are well approximated with few level sets. Numéro de notice : A2017-891 Affiliation des auteurs : LASTIG MATIS+Ext (2012-2019) Thématique : IMAGERIE/MATHEMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1137/17M1113436 Date de publication en ligne : 10/10/2017 En ligne : https://doi.org/10.1137/17M1113436 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=91884
in SIAM Journal on Imaging Sciences > vol 10 n° 4 (November 2017) . - pp 1724 - 1766[article]Tree size thresholds produce biased estimates of forest biomass dynamics / Eric B. Searle in Forest ecology and management, vol 400 (15 September 2017)
[article]
Titre : Tree size thresholds produce biased estimates of forest biomass dynamics Type de document : Article/Communication Auteurs : Eric B. Searle, Auteur ; Han Y.H. Chen, Auteur Année de publication : 2017 Article en page(s) : pp 468 - 474 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Termes IGN] biomasse aérienne
[Termes IGN] changement climatique
[Termes IGN] diamètre des arbres
[Termes IGN] échantillonnage
[Termes IGN] erreur systématique
[Termes IGN] estimation statistique
[Termes IGN] forêt boréale
[Termes IGN] inventaire forestier (techniques et méthodes)
[Termes IGN] inventaire forestier étranger (données)
[Termes IGN] Manitoba (Canada)
[Termes IGN] placette d'échantillonnage
[Termes IGN] seuillage
[Vedettes matières IGN] Inventaire forestierRésumé : (auteur) Studies that examine forest biomass dynamics often rely on long-term, spatially extensive, repeatedly measured permanent sample plots. Due to the intensive cost of sampling all trees within these plots, an arbitrary size threshold is typically imposed, which leads to only larger trees being sampled. However, it remains unclear whether the sampling of only large trees is representative of the entirety of stands of diverse sizes; the sampling of only large trees may produce biased estimates of biomass dynamics (growth, ingrowth, and mortality). Using a network of 141 permanent sample plots from Manitoba, Canada, with all trees of >1.3 m in height repeatedly measured, we constructed three distinct data sets, with 10 cm, 5 cm, and no diameter at breast height threshold, to illustrate that total productivity and mortality are increasingly underestimated with increasingly larger diameter at breast height thresholds. This effect is particularly significant in young stands, where productivity estimates peak at least 20 years earlier than the determined estimates under large thresholds. We highlight the need to account for smaller trees in long-term observational studies to ensure unbiased estimates of stand level aboveground biomass productivity and loss. Numéro de notice : A2017-807 Affiliation des auteurs : non IGN Thématique : FORET/MATHEMATIQUE Nature : Article DOI : 10.1016/j.foreco.2017.06.042 En ligne : https://doi.org/10.1016/j.foreco.2017.06.042 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=89245
in Forest ecology and management > vol 400 (15 September 2017) . - pp 468 - 474[article]Mapping theories of transformative learning / Daniel Casebeer in Cartographica, vol 52 n° 3 (Fall 2017)PermalinkJoint classification and contour extraction of large 3D point clouds / Timo Hackel in ISPRS Journal of photogrammetry and remote sensing, vol 130 (August 2017)PermalinkVertical stratification of forest canopy for segmentation of understory trees within small-footprint airborne LiDAR point clouds / Hamid Hamraz in ISPRS Journal of photogrammetry and remote sensing, vol 130 (August 2017)PermalinkA novel semisupervised active-learning algorithm for hyperspectral image classification / Zengmao Wang in IEEE Transactions on geoscience and remote sensing, vol 55 n° 6 (June 2017)PermalinkUrban 3D segmentation and modelling from street view images and LiDAR point clouds / Pouria Babahajiani in Machine Vision and Applications, sans n° ([01/06/2017])PermalinkGeometric features and their relevance for 3D point cloud classification / Martin Weinmann in ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol IV-1/W1 (May 2017)PermalinkThe differentiation of point symbols using selected visual variables in the mobile augmented reality system / Łukasz Halik in Cartographic journal (the), Vol 54 n° 2 (May 2017)PermalinkSemantic segmentation of forest stands of pure species combining airborne lidar data and very high resolution multispectral imagery / Clément Dechesne in ISPRS Journal of photogrammetry and remote sensing, vol 126 (April 2017)PermalinkUAS, sensors, and data processing in agroforestry: a review towards practical applications / Luis Padua in International Journal of Remote Sensing IJRS, vol 38 n° 8-10 (April 2017)PermalinkA classification-segmentation framework for the detection of individual trees in dense MMS point cloud data acquired in urban areas / Martin Weinmann in Remote sensing, vol 9 n° 3 (March 2017)Permalink