Centre de documentation
scientifique

Contextual classification of point cloud data by exploiting individual 3d neigbourhoods / Martin Weinmann in ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol II-3 W4 (March 2015)    

Public [article]  inISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences > vol II-3 W4 (March 2015) . - pp 271 - 278 Titre : Contextual classification of point cloud data by exploiting individual 3d neigbourhoods Type de document : Article/Communication Auteurs : Martin Weinmann, Auteur ; A. Schmidt, Auteur ; Clément Mallet , Auteur ; Stefan Hinz, Auteur ; Franz Rottensteiner, Auteur ; Boris Jutzi, Auteur Année de publication : 2015 Conférence : ISPRS 2015, PIA 2015 - HRIGI 2015 Joint ISPRS conference 25/03/2015 27/03/2015 Munich Allemagne ISPRS OA Annals Article en page(s) : pp 271 - 278 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] analyse d'image orientée objet [Termes IGN] classification contextuelle [Termes IGN] données lidar [Termes IGN] données localisées 3D [Termes IGN] scène urbaine [Termes IGN] semis de points [Termes IGN] voisinage (relation topologique) [Termes IGN] zone urbaine Résumé : (auteur) The fully automated analysis of 3D point clouds is of great importance in photogrammetry, remote sensing and computer vision. For reliably extracting objects such as buildings, road inventory or vegetation, many approaches rely on the results of a point cloud classification, where each 3D point is assigned a respective semantic class label. Such an assignment, in turn, typically involves statistical methods for feature extraction and machine learning. Whereas the different components in the processing workflow have extensively, but separately been investigated in recent years, the respective connection by sharing the results of crucial tasks across all components has not yet been addressed. This connection not only encapsulates the interrelated issues of neighborhood selection and feature extraction, but also the issue of how to involve spatial context in the classification step. In this paper, we present a novel and generic approach for 3D scene analysis which relies on (i) individually optimized 3D neighborhoods for (ii) the extraction of distinctive geometric features and (iii) the contextual classification of point cloud data. For a labeled benchmark dataset, we demonstrate the beneficial impact of involving contextual information in the classification process and that using individual 3D neighborhoods of optimal size significantly increases the quality of the results for both pointwise and contextual classification. Numéro de notice : A2015--052 Affiliation des auteurs : LASTIG MATIS+Ext (2012-2019) Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.5194/isprsannals-II-3-W4-271-2015 Date de publication en ligne : 11/03/2015 En ligne : http://dx.doi.org/10.5194/isprsannals-II-3-W4-271-2015 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=82698 [article]

Documents numériques

en open access Contextual classification of point cloud data ... - pdf éditeur Adobe Acrobat PDF

Tree topology representation from TLS point clouds using depth-first search in voxel space / A. Schilling in Photogrammetric Engineering & Remote Sensing, PERS, vol 78 n° 4 (April 2012)  

Public [article]  inPhotogrammetric Engineering & Remote Sensing, PERS > vol 78 n° 4 (April 2012) . - pp 383 - 392 Titre : Tree topology representation from TLS point clouds using depth-first search in voxel space Type de document : Article/Communication Auteurs : A. Schilling, Auteur ; A. Schmidt, Auteur ; Hans-Gerd Maas, Auteur Année de publication : 2012 Article en page(s) : pp 383 - 392 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] arbre (mathématique) [Termes IGN] données lidar [Termes IGN] données localisées 3D [Termes IGN] relation topologique [Termes IGN] semis de points [Termes IGN] transformation de Hough [Termes IGN] voxel Résumé : (Auteur) For a fundamental understanding of environmental processes and for the management of forests, information on the tree structure, preferably in 3D, is vital. Therefore, we propose a method to retrieve the spatial tree structure from 3D point clouds captured by a terrestrial laser scanner. The procedure addresses dense and noisy data sets of separate trees. Our method involves a variation of the Circular Hough Transform to determine trunk positions and a sequence of operations in voxel space. The core of the approach is the depth-first search algorithm, known from graph theory, to actually recover the tree as a graph. Furthermore, we compare results obtained from the tree graph to reference measurements of forest inventory parameters. The computation time of our method for topology representation is low and the method provides a reasonably accurate approximation of the 3D tree structure. Numéro de notice : A2012-181 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.14358/PERS.78.4.383 En ligne : https://doi.org/10.14358/PERS.78.4.383 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=31628 [article]