0
  • DE
  • EN
  • FR
  • Base de données et galerie internationale d'ouvrages d'art et du génie civil

Publicité

Tunneling Appropriate Computational Models from Laser Scanning Data

 Tunneling Appropriate Computational Models from Laser Scanning Data
Auteur(s): ,
Présenté pendant IABSE Symposium: Engineering the Future, Vancouver, Canada, 21-23 September 2017, publié dans , pp. 175-182
DOI: 10.2749/vancouver.2017.0175
Prix: € 25,00 incl. TVA pour document PDF  
AJOUTER AU PANIER
Télécharger l'aperçu (fichier PDF) 0.22 MB

Tunneling projects often require computational models of existing structures. To this end, this paper demonstrates the viability of automatically, robustly reconstructing an individual building mod...
Lire plus

Détails bibliographiques

Auteur(s): (Urban Modelling Group, University College Dublin, Ireland)
(Urban Modelling Group, University College Dublin, Ireland)
Médium: papier de conférence
Langue(s): anglais
Conférence: IABSE Symposium: Engineering the Future, Vancouver, Canada, 21-23 September 2017
Publié dans:
Page(s): 175-182 Nombre total de pages (du PDF): 8
Page(s): 175-182
Nombre total de pages (du PDF): 8
Année: 2017
DOI: 10.2749/vancouver.2017.0175
Abstrait:

Tunneling projects often require computational models of existing structures. To this end, this paper demonstrates the viability of automatically, robustly reconstructing an individual building model from laser scanning data for further computational modeling without any manual intervention. The resulting model is appropriate for immediate importation into a commercial finite element method (FEM) program. The method combines a voxel-based technique with an angle criterion. Initially, the voxelization model is used to represent the façade model, while an angle criterion is implemented to determine boundaries of the façade and its openings (doors and windows). The algorithm overcomes common problems of occlusions or artefacts that arise during data acquisition. The resulting relative errors of overall dimensions and opening areas of geometric models were less 2% and 6%, respectively, which are generally within industry standards for this type of building modeling.