3D building reconstruction based on given ground plan information and surface models extracted from spaceborne imagery


Creative Commons License

Tack F., Buyuksalih G., Goossens R.

ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING, cilt.67, ss.52-64, 2012 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 67
  • Basım Tarihi: 2012
  • Doi Numarası: 10.1016/j.isprsjprs.2011.10.003
  • Dergi Adı: ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.52-64
  • Anahtar Kelimeler: 3D city modeling, DSM, Photogrammetry, High resolution satellite imagery, 2D ground plans, DSM GENERATION, MULTIPLE, ALGORITHMS
  • İstanbul Üniversitesi Adresli: Hayır

Özet

3D surface models have gained field as an important tool for urban planning and mapping. However, urban environments have a complex nature to model and they provide a challenge to investigate the current limits of automatic digital surface modeling from high resolution satellite imagery. An approach is introduced to improve a 3D surface model, extracted photogrammetrically from satellite imagery, based on the geometric building information embodied in existing 2D ground plans. First buildings are clipped from the extracted DSM based on the 20 polygonal building ground plans. To generate prismatic shaped structures with vertical walls and flat roofs, building shape is retrieved from the cadastre database while elevation information is extracted from the DSM. Within each 2D building boundary, a constant roof height is extracted based on statistical calculations of the height values. After buildings are extracted from the initial surface model, the remaining DSM is further processed to simplify to a smooth DTM that reflects bare ground, without artifacts, local relief, vegetation, cars and city furniture. In a next phase, both models are merged to yield an integrated city model or generalized DSM. The accuracy of the generalized surface model is assessed according to a quantitative-statistical analysis by comparison with two different types of reference data. (C) 2011 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS). Published by Elsevier B.V. All rights reserved.