GAP-CLOSING 3D BUILDING RECONSTRUCTION BY ALIGNING BOUNDARIES OF ROOF SEGMENTS AND DETECTING UNCOVERED DETAILS
We describe a work flow to border building faces which aims to obtain a detailed and closed building model. Initially, we use the estimated roof planes and the rasterized binary mask of the corresponding inlier set to generate bordering polygons. To close the gaps between the initial boundary polygo...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
Copernicus Publications
2015-03-01
|
Series: | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
Online Access: | http://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/II-3-W4/173/2015/isprsannals-II-3-W4-173-2015.pdf |
id |
doaj-54a7cdb76e71475ba733781d4e5fa0a0 |
---|---|
record_format |
Article |
spelling |
doaj-54a7cdb76e71475ba733781d4e5fa0a02020-11-24T20:49:15ZengCopernicus PublicationsISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences2194-90422194-90502015-03-01II-3/W417318010.5194/isprsannals-II-3-W4-173-2015GAP-CLOSING 3D BUILDING RECONSTRUCTION BY ALIGNING BOUNDARIES OF ROOF SEGMENTS AND DETECTING UNCOVERED DETAILSM. Pohl0D. Bulatov1Fraunhofer Institute of Optronics, System Technologies and Image Exploitation, Gutleuthausstr. 1, 76275 Ettlingen, GermanyFraunhofer Institute of Optronics, System Technologies and Image Exploitation, Gutleuthausstr. 1, 76275 Ettlingen, GermanyWe describe a work flow to border building faces which aims to obtain a detailed and closed building model. Initially, we use the estimated roof planes and the rasterized binary mask of the corresponding inlier set to generate bordering polygons. To close the gaps between the initial boundary polygons and between the polygons and the building ground outline, we introduce an algorithm to align boundaries which successfully works in 2.5D and 3D. To enhance the accuracy of the boundary alignment, we use additional reliable model entities such as cut lines and step lines between the initial estimated roof planes. All gaps that cannot be avoided by this procedure are afterwards covered by a method searching for uncovered details.http://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/II-3-W4/173/2015/isprsannals-II-3-W4-173-2015.pdf |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
M. Pohl D. Bulatov |
spellingShingle |
M. Pohl D. Bulatov GAP-CLOSING 3D BUILDING RECONSTRUCTION BY ALIGNING BOUNDARIES OF ROOF SEGMENTS AND DETECTING UNCOVERED DETAILS ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
author_facet |
M. Pohl D. Bulatov |
author_sort |
M. Pohl |
title |
GAP-CLOSING 3D BUILDING RECONSTRUCTION BY ALIGNING BOUNDARIES OF ROOF SEGMENTS AND DETECTING UNCOVERED DETAILS |
title_short |
GAP-CLOSING 3D BUILDING RECONSTRUCTION BY ALIGNING BOUNDARIES OF ROOF SEGMENTS AND DETECTING UNCOVERED DETAILS |
title_full |
GAP-CLOSING 3D BUILDING RECONSTRUCTION BY ALIGNING BOUNDARIES OF ROOF SEGMENTS AND DETECTING UNCOVERED DETAILS |
title_fullStr |
GAP-CLOSING 3D BUILDING RECONSTRUCTION BY ALIGNING BOUNDARIES OF ROOF SEGMENTS AND DETECTING UNCOVERED DETAILS |
title_full_unstemmed |
GAP-CLOSING 3D BUILDING RECONSTRUCTION BY ALIGNING BOUNDARIES OF ROOF SEGMENTS AND DETECTING UNCOVERED DETAILS |
title_sort |
gap-closing 3d building reconstruction by aligning boundaries of roof segments and detecting uncovered details |
publisher |
Copernicus Publications |
series |
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
issn |
2194-9042 2194-9050 |
publishDate |
2015-03-01 |
description |
We describe a work flow to border building faces which aims to obtain a detailed and closed building model. Initially, we use the
estimated roof planes and the rasterized binary mask of the corresponding inlier set to generate bordering polygons. To close the
gaps between the initial boundary polygons and between the polygons and the building ground outline, we introduce an algorithm to
align boundaries which successfully works in 2.5D and 3D. To enhance the accuracy of the boundary alignment, we use additional
reliable model entities such as cut lines and step lines between the initial estimated roof planes. All gaps that cannot be avoided by this
procedure are afterwards covered by a method searching for uncovered details. |
url |
http://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/II-3-W4/173/2015/isprsannals-II-3-W4-173-2015.pdf |
work_keys_str_mv |
AT mpohl gapclosing3dbuildingreconstructionbyaligningboundariesofroofsegmentsanddetectinguncovereddetails AT dbulatov gapclosing3dbuildingreconstructionbyaligningboundariesofroofsegmentsanddetectinguncovereddetails |
_version_ |
1716806276266065920 |