SEMANTIC BUILDING FAÇADE SEGMENTATION FROM AIRBORNE OBLIQUE IMAGES
With the introduction of airborne oblique camera systems and the improvement of photogrammetric techniques, high-resolution 2D and 3D data can be acquired in urban areas. This high-resolution data allows us to perform detailed investigations on building roofs and façades which can contribute to LoD3...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Copernicus Publications
2018-05-01
|
Series: | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
Online Access: | https://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/IV-2/209/2018/isprs-annals-IV-2-209-2018.pdf |
id |
doaj-11fa51b044074558ba4be2d1952e2f50 |
---|---|
record_format |
Article |
spelling |
doaj-11fa51b044074558ba4be2d1952e2f502020-11-24T21:53:27ZengCopernicus PublicationsISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences2194-90422194-90502018-05-01IV-220921610.5194/isprs-annals-IV-2-209-2018SEMANTIC BUILDING FAÇADE SEGMENTATION FROM AIRBORNE OBLIQUE IMAGESY. Lin0F. Nex1M. Y. Yang2ITC Faculty of Geo-Information Science and Earth Observation, University of Twente, the NetherlandsITC Faculty of Geo-Information Science and Earth Observation, University of Twente, the NetherlandsITC Faculty of Geo-Information Science and Earth Observation, University of Twente, the NetherlandsWith the introduction of airborne oblique camera systems and the improvement of photogrammetric techniques, high-resolution 2D and 3D data can be acquired in urban areas. This high-resolution data allows us to perform detailed investigations on building roofs and façades which can contribute to LoD3 city modeling. Normally, façade segmentation is achieved from terrestrial views. In this paper, we address the problem from aerial views by using high resolution oblique aerial images as the data source in urban areas. In addition to traditional image features, such as RGB and SIFT, normal vector and planarity are also extracted from dense matching point clouds. Then, these 3D geometrical features are projected back to 2D space to assist façade interpretation. Random forest is trained and applied to label façade pixels. Fully connected conditional random field (CRF), capturing long-range spatial interactions, is used as a post-processing to refine our classification results. Its pairwise potential is defined by a linear combination of Gaussian kernels and the CRF model is efficiently solved by mean field approximation. Experiments show that 3D features can significantly improve classification results. Also, fully connected CRF performs well in correcting noisy pixels.https://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/IV-2/209/2018/isprs-annals-IV-2-209-2018.pdf |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Y. Lin F. Nex M. Y. Yang |
spellingShingle |
Y. Lin F. Nex M. Y. Yang SEMANTIC BUILDING FAÇADE SEGMENTATION FROM AIRBORNE OBLIQUE IMAGES ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
author_facet |
Y. Lin F. Nex M. Y. Yang |
author_sort |
Y. Lin |
title |
SEMANTIC BUILDING FAÇADE SEGMENTATION FROM AIRBORNE OBLIQUE IMAGES |
title_short |
SEMANTIC BUILDING FAÇADE SEGMENTATION FROM AIRBORNE OBLIQUE IMAGES |
title_full |
SEMANTIC BUILDING FAÇADE SEGMENTATION FROM AIRBORNE OBLIQUE IMAGES |
title_fullStr |
SEMANTIC BUILDING FAÇADE SEGMENTATION FROM AIRBORNE OBLIQUE IMAGES |
title_full_unstemmed |
SEMANTIC BUILDING FAÇADE SEGMENTATION FROM AIRBORNE OBLIQUE IMAGES |
title_sort |
semantic building façade segmentation from airborne oblique images |
publisher |
Copernicus Publications |
series |
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
issn |
2194-9042 2194-9050 |
publishDate |
2018-05-01 |
description |
With the introduction of airborne oblique camera systems and the improvement of photogrammetric techniques, high-resolution 2D and 3D data can be acquired in urban areas. This high-resolution data allows us to perform detailed investigations on building roofs and façades which can contribute to LoD3 city modeling. Normally, façade segmentation is achieved from terrestrial views. In this paper, we address the problem from aerial views by using high resolution oblique aerial images as the data source in urban areas. In addition to traditional image features, such as RGB and SIFT, normal vector and planarity are also extracted from dense matching point clouds. Then, these 3D geometrical features are projected back to 2D space to assist façade interpretation. Random forest is trained and applied to label façade pixels. Fully connected conditional random field (CRF), capturing long-range spatial interactions, is used as a post-processing to refine our classification results. Its pairwise potential is defined by a linear combination of Gaussian kernels and the CRF model is efficiently solved by mean field approximation. Experiments show that 3D features can significantly improve classification results. Also, fully connected CRF performs well in correcting noisy pixels. |
url |
https://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/IV-2/209/2018/isprs-annals-IV-2-209-2018.pdf |
work_keys_str_mv |
AT ylin semanticbuildingfacadesegmentationfromairborneobliqueimages AT fnex semanticbuildingfacadesegmentationfromairborneobliqueimages AT myyang semanticbuildingfacadesegmentationfromairborneobliqueimages |
_version_ |
1725872203739693056 |