A Hybrid Approach for Building Extraction From Spaceborne Multi-Angular Optical Imagery
- 4 January 2012
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 5 (1), 89-100
- https://doi.org/10.1109/jstars.2011.2179792
Abstract
The advent of high resolution spaceborne images leads to the development of efficient detection of complex urban details with high precision. This urban land use study is focused on building extraction and height estimation from spaceborne optical imagery. The advantages of such methods include 3D visualization of urban areas, digital urban mapping, and GIS databases for decision makers. In particular, a hybrid approach is proposed for efficient building extraction from optical multi-angular imagery, where a template matching algorithm is formulated for automatic estimation of relative building height, and the relative height estimates are utilized in conjunction with a support vector machine (SVM)-based classifier for extraction of buildings from non-buildings. This approach is tested on ortho-rectified Level-2a multi-angular images of Rio de Janeiro from WorldView-2 sensor. Its performance is validated using a 3-fold cross validation strategy. The final results are presented as a building map and an approximate 3D model of buildings. The building detection accuracy of the proposed method is improved to 88%, compared to 83% without using multi-angular information.Keywords
This publication has 32 references indexed in Scilit:
- Fast block matching algorithms using frequency domainPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2011
- Extraction of building polygons from SAR images: Grouping and decision-level in the GESTALT systemPattern Recognition Letters, 2010
- Shape-Based Human Detection and Segmentation via Hierarchical Part-Template MatchingIEEE Transactions on Pattern Analysis and Machine Intelligence, 2010
- Building Height Retrieval From VHR SAR Imagery Based on an Iterative Simulation and Matching TechniqueIEEE Transactions on Geoscience and Remote Sensing, 2009
- Stereo analysis of high-resolution SAR images for building height estimation in cases of orthogonal aspect directionsISPRS Journal of Photogrammetry and Remote Sensing, 2009
- Morphology-based Building Detection from Airborne Lidar DataPhotogrammetric Engineering & Remote Sensing, 2009
- Automated Building Extraction from High-Resolution Satellite Imagery in Urban Areas Using Structural, Contextual, and Spectral InformationEURASIP Journal on Advances in Signal Processing, 2005
- Digital roof model (DRM)using high resolution satellite image and its application for 3D mapping of city regionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2004
- A comparison of methods for multiclass support vector machinesIEEE Transactions on Neural Networks, 2002
- Template matching using fast normalized cross correlationPublished by SPIE-Intl Soc Optical Eng ,2001