Extraction of bright & dark features of buildings from VHR SAR images & detecting them automatically & 2-D reconstruction of radar footprints
Keywords:
Building detection, building reconstruction, synthetic aperture radar (SAR), very high spatial resolution (VHR)Abstract
Spaceborne synthetic aperture radar (SAR) systems such as COSMO-SkyMed, TerraSAR-X, and TanDEM-X acquire very high spatial resolution (VHR) imagery, enabling crucial applications like post-disaster damage assessment in urban environments. To support reliable, consistent, and rapid information extraction from such complex SAR data, automated processing methods are essential. This study presents an approach for detecting bright and dark linear features in single VHR SAR images and performing automated two-dimensional reconstruction of building radar footprints. The proposed method first extracts a set of low-level image features, which are then aggregated into more structured primitives using a production system. A semantic interpretation framework is introduced, assigning each primitive a probability of belonging to a specific scattering class. This semantic meaning, defined through fuzzy membership grades, compensates for the limited number of detectable features in single-image datasets. It also guides the selection of the most reliable primitives and footprint hypotheses, enabling accurate and robust radar footprint reconstruction in complex urban scenes.
References
E. Simonton, H. Oriot, and R. Garello, “Rectangular building extraction from stereoscopic airborne radar images,” IEEE Transactions on Geoscience and Remote Sensing, vol. 43, no. 10, pp. 2386–2395, Oct. 2005.
F. Zhang, Y. Shao, and X. Zhang, “Building extraction using dual-aspect high-resolution SAR images,” in Proceedings of the 18th International Conference on Geoinformatics, 2010, pp. 1–5.
F. Xu and Y.-Q. Jin, “Automatic reconstruction of building objects from multi-aspect meter-resolution SAR images,” IEEE Transactions on Geoscience and Remote Sensing, vol. 45, no. 7, pp. 2336–2353, Jul. 2007.
Y. Wang, F. Tupin, C. Han, and J.-M. Nicolas, “Building detection from high resolution PolSAR data by combining region and edge information,” in Proceedings of the IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Boston, MA, USA, Jul. 7–11, 2008, vol. 4, pp. IV-153–IV-156.
Downloads
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
