Urban building density detection using high resolution SAR imagery

Urban building density has always being an important indicator for city planning, land management and resident density evaluation. Recently, very high resolution (VHR) Synthetic Aperture Radar (SAR) sensors, delivering images with metric or sub-metric resolution, make it possible to extract urban fabric (like urban building density) with SAR imagery. In this article, we propose a highly efficient and more accurate method for computing pixel-level urban building density from single power SAR data based on a “building detection - building density estimation - amendment” procedure. The performance of the proposed method is evaluated on a TerraSAR-X image.