Building unit density detection from high resolution TerraSAR-X image based on mathematical morphological operators

A method for detecting building unit density information from high resolution SAR images is proposed. Like most research works in urban analysis on high resolution SAR images, it detects buildings by extracting the bright layover and double bounce regions. But there are several differences between this work and other works. First, it pays more attention on the accuracy of building unit number than the accuracy of the building area, position, height or other information. Second, it tries to detect building information from a single intensity SAR image, not from multi-aspect SAR, InSAR, PolSAR or multi-resource remote sensing data. At last, it does the building detection with a novel combination of thresholding and morphological operators. The performance of this method is tested on a high resolution TerraSAR-X image covering a part of the city of Wuhan, China.