Building L-shape footprint extraction from high resolution SAR image

Building extraction in SAR images is still a difficult task due to the geometric distortions, the electromagnetic wave interactions between adjacent targets, and the speckle noises. In this paper, a method for building L-shape footprint extraction from high resolution SAR images jointly using image segmentation and line detection is proposed. Experiments based on a TerraSAR-X image show that method proposed in this paper can accurately delineate the L-shape edges of buildings facing towards the SAR sensor.