COMPLEMENTARITY OF SAR POLARIMETRY AND TOMOGRAPHY FOR BUILDING DETECTION AND CHARACTERIZATION
- Computer Vision and Remote Sensing Group, Technical University of Berlin, Berlin, Germany
Keywords: SAR, polarimetry, tomography, object detection, segmentation
Abstract. In this paper we propose to study the potential of jointly using polarimetric and tomographic SAR data to recognize and localize buildings in complex scenarios. We present extraction methods for both polarimetric and tomographic features. One the one hand, we propose to use the polarimetric bilateral filter that has proved to be a powerful tool to retrieve the polarimetric covariance matrix while reducing speckle and preserving edges. Thus, polarimetric decompositions can be used for physical interpretation. On the other hand, the TomoSNI and TomoSeed algorithms allow to respectively extract interest points and segment geometric primitives in 3D point clouds obtained with tomographic focusing methods. We show how the output of such algorithms could be combined in order to allow the extraction of buildings. We also analyze different issues related to complex scenarios that may impede a correct detection and discuss some possible solutions.