Detection of buildings in spaceborne TomoSAR point clouds via hybrid region growing and energy minimization technique

Abstract

In this paper, an automatic approach is presented to detect/extract buildings from spaceborne TomoSAR point clouds. The approach is systematic and allows robust detection of both tall and low height buildings and is, therefore, well suited for urban monitoring of larger areas from space. The presented approach is illustrated and validated by examples using TomoSAR point clouds generated from a stack of TerraSAR-X high resolution spotlight images covering an area of approximately 1.5 km2 containing mostly moderate sized buildings in the city of Berlin, Germany. The depicted results validate the effectiveness of the proposed approach.