Building damage assessment in decimeter resolution SAR imagery: A future perspective

Damage assessment after natural disasters (e.g. earthquakes) is crucial for initiating effective post disaster relief actions. Synthetic aperture radar (SAR) sensors are an important source of information since they can map the extended areas quickly, in an uncensored manner, and independent from the weather conditions and the solar illumination. The spaceborne SAR sensors TerraSAR-X and COSMO-SkyMed reach spatial resolutions of about 1 meter and permit to analyze urban areas at the level of individual buildings. With this type of data completely destroyed buildings can be detected, while different types of damages can not be distinguished. In this paper we analize a set of decimeter resolution SAR data from an experimental airborne SAR system acquired from an artificial village of different types of destroyed buildings. We show that the increased resolution supports a more reliable identification of destroyed buildings, and allows the classification of destroyed buildings into several basic damage classes. Furthermore, we discuss some initial ideas for the development of novel automatic building damage assessment methods and give an outlook on how this type of data can be efficiently used in damage assessment scenarios.