Abstract : SAR sensors deliver precious information about our earth: they allow the mapping of different land covers with amplitude or polarimetric data, Digital Terrain Model reconstruction thanks to stereo-vision, interferometry or tomography, and movement monitoring with differential SAR interferometry and 4D SAR tomography.
Since the beginning of satellite SAR imagery and especially the ESA ERS-1 mission, multi-temporal SAR series are available and widely exploited for all these applications. However, the recent launch of the new SAR generation of Sentinel program and associated data policy will provide a huge amount of information with a high temporal frequency. Although having multitemporal and multisensor data increases the quantity of information, many challenges arise that must be overcome in order to take advantage of its full potential, especially in urban areas. The task of processing and analyzing this huge, fastly arriving and heterogeneous amount of data can benefit from recent advances in image processing, machine learning and temporal modeling.