CENTRE TECNOLÒGIC DE TELECOMUNICACI (ES)
The WISE (Wide-Area Sentinel-1 Deformation Classification for Advanced Data Exploitation) project focuses on developing advanced tools to exploit the large amount of satellite-borne Synthetic Aperture Radar (SAR) data. The main data source will be the Sentinel-1 SAR data: this Copernicus mission provides, with its fast revisit time, high-resolution and multi-polarization, an unprecedented amount of data, which need to be fully exploited.The thematic area covered by this project is the study of the ground displacements induced by different phenomena, affecting the natural and the built environment. The deformation will be derived using Persistent Scatterer Interferometry (PSI). The project shall produce maps of labelled deformation areas, geo-located and categorized by their type. This is key to perform a systematic and comprehensive exploitation of deformation datasets over wide areas. The project will exploit the PSI data containing the displacement time series of radar scatterers distributed all over Europe. A relevant data source that will be processed in this project is the re cently-launched European Ground Motion Service (EGMS) as part of the Copernicus Land Monitoring Services, based on Sentinel-1 PSI data. The European Environment Agency, deputing entity for the Service management, is intended as a fundamental stakeholder. The data analysis proposed in WISE unfolds following a rigorous processing chain. First, active deformation areas (ADAs), such as landslides, subsidence, infrastructure stability, will be detected employing a statistical hypothesis test. This operation allows preliminary data selection and removal of noisy pixels, yielding clusters of pixels whose deformation time series are correlated in both space and time. Second, the project shall develop automatic classification methods for the ADAs detected. A first method will identify the features necessary to uniquely define a deformation class, then traditional classifiers will yield the first set of labelled deformation areas. Machine Learning techniques will be adopted to derive more advanced classifiers. An accurate deformation classification technique cannot leave aside the temporal information associated to the persistent scatterers, hence classifiers using one or more memory layer will be employed to underpin this information. The novelty of such methods will be manifold, as they have been little utilised in the Earth Observation domain, and certainly not over an area as wide as Europe. The developed algorithms will be tested on SAR data collected over smaller areas of interest by different sensors, such as COSMO-SkyMed and TerraSAR-X, enabling the evaluation of frequency-diverse data. The labelled ADAs will be associated with metrics quantifying the effective likelihood of the classification performed, supporting potential users of the producedmethods and datasets. The produced maps will be validated using inventories and in-situ data of known deformation phenomena, by selecting representative test sites.The main output will be a database containing labelled deformation areas: a useful product for several end users of ground motion data, e.g. Civil Protection actors and risk management entities. Such a product will support the exploitation of theproducts of EGMS, Sentinel-1 and other SAR mission. The developed software will be freely available with different versions depending on the users’ experience in processing EO data.