Advanced methods using stacks of multi-pass interferometric SAR data, specifically Persistent Scatterers Interferometry (PSI), have a major application in the monitoring of ground displacements. Persistent Scatterers (PS) are typically detected using statistical tests, built to control the probability of false detection. Such a strategy can however lead to information losses for the community involved in the interpretation of the measurements in the risk monitoring context. In this work we discuss an innovative detection strategy based on the use of Deep Learning. We present first results carried out on data acquired by SAR sensors of the COSMO-SkyMed constellation.
A Deep Learning Based Solution for Persistent Scatterers Detection
Tang, Weili;Vitale, Sergio;Pascazio, Vito;
2025-01-01
Abstract
Advanced methods using stacks of multi-pass interferometric SAR data, specifically Persistent Scatterers Interferometry (PSI), have a major application in the monitoring of ground displacements. Persistent Scatterers (PS) are typically detected using statistical tests, built to control the probability of false detection. Such a strategy can however lead to information losses for the community involved in the interpretation of the measurements in the risk monitoring context. In this work we discuss an innovative detection strategy based on the use of Deep Learning. We present first results carried out on data acquired by SAR sensors of the COSMO-SkyMed constellation.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
