Simona Verde

Bridge Thermal Dilation Monitoring With Millimeter Sensitivity via Multidimensional SAR Imaging

The new generation of synthetic aperture radar (SAR) sensors is providing images with very high spatial resolution, improved up to the meter scale. Such a resolution increase allows more accurate monitoring capabilities by means of interferometric approaches. The use of higher frequency enhances the sensitivity of the system even to minute changes, such as thermal dilations. This phenomenon has an impact on the interferometric products, particularly on the deformation velocity maps, if not properly handled. Man-made structures, such as steel core bridges and specific buildings, may be very sensible to thermal dilation effects. By extending the multitemporal differential interferometry SAR processing chains, in our case based on the multidimensional imaging (MDI) approach, an additional parameter related to temperature differences at acquisition instants, the thermal coefficient, can be accurately estimated. This parameter provides interesting perspectives in application to infrastructure monitoring: It brings information about the thermal behavior of the imaged objects. In this letter, we investigate the thermal response of the Musmeci bridge (Potenza, Italy), by experimenting the extended MDI approach on a real TerraSAR-X data set. Results highlight the possibility of such a technique to obtain measurements of the motion that is highly correlated with temperature, thus providing useful information about the static structure of bridges.

CAESAR: An Approach Based on Covariance Matrix Decomposition to Improve Multibaseline–Multitemporal Interferometric SAR Processing

Synthetic aperture radar (SAR) tomography has been strongly developed in the last years for the analysis at fine scale of data acquired by high-resolution interferometric SAR sensors as a technique alternative to classical persistent scatterer interferometry and able to resolve also multiple scatterers. SqueeSAR is a recently proposed solution which, in the context of SAR interferometry at the coarse scale analysis stage, allows taking advantage of the multilook operation to filter interferometic stacks by extracting, pixel by pixel, equivalent scattering mechanisms from the set of all available interferometric measurement collected in the data covariance matrix. In this paper, we investigate the possibilities to extend SqueeSAR by allowing the identification of multiple scattering mechanisms from the analysis of the covariance matrix. In particular, we present a new approach, named “Component extrAction and sElection SAR” algorithm, that allows taking advantage of the principal component analysis to filter interferograms relevant to the decorrelating scatterer, i.e., scatterers that may exhibit coherence losses depending on the spatial and temporal baseline distributions, and to detect and separate scattering mechanisms possibly interfering in the same pixel due to layover directly at the interferogram generation stage. The proposed module allows providing options useful for classical interferometric processing to monitor ground deformations at lower resolution (coarse scale), as well as for possibly aiding the data calibration preliminary for the subsequent full-resolution interferometric/tomographic (fine scale) analysis. Results achieved by processing high-resolution Cosmo-SkyMed data, characterized by the favorable features of a large baseline span, are presented to explain the advantages and validate this new interferometric processing solution.

Multilook SAR Tomography for 3-D Reconstruction and Monitoring of Single Structures Applied to COSMO-SKYMED Data

With reference to the application to the imaging and monitoring of infrastructures and buildings in urban areas, SAR tomography has been mainly developed and tested at full resolution. In this work, we investigate the possibility related to the use of a multilook approach for fine resolution analysis of ground structures that combines SAR tomography and a method, CAESAR, recently proposed for classical DInSAR analysis at coarse resolution over large areas. Shown results, achieved by processing two 3 m spatial resolution (stripmap mode) COSMO-SKYMED datasets relative to the urban areas of Naples and Rome (Italy), clearly indicate that the proposed multilook-based method allows achieving an impressive density of detected scatterers over buildings and infrastructures, much higher than those achievable with standard full-resolution methods.

Detection and monitoring of facilities exposed to subsidence phenomena via past and current generation SAR sensors

The identification of facilities in areas affected by subsidence phenomena represents a fundamental activity in processes dealing with land management. For this kind of phenomena, the analyses may be hampered by the lack of official subsidence zoning maps because of the wide extension of the affected areas. This is mainly due to the costs necessary for measurements and surveys to be carried out via conventional in?situ techniques which can turn out to be unaffordable for the authorities in charge of land management. In this regard, during the last decade the use of remote sensing data, such as medium resolution synthetic aperture radar (SAR) images processed via differential interferometry algorithms (DInSAR), has proven its benefits for the detection and monitoring of facilities (i.e., buildings and infrastructures) in subsiding areas. Currently, the improved resolution and coverage of the ultimate generation SAR sensors seem very promising for consequence analyses of facilities, although displacement time series are still limited for long-term studies. In this paper, analyses of DInSAR data acquired via both medium (ERS-ENVISAT) and high (COSMO-SkyMed) resolution sensors are carried out over a densely urbanized flat area in southern Italy so as to show how the appropriate use of DInSAR data at different scales can valuably help in the detection and monitoring of damageable facilities.

Assimilation of GPS-Derived Atmospheric Propagation Delay in DInSAR Data Processing

Microwave radiation is almost insensitive in terms of power attenuation to the presence of atmosphere; the atmosphere is however an error source in repeat pass interferometry due to propagation delay variations. This effect represents a main limitation in the detection and monitoring of weak deformation patterns in differential interferometric Synthetic Aperture Radar (DInSAR), especially in emergency conditions. Due to the wavelength reduction current, X-Band sensors are even more sensitive to such error sources: procedures adopted in classical advanced DInSAR for atmospheric filtering may fail in the presence of higher revisiting rates. In this work, we show such effect on data acquired by the COSMO-SkyMed constellation. The dataset has been acquired with very high revisiting rates during the emergency phase. This feature allows clearly showing the inability of standard filtering adopted in common processing chains in handling seasonal atmospheric delay variations over temporal intervals spanning periods shorter than 1 year. We discuss a procedure for the mitigation of atmospheric propagation delay (APD) that is based on the integration of data of GPS systems which carries out measurements with large observation angles diversity practically in continuous time. The proposed algorithm allows a robust assimilation of the GPS atmospheric delay measurements in the multipass DInSAR processing and found on a linear approximation with the height of the atmospheric delay corresponding to a stratified atmosphere. Achieved results show a significant mitigation of the seasonal atmospheric variations.

Adaptive spatial multilooking and temporal multilinking in SBAS Interferometry

We present an adaptive filtering algorithm for accurate estimation of differential interferometric phase and coherence, in order to better support the SBAS approach. The typical method to enhance interferograms is a simple boxcar multilooking, that improves the signal-to-noise ratio (SNR) at the expense of the spatial resolution. The presented technique is based on an adaptive space filter that identifies and averages statistically homogeneous pixels in a given neighbourhood. It also consists of a phase filtering operation in the time domain via the temporal multilink for recovering the consistency of the interferometric phase (phase triangulation). Results on real Cosmo-SkyMed data, acquired over the city of Naples, are shown in order to demonstrate the effectiveness of the presented technique.

SAR coherence tomography: A new approach for coherent analysis of urban areas

In this paper a novel procedure for the filtering of interferograms based on the extension of SAR Tomography on multilook data is presented. This algorithm, named “Component extrAction and sElection SAR” (CAESAR) takes advantage of Principal Component Analysis to handle temporal and angular decorrelation effects as well as of SAR Tomography principles to solve the distributed layover at the interferogram generation stage. Results on Cosmo-Skymed real data confirm the effectiveness of the proposed procedure.

Non-local methods for filtering interferometric SAR datasets

Multipass DInSAR processing is a well established technique that allows accurate 3D reconstruction and monitoring of the Earth surface. Many multipass techniques, such as the Small Baseline Subset approach, carry out a spatial multilooking of interferograms to counteract the decorrelation effects and the thermal noise and, also, to reduce computational complexity through spatial subsampling, thus enabling large scale (even full frame) analysis. Multilooking, however, degrades the spatial resolution w.r.t the original full resolution SAR product and therefore does not preserve contours of structures and in general boundaries of areas characterized by different backscattering. In this paper we propose the use of more sophisticated non-local filtering algorithms for the implementation of the interferometric multilook operation to guarantee the same noise rejection power of multilooking, while preserving high-resolution detail. Different strategies are considered and discussed for the selection of the best predictor pixels. Results on real data acquired by the ENVISAT mission demonstrate the good performance of the proposed nonlocal filtering methods.

Potential of SAR for monitoring transportation infrastructures: an analysis with the multi-dimensional imaging technique

Differential interferometric synthetic aperture radar (SAR) has proven to be effective for accurate localization and monitoring of the displacement of ground targets. The high accuracy and spatial density of the measurements make this technique cost effective compared to the classical geodetic techniques typically used in the risk monitoring context. Ground infrastructure monitoring is typically carried out with in situ sensors. The new generation of high-resolution SAR sensors, however, allows one to acquire data sets with a spatial resolution reaching metric/submetric values. Here we investigate the application of a multi-dimensional SAR imaging technique, which is an extension of classical differential interferometric techniques, to very high resolution TerraSAR-X data in order to demonstrate the potential of this technology for monitoring of transportation infrastructures.

Potentialities of the use of spaceborne radar systems in the monitoring of structures and infrastructures

The processing of data acquired by Synthetic Aperture Radar (SAR) with advanced methods based on multidimensional imaging opens new perspectives in the monitoring of facilities in built environment. The availability of data acquired by recent meter-resolution satellite SAR sensors allows precise monitoring of even single buildings and infrastructures, therefore valuably helping in the detection and monitoring of exposed facilities. This paper deals with the potentialities offered by the use of SAR data in the framework of the built environment monitoring by discussing results obtained by the processing of Cosmo-Skymed data.