PSI - Issue 84

Michela Pulsoni et al. / Procedia Structural Integrity 84 (2026) 214–222 M. Pulsoni et al. / Structural Integrity Procedia 00 (2026) 000–000

216

3

creating reliable Persistent Scatterers (PS) for time-series analysis. In contrast, water bodies (rivers, lakes) beneath bridges absorb radar signals almost completely, providing negligible backscatter. This contrast creates an ideal scenario: the bridge structure appears as a clear series of highly coherent measurement points against a background of low or no signal from the water below. Advanced Differential Interferometric Synthetic Aperture Radar (A-DInSAR) detects surface deformations by analyzing phase differences between successive satellite radar observations of the same area. These phase variations in the reflected radar signal from stable measurement points, termed PS (an object that maintains constant reflectivity over time), are directly correlated with ground movements, enabling the generation of high-resolution deformation maps. Post-processing analysis was conducted using the PS-Toolbox software suite, developed by NHAZCA S.r.l. and integrated into QGIS. This platform enabled the visualization of displacement time series, Line-of-Sigh decomposition into vertical and horizontal components, interferometric cross sections. 2.2. Photo-monitoring Among the most innovative remote sensing techniques currently available, photo-monitoring undoubtedly plays a key role thanks to specific characteristics that make it complementary to many other techniques and systems. PhotoMonitoring is a non-invasive monitoring solution that exploits the widespread availability of optical sensors to remotely control deformations or changes affecting the object of interest, whatever its nature (Cosentino et al., 2022). This technique is based on the concept of digital image processing , i.e., the processing of digital images with the aim of extracting data and information for remote sensing applications. Essentially, the technique relies on the extraction of information through the comparison of different types of images (e.g., satellite, aerial or terrestrial) acquired at different times over the same area and scene. In particular, the technique mainly enables the following types of analysis: • Change Detection (CD), which is an analysis aimed at identifying the location and magnitude of changes between a pair of images acquired at different times. This analysis can be performed through a simple pixel-by-pixel difference, or by means of more advanced algorithms that allow full-field analyses ( • Fig. 1a). • Digital Image Correlation (DIC), which is an optical-numerical measurement technique capable of estimating 2D full-field surface displacements or strains of any object present in the scene. Deformations are computed by comparing and processing digital images of the surface of the same object, acquired before and after the deformation/displacement event, through a pattern tracking process ( • Fig. 1b).

a)

b)

Fig. 1. Schematic examples of (a) Change Detection (CD) and (b) Digital Image Correlation (DIC). To perform monitoring through photo-monitoring – whether based on DIC or CD – and to extract reliable information on a variable quantity to be controlled (e.g., displacement magnitude, displacement rate, extent of changes, etc.), the images must first undergo an accurate alignment process, i.e., a co-registration step. The objective of this phase is to make two or more images directly comparable through appropriate transformations applied to the