PSI - Issue 78

Mario Graniero et al. / Procedia Structural Integrity 78 (2026) 1040–1047

1042

The versatility of FW-UAVs is significantly enhanced by their payload capacity, allowing them to carry a diverse range of advanced sensors crucial for comprehensive damage assessment and vulnerability analysis. These payloads commonly include high-resolution optical cameras for detailed visual inspections, LiDAR (Light Detection and Ranging) systems for generating precise 3D point clouds, multispectral and hyperspectral sensors for material characterization and stress detection, and thermal cameras for identifying anomalies related to structural integrity through infrared thermography (IRT) (Nooralishahi et al., 2021). The integration of various UAV data types, whether analyzed independently or in conjunction with other sources, allows for the exploration of multiple engineering solutions (Gaspari et al., 2022). The data collected by these sensors provides a rich dataset for various analytical purposes. High-resolution optical imagery, for instance, is invaluable for visual crack detection, subsidence monitoring, and the assessment of pre existing weaknesses in structures. The application of computer vision techniques in Structural Health Monitoring (SHM) has gained considerable traction for processing UAV-acquired data (Ngeljaratan et al., 2024). A single UAV can be effectively utilized for seismic monitoring and safety assessment of linear infrastructures, ensuring their service continuity. Advanced computer vision-aided procedures, such as those employing the Maximally Stable Extremal Region (MSER) method for covariant region extraction, Speeded-Up Robust Features (SURF) and K-nearest Neighbor (KNN) algorithms for feature detection and matching, and the Maximum Sample Consensus (MSAC) algorithm for model fitting, enable precise calculation of physical dynamic displacement by correlating pixel displacement with a scale factor based on the pinhole camera model. Such procedures have been validated through full-scale shake-table tests of natural gas pipeline assemblies, demonstrating the feasibility of UAV-based SHM for linear infrastructure monitoring (Ngeljaratan et al., 2024).

Fig. 2. Adapted from (Hu and Assaad, 2023), showing the distribution of UAV platforms cited in the reviewed articles.

Furthermore, LiDAR data is fundamental for creating highly accurate Digital Elevation Models (DEMs) and Digital Surface Models (DSMs), which are critical for analyzing ground deformation, landslide susceptibility, and subtle changes in infrastructure geometry. These capabilities are particularly advantageous compared to conventional space based optical remote sensing systems, which often provide lower ground resolution and can be hindered by cloud cover, delaying the acquisition of crucial information (Brauchle et al., 2024). Beyond pre-event monitoring and post-disaster assessment, FW-UAVs offer a significant advantage in ensuring operator safety. Traditional inspection methods often necessitate personnel working in hazardous environments, such as steep slopes, unstable terrain, or damaged structures, and involve expensive, less flexible, and risky in-situ technical tests like those performed with elevating platforms and underbridge units (Gaspari et al., 2022; Nooralishahi et al.,