Use of UAV-based photogrammetry products for semi-automatic detection and classification of asphalt road damage in landslide-affected areas

Transportation networks are severely affected by natural hazards, including landslides. The prioritization of maintenance works is required to preserve the efficiency and functionality road infrastructure. To overcome subjectivity traditional visual inspections for pavement condition assessment, advanced (semi-)automatic approaches have been emerging. Still, quantitative objective description damage typology extent, its severity classification remain major issues assessment landslide impacts on transportation routes. this work provide a ready-to-use tool semi-automatic asphalt-paved roads in landslide-affected areas support risk analysis planning mitigation measures. use 3D models 2D images as reconstructed from UAV-based photogrammetry investigated detect longitudinal transverse cracks assess their areas, rapid, systematic, less laborious alternative field surveys. A procedure proposed i) select exposed landslides, ii) rapidly map distresses selected sections, iii) quantitatively describe cracks, iv) classify according International Roughness Index (IRI). applied Province Como (northern Italy), where three test sites detailed analyses. results indicate that useful management purposes at different levels details providing four outputs: hotspot deformations, multi-criteria binary classifier damage, an IRI-based criterion rate quality, map. • combination improves characterization. Semi-automatic methods heterogeneous deformations asphalt pavements. used severity. Drones flying 10 m height fractures wider than 1 cm Landslides can cause roads.