Analysis of the soil sealing in the urban area of Rome through automatic processing of satellite data with neural networks

Abstract

One of the last and more pressing requests to the researchers working in the field of Geomatics is to research, validate, and propose new strategies for the rapid mapping of different contexts, with low-cost solutions. The continuous implementation of image-matching algorithms and their use in structure from motion (SfM) software allow using new sensors and implementing new strategies for the production of 3D models starting from an image-based approach. In the last years, another central issue for the researchers has been related with the documentation of cultural heritage (CH) artifacts using different sensors and techniques. In the experience presented in this paper, the attention was focused on these two central aspects: a test of a commercial off-theshelf (COTS) steadicamera for the rapid 3D documentation of two cultural heritage artifacts was proposed. The two sites chosen to evaluate this mass market sensor were as follows: the Basilica of San Nicola (Tolentino, MC, Marche Region, Central Italy) and the Castello del Valentino (Torino, TO, Piedmont Region, North Italy). The metric products obtained with the Steadicam were compared with more consolidated techniques such as close-range photogrammetry (CRP) and terrestrial laser scanner (TLS). The products derived from the different techniques were then evaluated and compared, and an overall assess of the use of this new solution was made.