Automation of urban technological census. The historical centre of Bethlehem.

The research proposal reports the outcomes of a research track concerning the automation of the architectural census of technological elements in urban environments, aiming at the development of a monitoring and management system for the built heritage. The proposal is focused on a set of specific elements (water tanks) stacked on the coverings of the historical centre of Bethlehem and leverages close-range photogrammetric acquisitions to train Deep Learning models. The model lifecycle management, from training to prediction and deployment, as well as the storage of both image data and metadata, is performed through the scalability of a Cloud enterprise architecture. Periodical scheduled monitoring enables comparisons across different periods, allowing the detection of modifications, removals, and additions, therefore identifying the insurgence of potential criticalities. The goal of the project is the definition of a protocol to automate the identification of recurrent elements and monitor their evolution through time.