This paper presents an integrated methodology for the survey and semantic analysis of the Fossanova Abbey complex, combining architectural survey, laser scanning, UAV photogrammetry, LiDAR acquisition, and machine learning techniques for semantic segmentation of point clouds. The study investigates how supervised learning algorithms, particularly Random Forest, can support the classification and decomposition of complex architectural heritage datasets. The research also discusses the relationship between geometric interpretation, semantic annotation, and multiscale architectural representation, highlighting the potential of AI-assisted workflows for cultural heritage documentation and analysis.
Categories:
3_Architectural scale