VALIDATION OF A UAV-DERIVED POINT CLOUD BY SEMANTIC CLASSIFICATION AND COMPARISON WITH TLS DATA

Abstract. Recent years showed a gradual transition from terrestrial to aerial survey thanks the development of UAV and sensors for it. Many sectors benefited by this change among which geological one; drones are flexible, cost-efficient can support outcrops surveying in many difficult situations such as inaccessible steep high rock faces. The experiences acquired survey, with total stations, GNSS or laser scanner (TLS), not yet completely transferred acquisition. Hence, quality comparisons still needed. present paper is framed perspective aiming evaluate point clouds generated an context; data analysis was conducted comparing product homologous TLS system. Exploiting modern semantic classification, based on eigenfeatures vector machine (SVM), two were compared terms density mutual distance. proves its usefulness situation uniform distribution whole area producing cloud comparable more traditional systems.