Low-cost Uav for the Environmental Emergency Management. Photogrammetric Procedures for Rapid Mapping Activities

The main research topic of this PhD thesis is the analysis and the development of photogrammetric procedures for the processing of digital images acquired by an Unmanned Aerial Vehicle (UAV). The mini-UAV “Pelican” is a low-cost aerial platform capable of autonomous flight and equipped with a photogrammetric payload for rapid mapping purposes. It has been developed by ITHACA (Information Technology for Humanitarian Assistance, Cooperation and Action) a non-profit association with the main goal to carry on operational and research activities in the field of geomatics for analysis, evaluation and mitigation of natural and manmade hazards. The UAV “Pelican” is capable to perform a complete automatic image acquisition, according to a predefined flight plan specifications. This work was aimed at developing an automatic approach for the digital image processing. The photogrammetric processing led to the automatic extraction of Digital Surface Models (DSMs) and the generation of Solid True Orthophotos (STOP). Robust and reliable feature extraction and matching techniques were developed and implemented in Matlab code. In particular SIFT operator, was employed for the automatic extraction of tie points. Moreover, a multi-image matching method (Multi-Image Geometrical Constraint Cross-Correlation, MIGC3) and a robust filter (Self-tuning Standard deviation Median Filter, S2MF) were developed for the automatic DSM extraction. The algorithms were validated and tested on aerial images and on images acquired by the UAV “Pelican” over an archaeological area. The experimental results confirm the high performance of these techniques and the suitability of the photogrammetric process for rapid map production, according to the requirements of the ITHACA project.