Generating UAV high-resolution topographic data within a FOSS photogrammetric workflow using high-performance computing clusters

• DOI: 10.1016/j.jag.2021.102600
• Publication Date: 2021-10-22T15:52:00Z
• Authors: Marco La Salandra; Giorgia Miniello; Stefano Nicotri; Alessandro Italiano; Giacinto Donvito; Giorgio Maggi; Pierfrancesco Del...; Domenico Capolongo
• Keywords: Orthophoto; Landform; Aerial image; Position (finance)

Photogrammetry is one of the most reliable techniques to generate high-resolution topographic data and it key territorial mapping change detection analysis landforms in hydro-geomorphological high-risk areas. Specifically, Structure from Motion (SfM) an emerging survey technique that addresses problem determining 3D position image descriptors estimate three-dimensional structures. Thanks potential SfM algorithm development Unmanned Aerial Vehicles (UAVs) allow on-demand acquisition aerial images, possible extended areas Earth surface monitor active phenomena through multi-temporal surveys. However, ability detect remote wide with a very countered by need capture large datasets which can limit photogrammetric process, due for high-performance hardware. This paper presents workflow based on Free Open-Source Software (FOSS), able return different outputs manage amount reasonable time, distribution computationally expensive steps computing clusters hosted ReCaS-Bari center scientific research. The results are given terms performance evaluations configurations setups workflow. HTC cluster test parallel SSH approach involved important reduction several hours processing time thousands UAV especially compared classic process single workstation commercial software. A test, aimed validate sever new HPC cluster, really good halving respect test.