Technical University of Denmark
5 files

Orthoimages used for the study "Comparative Assessment of Five Machine Learning Algorithms for Supervised Object-Based Classification of Submerged Seagrass Beds Using High-Resolution UAS Imagery".

posted on 2023-09-29, 09:26 authored by Aris ThomasbergerAris Thomasberger

The UAS platform used to take the images was a consumer grade, low weight quadcopter of the type DJI® Phantom 4 RTK. The payload was a 20 million effective pixels 1-inch CMOS sensor-equipped RGB-camera with an 84° field of view, 8.8 mm/24 mm focal length and f. 2.8–11 aperture. In total, four flights were conducted in 2021. Two on April 7th and two on September 9th. At each date, the first flight was performed at an altitude of 100m and the second flight at 30m, resulting in a Ground Sample Distance (GSD) of 27.41mm and single image dimensions of 150x100m as well as 8.22mm GSD and single image dimensions of 45x30m, respectively. All images were taken with a nadir-viewing angle (90◦). Image front and side overlap were set to 75% and flight speed to 3.5m/s during flights at 100m altitude and 1.5m/s during flights at 30m. Both flights conducted at 100m altitude produced 156 images each, while 415 images were obtained during the 30m altitude flights. One georeferenced orthomosaic was created for each of the four flights by stitching the obtained images using the image processing software Agisoft Metashape Professional® ver. 1.7.4. The actual spatial resolution of the generated orthomosaics resulting from the 100m altitude flights conducted in April and September was 2.9cm/pixel and 3.1cm/pixel, respectively. The orthomosaics resulting from the 30m altitude flights conducted in April and September had a spatial resolution of 0.87cm/pixel and 0.84cm/pixel, respectively. 


Development of tools for economically efficient mapping of eelgrass in Natura 2000 areas, journal no. 33113-B-19-141, European Maritime and Fisheries Fund and the Danish Fisheries Agency


Related publications (DOI or link to DTU Orbit, DTU Findit)

ORCID for corresponding depositor