Detection of Stone Circles in Periglacial Regions of Antarctica in UAV Datasets

This paper tests three methods, based on template matching, the watershed transform and the sliding band filter, for the identification of stones circles, a natural pattern formed in periglacial regions of the Earth. All the methods take advantage of the 3D shape of the structures conveyed by Digital Elevation Models (DEM). The DEMs were built from milimetric imagery captured by Unmanned Aerial Vehicles (UAV) in Barton Peninsula, King George Island, Antarctica ($$62^\circ $$) during a field campaign developed in 2018. The best results were achieved by the sliding band filter method with a F score of $$83.9\%$$.