Automatic Detection and Segmentation of Barchan Dunes on Mars and Earth Using a Convolutional Neural Network
Lior Rubanenko, Sebastian Pérez-López, Joseph Schull, M. G. A. Lapôtre
Abstract
The morphology of isolated barchan dunes on Mars and Earth may shed light on the dynamic conditions that form them, their migration direction and the physical properties of the sediments composing them. Thusfar, dune fields have been largely manually analyzed from aerial and satellite imagery, as automatic detection techniques are often not sufficiently accurate in outlining dunes. Here, we employ an instance segmentation neural network to detect and outline isolated barchan dunes on Mars and Earth. We train and test the model on martian targets using Mars Reconnaissance Orbiter (MRO) Context Camera (CTX) images, and find it to be robust, with sufficient accuracy ~80% to characterize dune field dynamics. Using our trained model we detect and map the global distribution of barchan dunes relative to previously mapped dune fields, and find that barchan dunes are more abundant in the northern hemisphere than in the southern hemisphere. These contrasting abundances of barchans may reflect latitudinally dependent wind regimes, sediment supply, or sediment availability.