Faculty Mentor(s)

Dr. Darin Stephenson, Mathematics and Statistics; Dr. Brian Yurk, Mathematics and Statistics

Document Type

Poster

Event Date

4-22-2022

Abstract

Lake Michigan dune complexes evolve as winds and waves move the sand, causing major topographic changes over time. The Hope College Dune Group has been using remote sensing data from drone imagery to model various aspects of these dunes. In order to better understand the dune dynamics, a digital terrain model (DTM) is desired. A DTM is an elevation map of the dune’s bare ground surface which excludes ground obstructions such as trees and bushes. We have attempted to use traditional methods that have been developed for processing LIDAR data in order to construct these DTMs from our drone imagery, although these perform poorly. In order to improve the quality of our DTMs, we have developed a new approach and have performed initial testing with this technique. This approach uses an artificial neural network to classify the ground elevation of small 1-meter by 1-meter tiles within the drone imagery. This neural network is given the elevation of all points within the tile, including points that are positioned on nonground objects such as tree canopies and bushes. Preliminary results have shown promise from this technique, although more work needs to be performed in order to further fine-tune the model.

Comments

Research reported in this publication was supported in part by funding provided by the National Aeronautics and Space Administration (NASA), under award number 80NSSC20M0124, Michigan Space Grant Consortium.

Included in

Mathematics Commons

Share

COinS