Kaleb Alekel is a student at the State University of New York at Geneseo. He is currently working toward a bachelor’s degree in theoretical physics. His current favorite topic in physics is fluid dynamics. Aside from physics, he has an interest in geography and has studied population decline related to interstate construction in the Great Plains region of the United States. He is interested in using the Wolfram Language to apply the research methods he has learned in physics to other disciplines. His hobbies include camping, hiking and making pottery.
Project: 3D Modeling of Topographical Maps
Goal of the project:
Create a function that makes a 3D image of a landscape from a corresponding topographical map.
Summary of work:
Using a convolutional neural network, we can input an image of landscape contour lines and get a corresponding grayscale heat map. This encodes the height values as pixels with intensity values ranging from 0 to 1. These height values can then be extracted and used to reconstruct the landscape using ListPlot3D.
Results and future work:
I was able to make a neural network capable of outputting a heat map from a randomly generated topographical map. However, due to the small set data used to train the network, it is far from as accurate as it would need to be to reconstruct the landscape as a 3D image.