Approximate computationally expensive geospatial algorithms with machine learning
Many geospatial algorithms are computationally expensive. This prevents them from being used in interactive scenarios where the user can dynamically change the input parameters.
One example is the cartographic “depth to water index” which classifies terrain data according to the minimum elevation change to standing water. This index can be used to estimate soil moisture, an important factor when routing heavy vehicles in terrain.
The purpose of the Master’s Thesis is to apply machine learning methods to find ways to approximate this and other computations so that they can be used in interactive applications.
Fill in the application form below and we will contact you as soon as possible.