Term project draft idea

Digital Elevation Models (DEM) generation using evolutionary computation

The use of Geographic Information Systems (GIS) for watershed delineation and stream definition is a current practice in water resources engineering today. This technique is mainly based on Digital Elevations Models (DEM) developed by interpolation of topographic data. Although more accurate data can be obtained using Light Detection And Ranging (LiDAR), traditional GIS-based interpolation methods for the creation of DEMs have difficulty to process such large amounts of data. To address this issue, my current research is to evaluate different forms of DEM generation from LiDAR data to find the best way to generate DEMs that are adequate to watershed delineations. My idea for the term project is to explore the interpolation techniques for terrain creation available in MATLAB and apply evolutionary computation to search for solutions that can handle LiDAR data and also result in a more accurate representation of the earth surface. At this point I think my decision variables will be the different parameters of the interpolation methods and my fitness functions will be an error metric that will evaluate the accuracy of the interpolated values. Future work and applications to my research line would be to use the best DEMs generated during the exercise to watershed delineation in GIS and to evaluate the accuracy of the resultants watersheds.