Skip to content
Final Project
The main focus of this final project was pulling everything together. It was the most end-to-end assignment of the course, and that's what made it both the most challenging and the most rewarding.
The study area was Plot 3 at Purdue Wildlife Area, which covered a mix of terrain dense woodland, an open grass field, bare ground, and some farm structures. My lab parther flew the DJI Mavic 4 Pro at 122 meters AGL with 80% overlap, and despite some difficulties with the file sizes, I was ultimately able to process everything through Drone2Map and produce a true orthomosaic, a DSM, and a DTM to work from.
From there the analysis work in ArcGIS Pro covered a lot of ground. The orthomosaic and DTM comparison was one of the more intuitive outputs seeing how the spectral imagery and elevation data lined up. The classification work both unsupervised ISO Cluster and the supervised object-based approach showed how much the method matters depending on what you're trying to extract from the imagery.
The analysis that probably had the most real-world relevance was the tree presence within the 10-meter road buffer. Knowing where the tree canopy is encroaching on the corridor is exactly the kind of information a project manager would be looking for.
Overall this project was a strong capstone for the course. It tied together flight planning, photogrammetric processing, raster analysis, classification, and vector work into one cohesive workflow which is ultimately what UAS-based GIS looks like in practice.
AT319_Final_Report_Draft.docx.pdf
30.8 MB
Want to print your doc?
This is not the way.
This is not the way.
Try clicking the ··· in the right corner or using a keyboard shortcut (
CtrlP
) instead.