Archaeological Excavation at Vulci, Italy. Image Credit: Duke University
Aerial Reconstructed View. Image Credit: Duke University
In 2019-2021 I had the great opportunity to learn more about how technology is being used to tackle challenges in archaeology as a Bass Connections Research Fellow in the Dig@Lab at Duke University. As someone who loved learning about history in high school, this was a super exciting project where I could apply engineering to the study of the ancient world.
There are many challenges associated with the excavation process in field archaeology. Traditionally, observations were captured and recorded in a manual process by researchers. Not only is this time-consuming, this can also lead to subjective data and introduce bias. So, many new tools have been introduced to facilitate the data collection and analysis process. This may include ground penetrating radar, drones, and even autonomous robots (some of which have been developed at the Dig@Lab).
The problem I tackled during my time in the lab was more focused on the latter half of the equation: how can data collected from new technologies in the field be processed effectively in order to generate useful insights? As researchers uncover a layer at a dig site, a 3D scan (through methods like photogrammetry) can be created to capture what the layer looks like. This is done because many artifacts may occupy the same site, but at different depths depending on age. For example, an uncovered foundation of a Roman building may be itself built upon the site of an Etruscan structure from centuries prior. A 3D scan can be very detailed, but is time-consuming and resource-intensive to record and process. An alternative lower-fidelity yet faster method is to capture a "2.5D" height map of the dig site. Working with this type of data poses its own challenges, and was the focus of my work.
The first part of my project focused on developing an analysis tool for 2.5D height map data. Please see the document below for more info.
In the second part of my project, I focused my efforts on developing higher fidelity visualizations of the archaeological layers. The data I used came from the archaeological site at Vulci, Italy. The Blender 3D rendering engine worked perfectly for this. This was the first ever three-dimensional rendering of the dig site created from 2.5D height map data.