Close up of skull and other bones CTC14 by Ian Watt

Code The City 14 – Archaeology Meets Data Science

In 2006-2007 an excavation took place of St Nicholas Kirk. It uncovered over a thousand sets of human remains, a rich collection of artefacts, and structural parts of the building going back over a thousand years.

At CTC14 we brought archaeologists, historians, medical specialists, data experts, designers, developers and many others to look at what we could do with the data that emerged from the excavation as well as some the remains and objects retrieved during the dig.

We also had a number of sets of human remains, a first for Code The City, which were laid out correctly for experts to work on photographing them in a side room from the main event.

What happened at the Weekend?

We commenced the Saturday session with 28 people in attendance. Following a talk on the archaeology of Aberdeen from two researchers from Aberdeen University, we then followed the normal CTC approach, ideas were put forward for what we could work on, based on challenges, opportunities and barriers that had been identified.

Intro talk CTC14 - photo by Ian Watt
Intro talk CTC14 – photo by Ian Watt

These fed into six projects around which  teams were formed.

The purpose of the weekend was to generate and test rapid prototypes of approaches that could form longer projects that used the data, photographs, information and other assets from the original dig.

Two of the teams were involved in photography and examination of a number of skeletons and bones which were laid out in a separate room. These were teams SK and team Spondylosis.

In the other main room four further teams were formed: Unity Burials, Skelocator, Skeleton Bridge and PR and Marketing.

Each team worked in sprints of two hours or less. At the end of these sprints the teams came back together in the main area and we had stand-up reports from each one on their progress. This continued throughout Saturday after which we had pizza and refreshments together.

On Sunday the teams went straight back into their projects and continued working in short sprints up to the late afternoon at which point each team did a complete show and tell of their projects: using the main projector to show what problem they had been working on, what tasks they’d completed, what outputs they had achieved, what had gone well and less well – and what they would do if they had more time or budget to work on the project beyond the confines of the weekend.

All code developed over the weekend was captured in Github as were some other digital artefacts. Each team had its own repository.

The projects

Below is a summary and a link to each of the six teams’ work.

Team Skelocator

This team set out to identify and extract data from miscellaneous sources arising from the excavation of the Mither Kirk in 2006.

This addressed two specific challenges

  • converting data from a Microsoft Access 2 Database
  • identifying co-ordinate information for skeletons and other remains from scanned plans.
Team Skelocator - photo by Carolyne Smith
Team Skelocator – photo by Carolyne Smith

The former was a 2006 database of transcribed data from the levels notebooks created at the dig and which recorded the various levels of the excavation. This in itself exposed the issued with storing data in proprietary, non-standard formats and the challenges of long-term access.

Much more detail is contained in the project’s README.MD file on Github.

Team Unity Burials

This project sought to take output data from Skelocator and create a 3D model of the burial and show how each set of human remains could be located in 3D space. In time, the scans and models from the three teams working with Skeletons could also be placed in the space – moving from generic representations of the skeletons to accurate models showing orientation, missing limbs etc. The project can be found on Github.

Team PR & Marketing

This small team worked on a knowledge-base of facts about the excavation of the Kirk and a comprehensive Q&A document which could be reused to complement existing website information about the dig. These are contained in the project repo.

Team Spondylosis

This team helped and instructed on laying out skeletons and choosing bones for 3D imaging.  They found a 1st rib of Skeleton SK176 which appears not to have been fully recorded/photographed and appears to have 2 cut marks on it. They worked out how the bone may have become cut and did 3D scanning.

They identified questions for Osteoarcheologist’s use of photographs of skeletons

  • What is the best way they would like to use photographs/ images to examine skeletal material?
  • Quality of image/ photograph?
  • Scale
  • Virtual reality?
  • 3D model
  • Choose best/ rare examples of pathology and preserve them in 3D?
  • Dental record of each skull
  • Pelvis- articulated?

A sub team used the turntable for skull 360 degree and 3D imaging.  They identified that the process needs plenty of height, and a lot of light and a white screen behind the subject.

If the embedded 3d model does not show in this page, click on the link to view it on Sketchfab.


They took many photographs from various angles. On Sunday whole skeletons were laid out on blue fabric on long table.

Another sub-team initially started looking at four fundamental aspects of 3D modelling of archaeological artefacts:

  • How to produce 3D Photogrammetry Models of archaeological artefacts, such as bones and skeletons
  • What 3D software is available (commercial and free)
  • How to use the software
  • How to transfer the 3D models into Unity

A first aim was to confirm very early on that any 3D models they may produce could be imported on to the Unity platform.  As a ‘Proof of Concept’, they took an existing 3D model of an archaeological trench produced  earlier in the summer, which was saved on Sketchfab (which is one of the main internet web sites for the storage and display of 3D models).  They exported this example of a 3D model to a USB storage device (including point cloud, mesh model and texture model).  The Unity Team very quickly proved that they could import the mesh and texture models and spatially place them within the Laser 3D Model of the Kirk on the Unity platform.

As they had already used AGIsoft PhotoScan 3D photogrammetry software (student edition) for trial 3D models such as  archaeological trench model, they decided to try and produce some initial example models of bones.  As the model software can take a long time to produce models from sets of photographs at high resolution (10+ megapixel), it was agreed to produce a few photographs set at low resolution (3 to 5 megapixel).  These low-resolution photographic sets would still take many hours to produce the 3D models of bones.  Also, there was some concern on how best to photograph the bones, so some early test models of the bone artefacts would help guide follow up photographic tasks.

One team member had also identified a mobile phone application for 3D photography of museum artefacts suggested by Historic Environment Scotland (HES) called Qlone.  They carried out a quick trial (with great success) and passed the application and method of use to other team members to explore and use, while we continued to look at the use of professional commercial software such as AGIsoft PhotoScan.

At the end of Day 1, producing medium resolution 3D Photogrammetry models of bone artefacts up to the size of a skull, was going well using Qlone.  However, producing models of anything bigger, such as collections of bones and skeletons would be a problem due to the techniques Qlone uses such as a photography grid and the base line resolution of the Qlone application itself (big loss of quality and resolution) above about 25cm x 25cm items.

The team identified a number of challenges in producing 3D Photogrammetry models using PhotoScan and similar software, which were documented.

Next day, they aimed at trying to produce some 3D models of whole skeletons, with PhotoScan at a high resolution, with different background and using as small a number of photographs (to minimise the time to run the software).  This produced a number of observations which were recorded.

Another sub-team experimented with QLONE and iphone and other mobiles. The QLONE mat was downloaded and printed at A4 for the smaller bones, A3 for the medium bones and A2 for skulls but the larger mat needed extra lighting and at least 2 people to travel around the mat keeping all 4 corners in the shot as well as lighting the face being scanned. Software easy to use for smaller bones.  The scans were imported into Sketchfab and uploaded to various drives and onto Facebook.  Scans were done of six or more bones including SK176 skull – this is the SK with 6 blade wounds which had dried and distorted and so was scanned in pieces with the hope that they can be put back together digitally.

You can view a range of scanned objects in this Sketchfab gallery.

Team SK

This was a team of one who went through  all of the pictures taken of the skeletons in St Nicholas Church before they were removed. This is the digital photos taken by the digging team for SK recording only, not the photos taken by the 2 professional archaeologists on site or other photos taken by the team

They opened all the folders with the SK numbers, checked the photo taken for the skeleton number and have renamed the photos so that they are named the SK number.jpg making it easier for people to find the photos of the skeleton. All small folders have been put into one large folder keeping them together.

sTeam Skeleton Bridge

This team’s project was to provide data from the Team Skelocator team in a format that the Team Unity burial team can use. See


A much more detailed report from the three Bone Teams is available to download: CTC14 – Notes from the bone teams


This event was well attended and spawned six mini-projects.  These explored some very interesting techniques for how to catalogue, explore and present data and other artefacts from excavations. The prototypes created shows how with further investment of time and some budget, that human remains could be scanned as 3D objects and located in a three-dimensional space as they had been in the original burial site.

Header photo by Ian Watt CC_BY_SA