How can AI be used to conserve HMS Victory?
- View news filtered by: News / Stories
- View news filtered by: Portsmouth
- View news filtered by: Historic Ships
- View news filtered by: Victory Live: The Big Repair
The conservation of an international icon like HMS Victory requires all-hands on deck, making use of both the tried-and-true historic techniques, and the cutting-edge of technology. AI and machine learning algorithms are a hot topic now, but can it help with the conservation of Vice Admiral Lord-Nelson’s flagship and the museum’s collection?
Three master’s students from the University of Southampton (Siddhi Mahendra Pawar, Donheng Wang and Arundati Roy), alongside Dr. Rodrigo Pacheco-Ruiz, The National Museum of the Royal Navy’s Archaeological Data Manager and Amy Adams Collections Information and Access Manager, have set out to do just that. Since this decade long restoration began in May 2022 archaeologists have taken over 3,000 images in addition to the high-resolution images taken to produce multiple 3D digital models of the ship. These additional images had to be sorted and analysed manually, obviously a time-consuming process. But now, but with the aid of AI engines, this process has been automated, increasing the quality and resolution of the recording of the works.
The AI-based algorithm matches images stored in different locations and adds them to the 3D model of HMS Victory to ensure it is as detailed as possible. Keeping a detailed and accurate record of the conservation of HMS Victory is vitally important considering its scale. This record helps capture vital information about historical repairs to the ship, parts that have degraded and can even help inform the next stages of the project.
More than matching simple images this project has pushed the boundaries of what AI is capable of, working on incredibly high resolution, complex images. This project is also no doubt at the forefront of AI usage in archaeology, although it is a far cry from the traditional perception of archaeology.
Credit Dr Rodrigo Pacheco-Ruiz
This kind of technology is not only being used on HMS Victory, and is being applied to the collection of the National Museum of the Royal Navy more broadly. The collection of the museum encompasses everything from figureheads to documents to videos, and the ways in which these items are catalogued has changed considerably, requiring more detailed information. Collections are also expected to accessible digitally, all of which can increase the workload of curators in museums.
Utilising AI for ‘entity recognition’, it can retrieve and sort digital records, and can be used to label records in a way that is more accessible to visitors. This means that the AI could label something as ‘WWII’ ship, as well as its official name, which may not be known to the average visitor. It is clear that AI has applications both in collections, and in conservation projects such as HMS Victory.
For HMS Victory, both the modern and historic techniques work together, as the 3D model helps inform the teams hard at work using historic techniques to conserve this icon.