Archeology meets AI to help preserve maritime heritage

Posted by

The University of Southampton has completed a world-first collaboration with the National Museum of the Royal Navy to explore how Artificial Intelligence (AI) can support the museum’s vital work in preserving the country’s maritime heritage.

Laser scanning HMS Victory to create a virtual 3D model of the ship.

Laser scanning HMS Victory to create a virtual 3D model of the ship. Image Credit: NMRN

The project sees three Masters students from Southampton work with archaeologists and curators at the museum to apply the latest AI technologies across a range of projects, including the restoration of HMS Victory – site of the Battle of Trafalgar. The famous flagship of Vice-Admiral Lord Nelson during.

Directed by the University of Southampton CORMSIS Research Group – a collaboration between southampton business school And this School of Mathematical Sciences – The project has given students the opportunity to push AI technologies to the limits when it comes to managing and applying large amounts of information held by knowledge institutions.

Currently undergoing an ambitious ten-year restoration HMS Victory ‘major repairs’ The project involves a team of shipbuilders, riggers, conservators and archaeologists who are diligently documenting every detail of the work. This helps to capture important information about the ship’s historical repairs, damaged parts and informs planning for the next stages of work.

Since the restoration of HMS Victory began in May 2022, archaeologists have taken more than 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 stored and analyzed manually, but with the help of AI, this process has now been automated, increasing the quality and resolution of recording actions.

Dr. Rodrigo Pacheco-RuizVisiting Fellow in Maritime Archeology at the University of Southampton and Archaeological Data Manager for HMS Victory at the National Museum of the Royal Navy, explains: “Archaeologists are obsessed with detail and if records are not stored correctly, important historical information is forever lost. Can get lost. This is where the help of our University of Southampton students has been invaluable. They have developed an AI-based algorithm to match images stored in different locations and add them to our digital 3D model to ensure it is as accurate as possible.

“This work has pushed the limits of the capability of this type of AI, as our images are extremely high resolution, complex and detailed. This project is really at the forefront of how AI is being used in archeology – we are demonstrating what is possible. This is certainly a far cry from the traditional notion of how an archaeologist spends his time.

The collaboration also included a project to help catalog the museum’s collection, which includes everything from figureheads, uniforms and maps to photographic, film and video materials.

In recent years there has been a huge change in the way objects are recorded for the museum sector. Where previously a curator would add a brief description of an object to a physical catalog card, they now must capture detailed information to allow digital cataloging and, importantly, retrieval.

For Archive Information and Access Manager National Museum of the Royal Navy, Amy Adams, comments: “As members of the public increasingly want access digitally, the curators’ workload in terms of capturing the necessary data has increased significantly. Of course, we still want people to visit in person, but the opportunity to share knowledge around the world and engage more people in our country’s maritime history is huge.

“By working with university students, we were able to use AI for entity identification – to retrieve and sort digital records. This has huge time-saving potential, but what was really interesting was the AI’s ability to label records from a visitor’s perspective, so records are categorized based on what people are searching for (e.g. ‘Ships of World War II’) as well as its official name. The human touch is not obsolete, as technology still needs that context to focus and move it forward, but it is certainly exciting to think about the potential of AI for the museum sector.

Professor Bart Bessens The Business School of the University of Southampton and KU Leuven University in Belgium supervised the students. Commenting on the collaboration he said: “This project has provided a fantastic experience for our students, giving them the opportunity to apply what they have studied in a ‘real world’ situation. They have used large language models – deep learning algorithms that can identify, summarize and generate content – ​​to automatically classify images and artifacts curated by museums. We believe that this could be of great interest to the wider cultural world, to institutions that want to enrich data related to their collections quickly, efficiently and accurately.

Source: university of southampton


#Archeology #meets #preserve #maritime #heritage

Leave a Reply

Your email address will not be published. Required fields are marked *