RESEARCHERS at the University of Texas have taught artificial intelligence how to spot shipwrecks on the ocean surface from the air.
In collaboration with the United States Navy’s Underwater Archaeology Branch, the computer model is 92% accurate in finding known shipwrecks.
The project is focused on the coasts of the mainland United States and Puerto Rico, and is now ready to be used to find unknown or unmapped shipwrecks.
The overall goal of the project is to build a customisable archaeological machine learning model, which would be able to quickly switch between predicting different types of archaeological features underwater as well as on land, including Maya archaeological structures and caves.