“ The wrecks in the Royal Port were discovered through a new high-resolution sonar survey of the seabed. This produced enormous quantities of data that was fed into a machine learning algorithm trained to recognise the “signatures” of shipwrecks. The initial results were promising, with excavations on targets generated by the algorithm revealing a small boat and a 30m-long merchant ship. “

Any tips on how machine learning can be used to find hints of sunken ships from sonar images?