Tom Morgan

        Tom Morgan

Research Student

My research focuses on developing and optimising automated analysis of image and acoustic data as applied to seabed characterization. For my PhD, I am using a combination of autonomous underwater vehicle technology and machine learning for image and audio analysis.

Contact details:
  • +44 (0)01631 559 000

Optimising marine image capture and analysis from autonomous underwater vehicles (MICA)

Using Autonomous Underwater Vehicles image and audio data of the seafloor is captured which will be fed into a machine learning algorithm which will identify protected marine features.

Salmon is the largest food sector in the UK economy and the sector has the objective of doubling in value. However, detritus from salmon firms is dispersed around the fish-cages which, determined by factors in the local environment (depth, local current regime etc) this material impacts the seabed. Of particular concern is the potential for this to damage priority marine features (PMFs) which are individual taxa/biotopes which are considered highly sensitive, rare and/or provide critical ecosystem functions. The PhD intends to utilise AUV’s to locate and survey biogenic reefs, the collected image and audio data will be returned and then combined with location data to create geo-referenced data ties. The images will then by annotated and run through a custom “multimodal” convolutional neural network. This should then allow the creation of a machine learning algorithm to categorise seabed data to allow salmon farms to identify if a potential location is likely to cause damage to PMFs and allow organisations to monitor the impact of salmon farms on PMFs.


Tom Wilding: SAMS

John Halpin: SAMS

Keiller Nogueira: University of Stirling

Jaime Zabalza: University of Strathclyde


SUPER, NatureScot, The Scottish Salmon Company Limited, Sustainable Aquaculture Innovation Centre

Start Date




2023 - Msc Data Science and Public Policy (Economics) - UCL

2021 - Bsc Mathematical Economics and Politics - University of Exeter