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

Apply Here

This is an exciting and unique opportunity to join a team of researchers developing and optimising automated (artificial intelligence (AI)-based) analysis of image and acoustic data as applied to seabed characterisation. This PhD combines cutting edge data collection and analysis. You will develop: 1. Practical boat-based data acquisition skills using remote and autonomous underwater vehicles and 2. Data analysis (image/acoustics) skills, including building, applying, and evaluating new algorithms, based on deep learning. 

Your PhD will include four highly publishable areas:

1.      An evaluation and application of methods for extracting, georeferencing and aligning image and acoustic data from the AUV, creating combined data tiles.

2.      The development of analysis pipelines that optimise the information that can be extracted, by a machine, from the tiles (above), including optimised image augmentation.

3.      The assessment of the influence of AUV altitude, and environmental factors (e.g., water clarity) on machine performance including/excluding the acoustic data.

4.      The development and promotion of AUV standard operating procedures (SOPs) for image and acoustic data collection when applied to assessing biogenic-reef PMFs. 

 The context of this PhD is in environmental monitoring and conservation and our partners include the fish-farm and regulatory sectors who are contributing an extra £15K to enable additional fieldwork. This PhD would suit a biologist/ecologist with an aptitude for programming or a data-scientist with an interest in ecology. This PhD will be based at SAMS, a highly supportive, well equipped, and internationally recognised research-orientated laboratory, located within a few metres of the sea. During your PhD you will be mentored by Tom Wilding (Director of Studies) and our in-house image analysis expert and co-supervisor, John Halpin, via ad-hoc/weekly catch-ups with further expert input from Keiller Nogueira and Jaime Zabalza (Universities of Stirling and Strathclyde respectively) as necessary and at least monthly. Field-work support will be provided by SAMS’ ROV/AUV technical team. You’ll be joining ca 40 SAMS PhD students and be able to develop your broader skill base (e.g. communication) via completion of the SUPER Postgraduate Certificate in Researcher Professional Development.

AI-assisted image analysis, is a massive subject area and the skills you will learn will establish you as a leader in this field. Employment prospects of PhD graduates with these skills are exceptional; PhD graduates from the group have gone onto highly successful careers in academic, regulatory and industry/consultancy sectors. 

Background: salmon farming, in Scotland, currently encompasses >230 sites, from Loch Fyne in the south to Shetland in the North. The industry seeks to expand, doubling economic value by 2030. 

The detritus from salmon farms (organic matter) is dispersed around the fish-cages to an extent that is determined by the receiving environment (depth, local current regime etc) and this material variously impacts the seabed. The Scottish regulator (SEPA) and their advisors (MICA-partners, NatureScot) are required to ensure sustainability of the sector and that the receiving environment is not adversely affected beyond local scales. Of particular concern are priority marine features (PMFs) which are individual taxa/biotopes considered to be highly sensitive, rare and/or which provide critical ecosystem functions. Examples of PMFs range between individuals of the coral Swiftia sp to sensitive biogenic reefs such as mussel-beds and maerl. The sector (industry and regulators) needs non-destructive, scale-appropriate methods for identifying and mapping PMFs. 

The successful candidate will initially focus on developing innovative, machine-assisted approaches to mapping the locations of blue and horse mussel reefs, both of which are local to SAMS (https://marinescotland.atkinsgeospatial.com/nmpi/) (e.g., Lochs Levan and Creran). Our surveys will extend to maerl as surveying resources/partner input allows. 

This project builds on and develops our existing expertise by bringing together hardware (AUV, ROV, graphics cards), software and expertise uniquely available within the MICA team. ROVs are excellent for smaller-scale surveys but the sector needs options for assessing prospective sites, and changes occurring at those sites, that are further from shore, more exposed and in deeper water. AUVs bring numerous cost and logistical advantages to surveys, both in relation to aquaculture and beyond in these more challenging environments, they offer the simultaneous collection of acoustic data and are highly efficient. 

Related projects currently underway (led by Tom Wilding).

NS3D: https://www.sams.ac.uk/science/projects/ns3d/ 

SEA-AI: https://www.sams.ac.uk/science/projects/sea-ai/

BacMetBar: https://www.sams.ac.uk/science/projects/bactmetbar/

For the supervisor team see:

https://www.sams.ac.uk/people/researchers/wilding-dr-thomas/

https://scholar.google.com/citations?pli=1&user=PqlaOMIAAAAJ

https://www.sams.ac.uk/people/researchers/halpin-dr-john/

https://www.stir.ac.uk/people/1427473

https://www.strath.ac.uk/staff/zabalzajaimemr/

See also: https://www.sams.ac.uk/facilities/robotics/ for details of SAMS’ robotics facilities.

The start date of this project is: 2 October 2023

The 3½ year studentships cover:

• Tuition fees each year at Home (UK) rate.
• A maintenance grant each of around £15,000 per annum (for full-time study)
• Funding for research training
• Part-time study is an option, with a minimum of 50% of full-time effort being required.

Applicants should normally have, or be studying for:

• A postgraduate Master’s degree from a degree-awarding body recognised by the UK government, or equivalent, or
• A first or upper second class honours degree from a degree awarding body recognised by the UK government, or equivalent, or
• Other qualifications or experience that affords sufficient evidence of an applicant’s ability to work at the academic level associated with doctoral study.

This PhD may require you to work, for limited durations, on small boats (full training and support will be given).  

Apply Here