Supervisors: Pete Atkinson, James Lawrence
Deadline for applications: 30 September 2018
Studentship funding: Full studentships (UK/EU tuition fees and stipend (£14,777 2018/19 [tax free])) for UK/EU students for 3.5 years. Unfortunately funding is not available for International (non-EU) students.
Why is this project interesting?
In recent years the number of satellites specialising in capturing imagery of the Earth’s surface has risen dramatically, and increased investment from both ESA and commercial satellite operators are likely to continue this trend over the next decade. With every point on the Earth’s surface captured multiple times per day there is now a huge opportunity to understand the world’s economies in more detail. Traditional techniques used throughout the remote sensing sector are no longer able to effectively handle the scale of data now available. A new approach based around machine learning is required to interrogate these data in a timely fashion.
This PhD will present the successful candidate with an excellent opportunity to develop world-leading capability in the use of machine learning techniques for imagery, video and trend analysis. The candidate will need to develop techniques to identify and understand the changes in key economic indicators such as mine output, urban growth, commodity extraction and storage and shipping.
The PhD will be supported by machine learning experts, physicists and data scientists at Geospatial Insight.
What’s in it for you?
Become expert in the application of machine learning for the analysis of satellite borne imagery. The Earth observation (EO) sector is in a phase of growth as businesses are beginning to understand the value of EO data, and machine learning approaches are reducing costs, making the business proposition viable. Thus, this PhD provides skills and expertise in a sector where demand for such skills is high.
Develop links with external organisations. This project benefits from linkages with Geospatial Insight, one of the UK’s leading technology businesses for the development of downstream imagery processing services and products.
Join an exciting research environment. You will benefit from the research training programmes offered by the Faculty of Science and Technology at Lancaster University, by being part of the large and vibrant Lancaster Environment Centre and by becoming a member of the Geospatial Data Science research group. This project is at the cutting edge of what is possible using EO data and there is great potential for high quality academic publication of the results.
Who should apply?
We are seeking applications from graduates with a good (i.e., 1st class or 2.1) Undergraduate degree, and preferably also a Masters degree, in a Machine Learning-related subject. You should have a strong background in computer science, mathematics, physical science or geography/environmental science with strong quantitative (e.g., programming) skills. You must have demonstrable potential for creative, high-quality PhD research.
How to apply
Please download the Use of Machine Learning for Regional and Country Wealth Prediction information for the application process.