About the project or challenge area

Remote sensing and Earth observation (EO) from diverse sources, including satellite, airborne, and in situ platforms and citizen observatories, offer great opportunities to identify changes to the Earth’s surface across different scales. For environmental systems, such as climate and weather, different types of EO sensors are adopted, with distinct spatial, spectral, and temporal resolutions. This generates substantial quantities of data, and environmental remote sensing is now considered an area of “massive big data”. Traditional remote sensing techniques are inadequate for extracting the most meaningful information from these data. This calls for novel technologies to mine the potential information in a robust, accurate, and automatic fashion.

The latest developments in artificial intelligence (AI), in particular deep learning (DL), have gained tremendous interest in the field of remote sensing. DL is considered a state-of-the-art breakthrough in AI that automates the process of feature learning and feature representation and is capable of hierarchically extracting valuable end-to-end information about natural phenomena. Latest development in AI techniques such as deep convolutional neural networks, attention mechanisms, transformers, graph neural networks, and explainable AI further boost the precision of many practical applications with ground-breaking performance. The combination of remote sensing and artificial intelligence can help improve experts’ understanding of land, ocean, and atmosphere systems. This can lead to many benefits, including more accurate predictions about the behaviour of such environmental systems, the automation of data analysis, improved management of resources, and the discovery of new insights from complex datasets.

Why choose this project?

This project has external partner from UK Centre for Ecology & Hydrology based on the Centre of Excellence in Environmental Data Science (CEEDS). This setting provides a unique and stimulating environment, with series of research seminars, workshops and training opportunities provided across CEEDS to support AI-based environmental innovations in which the student will be encouraged to participate actively. There will be opportunities for industry placement depends on specific topic related to the theme.

About you

We are inviting students to propose an idea linking to the theme/challenge area. Good conceptual and practical knowledge of AI and sensor technology is desirable; programming (ideally in Python) and machine learning skills are assets. 

The student will receive advanced training in programming, with focus on AI and remote sensing. However, enthusiasm for geospatial science and remote sensing are by far the most important requirements.  

How to apply

All students can apply using the button below, following the Admissions Statement (PDF, 188kB). Please note that this project requires a research statement - for further guidance, please see Section B of the Research Statement template (Office document, 74kB). 

This project is not funded, for further details please use this link.

Before applying, we recommend getting in touch with the project supervisor. If you are interested in this project and would like to learn more about the research you will be undertaking, please use the contact details on this page.