Be part of the world-renowned WorldPop Group at the University of Southampton, a global top 100 university (QS World Rankings 2021). WorldPop is an award winning, multi-sectoral team of researchers, technicians and project specialists working on producing data on population distributions and characteristics at high spatial resolution.
We are seeking a highly skilled and motivated Senior Research Fellow in image analysis and Artificial Intelligence to join our team. The successful applicant will be responsible for developing and implementing advanced algorithms and models to analyse and interpret images using AI techniques. Specifically, the candidate will be expected to extract meaningful data from time series satellite images to contribute towards improving the estimation of displaced populations, nomadic and seasonal migrant populations, abandoned, and repopulated settlements in low- and middle-income countries. This role requires expertise in image processing, machine learning and computer vision techniques, as well as a strong understanding of time-series satellite image data.
The applicant should ideally be familiar with various Earth observation platforms, aerial photography, and pre-processing of their products. The successful candidate will work closely with specialist GIS analysts and population modellers in the production of spatial estimates of temporary and mobile populations, contributing to work on constructing national high-resolution population estimates. They will primarily contribute to the work being undertaken for GRID3 (Geo-Referenced Infrastructure and Demographic Data for Development, grid3.org) in low- and middle-income countries, as part of a team of specialised WorldPop researchers and external partner organisations (e.g. Planet, Microsoft and Google). However, the candidate will have the opportunity to collaborate on projects funded by the United Nations, the European Commission and other international organisations such as GAVI. Experience working on population dynamics for humanitarian preparedness and response is welcomed.
The successful candidate will be responsible for (1) Designing and developing AI-based computer vision algorithms and machine learning to analyse and interpret images, such as feature extraction, segmentation and object recognition, (2) collect time series image data from various sources, ensuring data quality and integrity , (3) analyse time series image data to detect trends, anomalies and meaningful patterns (4) carrying out thorough experiments and assessments to evaluate the accuracy and efficiency of developed models and algorithms, (5) providing technical guidance and support to team members, as well as collaborate with external stakeholders. You will work as part of a multi-sectoral team of specialists, and under the supervision of Dr Sarchil Qader and the relevant project lead.
As well as joining the vibrant and well-connected WorldPop team based at the School of Geography and Environmental Science, you will have opportunities to lead high-impact publications, deliver workshops aimed at training project stakeholders and present at national and international conferences.
Essential and desired skills and backgrounds: Strong experience in computer vision, time series image analysis and AI. PhD or equivalent professional qualifications in a computer science/electrical engineering/quantitative discipline. Proficiency in programming languages such as Python, C++ or Java and familiarity with deep learning. Knowledge of optimization techniques, parallel computing and GPU programming for efficient model training and deployment. Familiarity and knowledge of coding in cloud-based platforms such as Google Earth engine or similar environments.
The funded position is until August 2025, with a strong likelihood of extension.