Summary
The project will develop interpretable machine learning models for estimating population of a region from satellite imagery and survey data.
Scientific Background and Motivation
More than one-third of the Sustainable Development Goals (SDGs) indicators established by the United Nations (UN) are defined in terms of total population or a specific demographic sub-population [1]. Up-to-date population information of a region is crucial for decision making including access to services, distribution of vaccinations, disaster relief, and many others. Traditional population data, such as census, are not adequate for this purpose since censuses are typically conducted decennially and countries with the greatest need for up-to-date population counts conduct them even less frequently. Population estimation using alternative data sources such as satellite imagery has received significant attention in the recent years. Both census-dependent [2] and census-independent [3] approaches have been explored with some success, and many of these methods have utilized advanced image analysis methods such as deep convolutional neural networks with promising results. This project will develop these methods further, to produce sustainable, interpretable and reliable machine learning models estimating population more effectively. It will involve utilizing contextual information, both spatial and temporal, integrating satellite imagery at multiple resolutions with publicly available data sources such as land cover map, etc., combining from census, surveys and microcensus data, and explaining the decisions made by these models to the end-users.
Aims and Objectives
The aim of the project is to develop sustainable, interpretable, and reliable machine learning models to effectively estimate the population of an area using satellite imagery and survey information. The project will investigate the following research questions: (1) Can contextual neighbourhood information, both short-range and long-range, improve population estimates by understanding the characteristics of the surrounding regions? (2) Can information from census, surveys, and micro-census be combined to track population reliably over time? (3) Can data from different sources and different resolutions be combined to acquire complementary information about an area? (4) Does uncertainty/bias differ in sparsely populated rural areas vary compared to densely populated urban areas? and (5) Can estimated population and associated uncertainty be explained to policymakers effectively?
Methodology
The project will incorporate satellite imagery of different resolutions and survey data using computer vision models, deep learning architecture and explainable model constructs to estimate population of a region.
Year 1: Familiarize yourself with the different datasets, i.e., satellite imagery, surveys, and censuses. We have several datasets available for a variety of countries from associated projects so experimentation and discussions with the supervisory team will help to define the case study examples. Familiarize yourself with different methods, i.e., predictive modelling, convolutional neural networks, spatio-temporal models, attention-based models, graph-based models, explainable models, Bayesian networks, etc. Build baseline models based on existing literature following earlier works done by our group, and assess their sustainability, interpretability and reliability.
Year 2: Investigate spatial contextual information to improve estimation of population by borrowing strength from neighbouring regions. Explore attention-based models to include short- and long-range neighbourhood information in population estimation. Integrate multiple data-sources to better identify zero-population areas. Assess the feasibility of estimating population from different resolutions. Investigate deep learning architectures for data assimilation, such as encoder-decoder neural networks.
Year 3: Investigate the potential of survey data in population estimation frequently and effectively. Develop spatio-temporal graph-based models for estimating population using census, surveys and micro-census data when available. Explore the explainability of the models using model agnostic methods, and Bayesian networks. Explore best practices to communicate the population maps to end-users and assess the utility of the population maps to policymakers.
This PhD is part of the NERC and UK Space Agency funded Centre for Doctoral Training "SENSE": the Centre for Satellite Data in Environmental Science. SENSE will train 50 PhD students to tackle cross-disciplinary environmental problems by applying the latest data science techniques to satellite data. All our students will receive extensive training on satellite data and AI/Machine Learning and field training. All students will experience extensive training on professional skills, including spending 3 months on an industry placement. See http://www.eo-cdt.org
More information for applicants can be found on our available Projects, How to Apply and dedicated FAQ webpages. You will also find additional resources such as How can a PhD help me with my career? and our ‘Demystifying the PhD application process’ webinar'
Continue with Facebook
