Tom joined the Campus as a Data Scientist in the Foreign, Commonwealth & Development Office (FCDO) hub in August 2019.
Prior to this he worked at Forest Research where he developed machine learning methods for detecting forest clear-felling from radar satellite imagery. Tom has also worked in private sector in the United States where he specialised in geodatabase technology for a software company.
Tom is particularly interested in how open source technologies commonly used in data science can be applied to geospatial and remote sensing imagery analysis. Tom has an MSc in Data Science and a BSc in Environmental Science.
We used machine learning to develop a model to identify areas of trees from satellite images in eastern Uganda, where the Mbale Trees Programme has been running since 2010.
Read more on Using Sentinel-2 images to measure the change in tree coverage in eastern Uganda: what does it mean for the Mbale Trees Programme?
Cattle are central to many people’s livelihoods in South Sudan, but there are very few recent data on the number of animals and their geographic distribution. Building on previous work, this blog explores new methods, based on convolutional neural networks, that can better distinguish cattle camps from similar landscape features.
Read more on Technical report: Predicting cattle camp locations in South Sudan from Sentinel 2 satellite imagery
On World Statistics Day 2020, our ONS FCDO Data Science Hub talk about exploring the use of satellite imagery to conduct a cattle census in South Sudan.
Read more on Counting cows in South Sudan