We are home to the world’s first Geospatial Slavery Observatory. The majority of today’s slaves live in developing countries where many slave-based activities are visible in satellite imagery (for example brick kilns, mines, fisheries and farms). By identifying slavery locations using geospatial intelligence, we answer the demand within the development and human rights communities for scientific data that can underpin policy formation and humanitarian operations. Our process is to:
• Compile, synthesise and integrate spatial data to detect and eventually prevent slavery.
• Develop (automated) methods with as much data as possible at as low as possible cost, with known levels of uncertainty.
• Act as a conduit for all observations of slavery activity.
Employing state-of-the-art techniques from geoinformatics and data modelling, geospatial intelligence, and location-based services, we conduct comparison and cross-corroboration of archival satellite imagery with various open source data, and use imagery analysis methodologies. We also use our expertise on volunteered geographic information to enable quality crowd-sourcing, and are applying machine-learning techniques that automate identification via a prototype feature extraction algorithm.
Our work was featured in a Telegraph article in October 2016. We are now developing new pilots in Ghana (fishing), India, Nepal and Pakistan (brick kilns), Thailand (fishing), Brazil (charcoal camps) and the Democratic Republic of the Congo (mining).