6,438 Decoders joined Amnesty to help find evidence of homes and schools in Darfur being destroyed

About Next Project


More than 6,000 volunteers from over 100 countries took part in comparing recent satellite images of villages in Darfur with older ones to determine where damage had occurred in the last two years

Using the remote locations mapped in our recent Decode Darfur project, we invited Decoders to compare two images of the same village and identify significant change in buildings and structures over time.

By studying these images and pinpointing villages that have been destroyed, Decoders have helped Amnesty International build evidence demonstrating that civilians have been systematically attacked.


Partnering with machine learning researchers from University College London, we are developing cutting edge artificial intelligence algorithms, using the large amount of data offered by the Decoders, aiming to extend the analysis to the whole of Sudan.


Decode Darfur was the second project for the Amnesty Decoders - a global network of digital volunteers for human rights research. Already there are more than 45,000 volunteers from more than 150 countries.

This project was developed in collaboration with Open Data Kosovo, Focal Labs and The Engine Room, with financial support from the Swedish Postcode Lottery. Powered by PyBossa, an open source crowdsourcing framework to analyse data that can't be processed by machines alone.

Amnesty International would like to thank all the volunteers who have helped so far with this project. We’d also like to say a very special thank you to those volunteers who helped out as moderators on the discussion forum. We couldn't have done it without you.

In particular we'd like to thank SBE, voz, Walter, JoostV, Dekker, yvesprigent, Sietse, hanny123, ellen-2016, anon2034, SarahN, Birdseye, Geromy, Max_B, Akkie, Ingewiel, Aiud, mohamedahmed, and many others who decoded hundreds of thousands of square kilometers of satellite images, and who participated in hundreds of conversations.