Malawi’s Dzaleka refugee camp was originally established in 1994 to shelter approximately 9,000 refugees fleeing violence and war in Burundi, Rwanda and the Democratic Republic of the Congo. Over the next twenty years, the camp’s population swelled as people continued to flee long-standing conflict in the region. Today, approximately 30,000 refugees live in the deeply congested camp, making it more difficult for refugees to access water, medical treatment and electricity.
A new approach to refugee camp planning and mapping is needed for both existing camps like Dzaleka and those that will be built in response to the next crisis.
To meet this growing need, The Hive, USA for UNHCR’s (United Nations High Commission for Refugees) innovation lab, is using satellite imagery to map and analyze Dzaleka camp and 115 other refugee camps to assist ground site planning teams in their decongestion efforts.
As the number of forcibly displaced people increases, construction of new refugee camps and the expansion of existing ones is vital. With this comes a need to ensure camps are being built and maintained efficiently and sustainably with respect to location of shelters, water sources and other variables.
Camp mapping is essential but labor intensive as field staff typically survey the camps on the ground and upload the data manually. By harnessing the power of satellite imagery from Maxar Technologies’ DigitalGlobe with machine learning, this mapping project aims to reduce the amount of labor and staff time needed to monitor camp growth and utilization and allow for UNHCR to expand other program areas that assist refugees.
Out of the 202 UNHCR refugee camps, 116 will be mapped using Tomnod and machine learning while following similar shelter mapping methodology currently used by UNOSAT and UNHCR.
Using DigitalGlobe satellite imagery, individuals can tag shelters in refugee camps. There are four types of structures that need to be identified within each refugee camp: tents, improvised, semi-permanent and administrative shelters.
Once the data is tagged, the Hive will work with Stanford University Sustainability and Artificial Intelligence Lab to automatically identify and count tent features from satellite imagery of other UNHCR refugee camps using convolutional neural network models.
Help us automate the process of mapping growing refugee camps. By tagging the shelters in refugee camps, you will assist UNHCR in their mission to safeguard the well-being of refugees.
To participate in helping the Hive to identify structures within refugee camps, please join the Tomnod campaign.