First STARS Mapping Party at ITC
On the 1st of May 2015, the STARS project team organized at ITC in Enschede a “Mapping party,” in which volunteers helped to digitize and classify remote sensing images. The purpose of this crowd-sourcing activity was to test whether a professional crowd can identify and digitize spatial features in support of smallholder farming in Africa (Mali case study). This was our first experiment in this domain with decently sized crowd.
The mapping party started with a training on visual image interpretation and crop land cover classification, specific to the cases of Mali and Nigeria. Then, participants digitized plot boundaries and other features, and in so doing classified very high resolution satellite images of Sub-Saharan African agricultural fields.
The participants (over 40 volunteers) were divided into 9 groups, in which each group focused on a particular class, namely: maize, sorghum, agricultural field, bare soil, single tree, group of trees, bare rock, irrigated agriculture and tree plantations.
The data gathered during the mapping party will be analyzed to determine how crowd-sourcing data and events of this type can help and improve image classification methods. We expect the results of preliminary analysis to become available in a few weeks.
The main organizer of the event was Alex Soloviov, a PhD student in the GIP department; he was helped tremendously by members of the ITC Student Association Board.