20 juin 2013
Ce document est lié à :
info:eu-repo/grantAgreement/EC/FP7/263479
info:eu-repo/semantics/openAccess
Alexis Comber et al., « Evaluating Global Land Cover Datasets: Comparing VGI on Cropland with Formal Data. GI_Forum 2013 – Creating the GISociety| », Elektronisches Publikationsportal der Österreichischen Akademie der Wissenschafte, ID : 10670/1.dnnomr
This paper explores the use of crowd sourced data on land cover. It compares data collected through the Geo-Wiki system describing cropland land cover with cropland data recorded in global land cover datasets. Using an African case study, local confusion matrices are calculated at discrete locations. At location the strongest correspondences are used to infer which global dataset best describes cropland at each location. Future research areas are suggested.