An automated methodology for converting OSM data into a Land Use/Cover map

Abstract

Land Use/Land Cover Maps (LULCM), fundamental for many areas of application, are usually generated through the classification of satellite imagery. However, their creation is time consuming and therefore updated LULCM are seldom available. The OpenStreetMap (OSM) collaborative project collects a rich set of vector data provided by volunteers at a global scale. It has already been shown that OSM data may be converted into LULCM, but data quality issues in OSM raise some challenges for this conversion, such as overlapping features that should be assigned to different classes. Thus, the creation of LULCM using OSM requires a solution for handling these inconsistencies. In this article an automated methodology is proposed using rules of decision and spatial analysis in a GIS environment to convert OSM features into LULCM, which automatically solves the inconsistencies mentioned above. The methodology is applied to two areas in Europe and the results are compared to available LULCM.