Identifying urban land use social functional units: a case study using OSM data

Urban land use is always the central focus of urban planners and policymakers when pursuing urban sustainable development. Currently, many studies have explored urban land use using satellite data, but few of them can identify the social function of land use through these data. In this research, OpenStreetMap (OSM) data are used to extract urban land use social functional units. Firstly, the OSM data are transferred to actual land use types based on the established mapping rules between OSM data and the urban land use classification standards. Then, the improved DBSCAN algorithm is applied with OSM data of the study area after estimating the parameters of the neighborhood radius (Eps) and the minimum polygons (MinPolys) based on the polygon density. With the two estimated parameters, the improved DBSCAN algorithm is used to cluster residential, commercial and public service land use social functional polygons of the OSM data. Following that, the clusters of the three land use types are processed into land use social functional units. The results show that OSM data perform well in determining these three land use social functional clusters using the improved DBSCAN algorithm, and the social functional units display obvious spatial features.