OpenStreetMap (OSM) is currently the largest openly licensed collection of geospatial data, widely used in many projects as an alternative to or integrated with authoritative data. One of the main criticisms against this dataset is that, being a collaborative product created mainly by citizens without formal qualifications, its quality has not been assessed and therefore its usage can be questioned for some applications. This paper provides a map matching method to check the spatial accuracy of the building footprint layer, based on a comparison with a reference dataset. Moreover, from the map matching and a similarity check, buildings can be detected and therefore an index of completeness can also be computed. This process has been applied in Lombardy, a region in Northern Italy, covering an area of 23,900 km2
and comprising respectively about 1 million buildings in OSM and 2.8 million buildings in the authoritative dataset. The results of the comparison show that the positional accuracy of the OSM buildings is at least compatible with the quality of the reference dataset at the scale of 1:5000 since the average deviation, with respect to the authoritative map, is below the expected tolerance of 3 m. The analysis of completeness, given in terms of the number of buildings appearing in the authoritative dataset and not present in OSM, shows an average percentage in the whole region equal to 57%. However, worth noting that the opposite, namely the number of buildings in OSM and not in the reference dataset, is not zero, but corresponds to 9%. The OSM building map can therefore be considered to be a valid base map for direct use (territorial frameworks, map navigation, urban analysis, etc.) and for derived use (background for the production of thematic maps) in all those cases where an accuracy corresponding to 1:5000 is required. Moreover it could be used for integrating the authoritative map at this scale (or smaller) where it is not complete and a rigorous quality certification in terms of metric precision is not required.
This is an open access article distributed under the Creative Commons Attribution License
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.