Open land cover from OpenStreetMap and remote sensing

In a recently published study (1), we produced a web based land use land cover (LULC) product based on OSM tags which are constantly updated by contributors/volunteers, and present a Remote Sensing based solution when tags were absent for a test site. We harness the combined benefit of an open source and ever-growing machine generated remote sensing time series, and thousands of people contributing land data every day. RS data were used as a source of information to extrapolate LC information provided by OSM tags into areas absent of such tags, where known areas were used as training to classify unknown areas. Three research questions were addressed:

  • What tags and relations in OSM can be used to create LULC classes from the CLC?
  • Can an open source LC product have complete coverage despite VGI’s spatial incompleteness?
  • How accurate is this LC product, and how does it compare to other existing products?

OpenStreetMap (OSM) tags were used to produce a global Open Land Cover (OLC) product with fractional data gaps available at Data gaps in the global OLC map were filled for a case study in Heidelberg, Germany using free remote sensing data, which resulted in a land cover (LC) prototype with complete coverage in this area. Sixty tags in the OSM were used to allocate a Corine Land Cover (CLC) level 2 land use classification to 91.8% of the study area, and the remaining gaps were filled with remote sensing data. For this case study, complete are coverage OLC overall accuracy was estimated 87%, which performed better than the CLC product (81% overall accuracy) of 2012. Spatial thematic overlap for the two products was 84%. OLC was in large parts found to be more detailed than CLC, particularly when LC patterns were heterogeneous, and outperformed CLC in the classification of 12 of the 14 classes. Our OLC product represented data created in different periods; 53% of the area was 2011–2016, and 46% of the area was representative of 2016–2017.

EU H2020 Project LandSense

New: (1)

Schultz, M.; Auer, A.; Voss, J. Carter,S.; Zipf, A. (2017): Open land cover from OpenStreetMap and remote sensing. International Journal of Applied Earth Observation and Geoinformation. Volume 63, December 2017, Pages 206-213.

Related earlier work:

Jokar Arsanjani, J., Mooney, P., Zipf, A., Schauss, A., (2015): Quality assessment of the contributed land use information from OpenStreetMap versus authoritative datasets. In: Jokar Arsanjani, J., Zipf, A., Mooney, P., Helbich, M., OpenStreetMap in GIScience: experiences, research, applications. ISBN:978-3-319-14279-1, PP. 37-58, Springer Press.

Jokar Arsanjani, J., Helbich, M., Bakillah, M., Hagenauer, J., & Zipf, A. (2013). Toward mapping land-use patterns from volunteered geographic information. International Journal of Geographical Information Science, 2264-2278. DOI:10.1080/13658816.2013.800871.

Dorn, H., Törnros, T. & Zipf, A. (2015): Quality Evaluation of VGI using Authoritative Data – A Comparison with Land Use Data in Southern Germany. ISPRS International Journal of Geo-Information. Vol 4(3), pp. 1657-1671, doi: 10.3390/ijgi4031657

Ballatore, A. and Zipf, A. (2015): A Conceptual Quality Framework for Volunteered Geographic Information. COSIT – CONFERENCE ON SPATIAL INFORMATION THEORY XII. October 12-16, 2015. Santa Fe, New Mexico, USA. Lecture Notes in Computer Science, pp. 1-20.

Törnros, T., Dorn, H., Hahmann, S., and Zipf, A. (2015): Uncertainties of completeness measures in OpenStreetMap – A Case Study for buildings in a medium-sized German city, ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., II-3/W5, 353-357, doi:10.5194/isprsannals-II-3-W5-353-2015.

Fan H., Zipf A., Fu Q. and Neis P. 2014. Quality assessment for building footprints data on OpenStreetMap. In: International Journal of Geographical Information Science. DOI: 10.1080/13658816.2013.867495