Using OpenStreetMap to Create Land Use and Land Cover Maps

Abstract

OpenStreetMap (OSM) is a bottom up community-driven initiative to create a global map of the world. Yet the application of OSM to land use and land cover (LULC) mapping is still largely unexploited due to problems with inconsistencies in the data and harmonization of LULC nomenclatures with OSM. This chapter outlines an automated methodology for creating LULC maps using the nomenclature of two European LULC products: the Urban Atlas (UA) and CORINE Land Cover (CLC). The method is applied to two regions in London and Paris. The results show that LULC maps with a level of detail similar to UA can be obtained for the urban regions, but that OSM has limitations for conversion into the more detailed non-urban classes of the CLC nomenclature. Future work will concentrate on developing additional rules to improve the accuracy of the transformation and building an online system for processing the data.

Automatic conversion of OSM data into LULC maps: comparing FOSS4G based approaches towards an enhanced performance

“…Arsanjani et al [3] tested the conversion of OSM data into a LULC map using the Urban Atlas nomenclature with encouraging results. Fonte et al [8,10] proposed an automated methodology to convert OSM data into LULC maps and Fonte et al [9] showed the potential of merging LULC data extracted from OSM with existing LULC products, namely the GlobeLand30. The use of OSM extracted LULC data to validate LULC maps was also tested with promising results, as for Urban Atlas level 1 nomenclature the accuracy indices obtained using OSM data (where available) to obtain the reference class were not very different from the ones obtained when the reference class was always obtained by photo interpretation [7].…”

“…Shultz et al [29] produced a global Land Cover product using OSM data, and used the available data to train a classifier that was used to classify satellite imagery in order to generate data for the regions when OSM data is not available. The encouraging results obtained in the previous efforts motivated the creation of OSM2LULCa FOSS4G aiming to automatically convert OSM data into LULC maps [8][9][10]. OSM2LULC 2 is part of the GeoData Algorithms for Spatial Problems (GASP) Python package, 3 whose ultimate aim is to provide various tools to extract, to convert, to analyse, and to validate geospatial information.…”

“…OSM2LULC has already been applied to different case study regions [8][9][10]. These case studies allowed the identification of the main problems of the algorithm.…”

Joint Learning from Earth Observation and OpenStreetMap Data to Get Faster Better Semantic Maps

“…Although the coverage and the quality of the annotations from open GIS vary a lot depending on the users' knowledge and number of contributors, this data may contain relevant information for mapping specific areas and classes. A deterministic framework to create land use and land cover maps from crowdsourced maps such as OSM data was proposed in [10]. Machine learning tools (a random forest variant) also allow coupling remote sensing and volunteered geographic information (VGI) to predict natural hazard exposure [11] and local climate zones [8], while active deep learning helps finding unlabeled objects in OSM [5].…”

Rapid Detection of Land Cover Changes Using Crowdsourced Geographic Information: A Case Study of Beijing, China

“…This inspired further investigation to employ CGI for changes detection to complement the shortcoming of remote sensing. A recent research study proposed a method of generating up-to-date maps using textual information from OSM [17,18]. Since these kinds of data were provided by volunteers, textual information is often insufficient and incomplete for rapid LCC detection.…”