Community detection in spatial networks: Inferring land use from a planar graph of land cover objects

Abstract

Land cover to land use Modularity a b s t r a c tThis paper applies three algorithms for detecting communities within networks. It applies them to a network of land cover objects, identified in an OBIA, in order to identify areas of homogenous land use. Previous research on land cover to land use transformations has identified the need for rules and knowledge to merge land cover objects. This research shows that Walktrap, Spinglass and Fastgreedy algorithms are able to identify land use communities but with different spatial properties. Community detection algorithms, arising from graph theory and networks science, offer methods for merging sub-objects based on the properties of the network. The use of an explicitly geographical network also identifies some limitations to network partitioning methods such as Spinglass that introduce a degree of randomness in their search for community structure. The results show such algorithms may not be suitable for analysing geographic networks whose structure reflects topological relationships between objects. The discussion identifies a number of areas for further work, including the evaluation of different null statistical models for determining the modularity of geographic networks. The findings of this research also have implications for the many activities that are considering social networks, which increasingly have a geographical component.

Urban land use extraction from very high resolution remote sensing images

“…This differs from other studies on land use, e.g. van der Kwast et al (2011) using regular grids, Comber et al (2012) using sub-graphs partitioned from a planar graph, and Hu and Wang (2013) using parcels. An existing urban land use map however may not be available for some areas.…”

“…Such a graph can represent the spatial relations between land cover objects, where a sub-graph corresponds to a basic land use unit. Interesting work on using graph-based methods for land use classification is given by Barnsley and Barr (1997); Comber et al (2012) and Walde et al (2014). The difficulty, however, is to effectively partition a graph into meaningful sub-graphs.…”

“…spatial arrangement, can be used as a good proxy to differentiate land use classes. Recent studies on urban land use extraction have also shown that the use of spatial arrangement of land cover features can improve the extraction performance (Zhan et al, 2002;van der Kwast et al, 2011;Comber et al, 2012;Vaduva et al, 2013;Walde et al, 2014;. Among those, graph-based methods have been regularly used for object-based land use extraction.…”

“…Detailed and accurate urban land use information is thus important to study the interaction between humans and the environment and to support sustainable urban development. Increasing attention has been paid to the extraction of urban land use from remote sensing images (Barnsley and Barr, 1997;Herold et al, 2003;van der Kwast et al, 2011;Comber et al, 2012;Voltersen et al, 2014;Li et al, 2016b). Particularly, satellites like GeoEye, Wordview, and Pléiades are providing large volumes of Very High Resolution (VHR) images, in single, stereo or tri-stereo modes, allowing for the extraction of land use information with fine spatial detail (Pesaresi et al, 2013).…”

“…A graph-based method starts from the construction of a planar graph, in which the contextual information can be calculated within a sub-graph. Interesting work on using graph-based methods for land use classification is given by Barnsley and Barr (1997);Comber et al (2012); Walde et al (2014). Such a graph can represent the spatial relations between land cover objects.…”