Decision-Level and Feature-Level Integration of Remote Sensing and Geospatial Big Data for Urban Land Use Mapping

Information about urban land use is important for planning and sustainable development. The emergence of geospatial big data (GBD), increased the availability remotely sensed (RS) development new methods integration to provide opportunities mapping types use. However, modes RS GBD are diverse due differences in data, study areas, classifiers, etc. In this context, aims summarize main evaluate them via a case Hangzhou, China. We first categorized into decision-level (DI) feature-level (FI) analyzed their by reviewing existing literature. two were then applied Hangzhou city, based on parcels derived from OpenStreetMap (OSM) road network, 10 m Sentinel-2A images, points interest (POI). corresponding classification results validated quantitatively qualitatively using same testing dataset. Finally, we illustrated advantages disadvantages both approaches bibliographic evidence quantitative analysis. showed that: (1) visual comparison indicates generally better performance DI-based than FI-based classification; (2) easy implement, while enables mixture features; (3) can be used together improve mapping, as they have different performances when classifying This provides an improved understanding terms strategy.