Anzeige
Erschienen in:
2014 | OriginalPaper | Buchkapitel
Estimation of Building Types on OpenStreetMap Based on Urban Morphology Analysis
verfasst von : Hongchao Fan, Alexander Zipf, Qing Fu
Erschienen in: Connecting a Digital Europe Through Location and Place
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
Buildings are man-made structures and serve several needs of society. Hence, they have a significant socio-economic relevance. From this point of view, building types should be strongly correlated to the shape and sized of their footprints on the one hand. On the other hand, building types are very impacted by the contextual configuration among building footprints. Based on this hypothesis, a novel approach is introduced to estimate building types of building footprints data on OpenStreetMap. The proposed approach has been tested for the building footprints data on OSM in Heidelberg, Germany. An overall accuracy of 85.77 % can be achieved. Residential buildings can be labeled with accuracy of more than 90 %. Besides, the proposed approach can distinguish industrial buildings and accessory buildings for storage with high accuracies. However, public buildings and commercial buildings are difficult to be estimated, since their footprints reveal a large diversity in shape and size.