Drone Imagery for OpenStreetMap Sidewalk Data Enrichment

Abstract

Several Location Based Services (LBS) are developed from open sources. The OpenStreetMap (OSM) is most stable and used one and, besides being open data source, has a collaborative characteristic, known as Volunteered Geographic Information (VGI). In this context, some routing services have emerged supported by LBS and VGI, and personalization has become a trendy. However, customization as wheelchair and pedestrian route planning requires specific and detailed information about sidewalk. Currently, most sidewalk in OSM are nearly missing or tags absent. Besides that, there is a lack of standardization for mapping this data. In a first moment, this study was developed aiming to fill this gap by using remote sensing, specifically Drone. The procedure for data enrichment proposed consists of: (i) data completeness search; (ii) data acquirement and processing, and (iii) data availability. The initial results showed that the pilot area has a high degree of incompleteness and, once drone imagery was provided, sidewalk could be mapped based on the resulting orthomosaic. The next step will be a study targeting sidewalk tags and features pattern definition. At the end of this study, besides the exploration of the possibility of high quality imagery sharing, sidewalk standardized mapping may improve location based services data for engines, such as wheelchair and pedestrian routing.