Abstract
Abstract. This paper describes the development of 3D database for Istanbul 3D city models. The schema is based on object-relational technology and also called spatially-enhanced relational database management system (SRDBMS). The 3D city models were generated based on LiDAR points cloud with other several typical GIS datasets like terrain, orthophoto, Point of Interests (POIs), and other attribute data. The database is based on PostGIS schema and CityGML schema (3DCityDB). Two major datasets, namely, terrain data (with several formats), and 3D city models were populated in the database. Terrain and attributes data retrieval are based on Web Feature Service (WFS) whereas 3D models were visualized via 3D Tiles format in Cesium platform. Two issues were also highlighted in the paper with respect to 3D attributes linkages and 3D complex objects.
Highlights
The Istanbul Greater Municipality has developed 3D city models by using 3D point clouds data, 2D digital map, and 3D laser scanned datasets, which are processed by some commercial software and in-house tools
Efficiently visualising 3D geometries and semantic information stored in City Geography Markup Language (CityGML) is complex (Arroyo et al, 2018)
This paper is to develop a database for spatial objects within 3D city model of Istanbul that based on LiDAR point clouds
Summary
The Istanbul Greater Municipality has developed 3D city models by using 3D point clouds data, 2D digital map, and 3D laser scanned datasets, which are processed by some commercial software and in-house tools. After quality check for the classified point cloud, the 3D building cubes (approximately 1.5 million) were obtained automatically using TerraModeller software as a 3D CAD files with the contributions of ortophotos and building footprints available in the 1/1000 scale base maps while some specific structures such as mosques, and bridges which have complex roof type and fancy structures were manually generated. These files were converted to the CityGML format using an FME Workbench. Efficiently visualising 3D geometries and semantic information stored in CityGML is complex (Arroyo et al, 2018)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
