Creating a workflow of 3D building data in a municipality context

anglais. 3D city models are valuable tools for visualization and analytical purposes in the urban environment. A popular format for city models is CityGML, an open data model that can be used for storage and exchange of 3D city models. However, using city models based on CityGML has appeared to be a challenge as there is a limited number of software applications that are able to read them. Complexity in CityGML data sets is also a problem as the format allows for many variations of geometry handling. Also, 3D models in general are often created individually by producers and exchanging them between platforms and users is an issue. As a response to the challenges of 3D city models not being created in a common way, national standards have been created to satisfy internal needs and making 3D city data more software compatible. Sweden has also recognized these needs and is in the process of creating a national standard for 3D buildings called NS (National Specification) building. The data in NS building will be represented as NS LOD (National Specification Level of Detail), a lightweight 3D data type that focuses on storing simple geometries that can be transformed to CityGML. Another 3D city format is CityJSON. CityJSON is a format that has been adapted to the data model of CityGML. The creators of CityJSON have attempted to reduce the complexity of 3D city models to allow for better software development and usage on the web. Because of the complexity 3D city models, robust database systems that can implement all the features of a 3D city model are required. 3D City Database (3DCityDB) is a geodatabase that has been created to properly transform and store CityGML datasets. This thesis has attempted to verify that NS LOD data can be used in NS building and be transformed to CityGML. This master thesis is also involved in a project (3CIM) between the three largest municipalities in Sweden (Stockholm, Gothenburg and Malmö) and Lund University. The objective of the project is to research how a 3D building model can be enhanced by making it interoperable to an information model. This will in return allow it to be used for analytical purposes. This thesis has created a 3D building environment based on 3DCityDB. The environment can import CityGML data (transformed from NS LOD) and export to CityGML and CityJSON. The environment can be further developed and used to control the information flow during import and export. The results of this thesis show that NS LOD data can be used in the national Swedish specification NS building to be transformed to CityGML. However, further development of the specification should focus on errors that are created during transformation from NS LOD to CityGML as they can hinder the usage of the transformed data. This thesis has also successfully created open-source processes that can be used to import CityGML data to 3DCityDB and export to CityGML and CityJSON. Still, if a user is to adapt the processes much work must be put in to ensure that data sets exceeding this master thesis can be handled.

anglais. 3D city models are valuable tools for planning and expanding new buildings in cities. A popular data format for representing these models is CityGML. However, using CityGML as a data format has appeared to be a challenge as there is a limited number of computer programs that are able to read them. Also, 3D city models in general are often created individually by producers and trading them between computer programs and users is an issue. As a response to the challenges of 3D city models not being created in a common way, national guidelines have been created to satisfy internal needs and making 3D city data more computer friendly. Sweden has also recognized these needs and is in the process of creating national guidelines for 3D buildings called NS (National Specification) building. The data in NS building will be represented as simple geometries called NS LOD (National Specification Level of Detail) that can be transformed to CityGML. Another data format for 3D city models is CityJSON. CityJSON is a format that has been based on the data format CityGML. The creators of CityJSON have attempted to reduce the complexity of 3D city models to allow for better software development and usage on the web. Because of the complexity 3D city models, robust systems for storing them that also can implement all the features of a 3D city model are required. 3D City Database (3DCityDB) is a database that has been created to properly transform and store CityGML datasets. This thesis has attempted to verify that data from the Swedish specification NS building can be transformed to CityGML. This master thesis is also involved in a project (3CIM) between the three largest municipalities in Sweden (Stockholm, Gothenburg and Malmö) and Lund University. The objective of the project is to research how a 3D building model can be made better for environmental purposes. This thesis has created a 3D building environment based on 3DCityDB. The environment can import CityGML data (transformed data from NS building) and export to CityGML and CityJSON. The environment can be further developed and used to control the data during import and export. The results of this thesis show that data from NS building can be used in the national Swedish specification to be transformed to CityGML. However, further development of the specification should focus on errors that are created during transformation from NS building to CityGML as they can hinder the usage of the transformed data. This thesis has also successfully created a system that can be used to import CityGML data to 3DCityDB and export to CityGML and CityJSON. Still, if a user is to adapt the processes much work must be put in to ensure that data sets exceeding this master thesis can be handled.