Visualizing Multi-dimensional Geospatial Data On The Cesium Platform

As the concept of smart cities and digital China grows and matures,spatial data have attracted widespread attention from various industries.In the era of big data,the demand for spatial data in various industries is increasing.Many interdisciplinary staff and non-professionals are also attracted to this field.Not only do they need a common way of storing data,they also put forward requirements for data sharing and visualization.First of all,the sharing of data and the public accessibility need to be improved so that more non-professionals can understand and use spatial data.Second,the dimensional scope of geographic data visualization needs to be extended.In addition to near-surface geospatial data,geospatial data in underground space and indoor space are also required by various industries.So as the spatio-temporal data.Scientific visualization of data can increase the level of public accessibility to the data.Therefore,there is an urgent need for a platform that is public-oriented,with strong capacities of sharing and visualizing spatial data in multi-space and multi-dimension.Virtual globe is a three-dimensional visualization platform for spatial data.It can display spatial data in a friendly and intuitive manner at different scales.In the era of Internet 2.0,the virtual globe has gradually gone from a client-side application to a web application,and its sharing and openness have also been continuously improved.In this paper,a spatial data visualization platform was developed with Web-based virtual globe engine Cesium,which expands the scope of visualization from near-surface to underground space and indoor space,as well as enrich the visualization of time dimension.This study builds a spatial relational database structure based on CityGML and 3D City Datebase,and stores multiple source spatial data formats in a unified manner.It includes unified workflows for Open StreetMap,CityGML,IFC,underground drilling,subterranean lithosphere,and time-space trajectory data;using JavaScript and python scripts,combined with the open source 3D modeling software Blender to achieve 3D spatial data slicing,simplification and format conversion.And finally the 3D models with a huge amount of geometry information are exported to a 3D Tiles format with a HLOD structure.The 3D Tiles format is used for the tile-based loading,out-of-core rendering and perspective-based visualization.Then python scripts are adopted to process and export trajectory data into CZML format files for description and dynamic visualization in the Cesium platform.Interactive events and visualization effects are improved based on the secondary development of the Cesium virtual globe,to achieve seamless underground and indoor roaming.This paper proposes a method to create a multi-spatial multi-dimensional GIS visualization platform based on Cesium,which provides a reasonable solution for the lack of visualization of underground space,indoor structure,and multi-dimensional space-time data in traditional geographic information platforms.This method can broaden the scope of the geographic information platform and enrich the diversity of spatial data.The platform has the ability to load massive amounts of multi-dimensional spatial data,as well as open and shared,mass-oriented platform features,and provides a good carrier for the construction of smart cities.