Low-Cost 3D Building Modelling Using National Digital Elevation Model in Urban Area of Bandung City, Indonesia

Abstract

Indonesia is one of the developing countries that is carrying out massive development in all sectors. Indonesia is currently implementing a one map policy accompanied by the massive production of 1:5,000 large-scale maps. It is hoped that the massive production of large-scale maps will assist in making plans and policies for each region in Indonesia. Indonesia also opens access to a lot of data that it can easily download. Large scale map data of 1:5000 and the National Digital Elevation Model (DEMNAS) with a spatial resolution of 8m are two data that can be accessed free of charge by the public that can be used for planning purposes whether for the benefit of individuals, groups, or countries. [1]. In preparing for regional development, 3D building models can help to improve the quality of planning. The formation of 3D buildings in an area can be used for various analyses such as urban planning, geohazard, building capacity determination, wireless network planning, 3D GPS Navigation, 3D Geographic Information System, and solar PV potential analysis [2-3]. When analysing the 2D map, the building has limited analysis, only on the horizontal axis. The limitations of this analysis make the modelling results carried out on 2D building maps often do not explain the actual situation in the field, mainly if the height of the building strongly influences the research. Therefore, a 3D building map is needed so that the analysis of the building will be closer to the actual situation. Currently, a method is required to create a 3D building map using data resources owned by Indonesia. However, in 3D mapping in Indonesia, some obstacles need a large amount of money to manufacture [4]. In this study, we will create a method for reconstructing 3D buildings using open-source data belonging to Indonesia so that the costs required are low. The data that will be used in this study are building data from the 1:5000 map [1], DEMNAS data with a spatial resolution of 8m[1], and some actual building height data obtained from Google Earth Pro[5]. DEMNAS data is a digital surface model data, so before determining the height of the building, first, create a digital terrain model from DEMNAS data. After obtaining the digital terrain model data, the next step is to determine the height of the building by subtracting the DEMNAS digital surface model from the DEMNAS digital terrain model. The next activity is to calibrate DEMNAS height data with actual building height data so that data on the height of all buildings can be obtained using DEMNAS data. DEMNAS data will be able to create 3D buildings, especially in buildings wider than the spatial resolution of DEMNAS, so that the costs incurred can be effective. Reconstructing 3D buildings using open-source data has been carried out by several previous studies. The research of Misra et al., 2018 and Kim et al., 2020 was able to identify building heights using SRTM data. However, this research has not estimated each building's footprint but has only estimated the height on raster data. In addition, Girindan et al., 2020 also carried out a height reconstruction using SRTM data to determine the building height for each building footprint. However, the resolution of the SRTM data is 30 m. In Indonesia, many buildings are small and dense, so that the spatial resolution of 30 m cannot represent the building. The novelty of this research is to reconstruct the building height on each building footprint using DEMNAS data with a resolution of 8 m to get good results. The main objective of this research is to reconstruct the height of a 3D building using open-source data for low-cost mapping. With this research, it is hoped that the planning and analysis of the building can be better because it uses 3D buildings. In addition, it is expected that the use of open-source data in reconstructing the height of 3D buildings can encourage lower mapping costs to be carried out by all regions in Indonesia. The community can use this research for 3D building analysis, starting from determining potential areas for solar PV installation, geohazard analysis, and building capacity to urban planning.