Developing a VGI method for 3D city modeling based on CityGML and Open Data Kit

Abstract

Due to technological developments, 3D city models have become valuable in various domains such as emergency services, facilities management, tourism and entertainment along with several applications such as the estimation of solar irradiation, routing, lighting simulations, etc. However, many cities in the world, especially in developing countries, still suffer from lack of 3D city models. It seems that the main reason for this deficiency is that 3D city models are expensive. Furthermore, acquiring semantic and thematic data as an indispensable part of 3D city models is an exhausting and time-consuming task. Nowadays, a geospatial data collecting technique, which is an inexpensive and promptness solution, has been developed. This technique is based on crowdsourcing concept and is recognized as Volunteered Geographic Information (VGI). In this paper, we have used VGI as a free and promptness technique for data gathering to solve the abovementioned problems in the Shahid Rajaee Teacher Training University as the study area. We gathered the minimum required data for creating a 3D city model based on the CityGML standard as the most well-known and acceptable standard by VGI. Also, 3DcityDB that supports CityGML was used for data storage task. In order to collect the required data, an Android mobile application was developed based on Open Data Kit (ODK). In this study, the volunteers were asked to provide their estimations of the heights of buildings as well as some other spatial and attribute data. Consequently, a 3D city model was produced based on the CityGML standard that achieved LOD 1 and 2. For validation, the heights of buildings obtained from VGI were compared to the accurately measured heights. The calculated RMSE for this comparison was 1.33 meter, proving the abilities of VGI in collecting reliable datasets.