Akademik Veri Yönetim Sistemi
EARTH SCIENCE INFORMATICS, cilt.16, sa.3, ss.2559-2571, 2023 (SCI-Expanded)
In a large variety of smart city applications, the processes are settled with LOD2 (Level of Details) and the generation of the LOD2 models requires the proper generation of the roof geometries. In general, obtaining roof type information and succeeding generations of the LOD2 models requires expensive aerial surveys and time-consuming construction processes. In this study, a methodology to generate LOD2 building models using only 2D building footprints and aerial imagery is explained to overcome these challenges. The roof type information has been obtained from an aerial image that covers the entire study area using a CNN (Convolutional Neural Network) model. Then, the roof geometries have been constructed procedurally by extending and implementing a well-known Straight Skeleton (SS) algorithm for three main types of roofs: flat, gable and hipped. These constructed roof geometries have been combined with LOD1 block models generated by extruding the 2D footprints according to the height attribute. The overall accuracy of the CNN is 89.9% and the class-wise accuracies are over 84% for all classes. The least recall value is observed for the gable roof class, the enhancement options are discussed in the relevant section. The proposed methodology has been developed as a web-based solution utilizing RESTful web services with modern web technologies. In summary, the main novelty of the study is based on two contributions: using DL for gathering roof-type information without any end-user interference and the extension of the SS algorithm for the construction of roof geometries. The final product of this study is a web-based architecture for the rapid generation of the LOD2 building models.