Knowledge graph assists in the automatic construction of historical building information models
This paper proposes an automatic generation method for historical building information models based on knowledge graph construction and inference technology,which addresses the issues of missing,inconsistent density,and abnormal data in the application of LiDAR and oblique photography technology in the digital protection process of historical buildings. Additionally,the 3D point cloud data obtained through point cloud fitting cannot be disassembled. Firstly,according to the rules of architectural construction techniques,construct a knowledge graph of the data structures and spatial topological relationships of each component of the building itself and between them. Then,combined with known point cloud data,scientifically infer missing component information. Finally,complete the component parameters and complete the automated model construction. Compared to the current mainstream reverse manual modeling methods based on point cloud data,this method has faster modeling speed,detachable model components,and more complete parameter information. The experimental results of the structural carpentry of the Sheng Mu Miao at Shanxi province Jinci Temple show that the parameterized model obtained by this method can comprehensively and realistically display the structural information and traditional construction techniques of the building,providing an important technical means for the digital protection of historical buildings.
knowledge graphknowledge reasoningontologystructural carpentry of the Sheng Mu Miao at Jinci TempleHBIM
万方数据
维普
