The content of this article comes from the "Surveying and Mapping Bulletin" No. 5, 2024, drawing review number: GS Jing (2024) No. 0777
Automatic generation method of building LOD2 model based on energy function primitive extraction
Wang Lin1,2, Chen Jianlong3, Liu Zhaoliang4,5, Liu Wenxuan4,5
1. Hefei Surveying and Mapping Design and Research Institute Co., Ltd., Hefei 230061, Anhui, China; 2. Hefei Engineering Research Center for Geographic Information and Smart City, Hefei 230061, Anhui, China; 3. Hefei Chaohu Space Surveying and Mapping Technology Co., Ltd., Hefei 230061, Anhui, China; 4. Wuhan Dashi Wisdom Technology Co., Ltd., Wuhan 430223, Hubei, China; 5. Engineering Technology Innovation Center for Real 3D Construction and Urban Fine Governance, Ministry of Natural Resources, Wuhan, 430223, Hubei, China
Keywords: Level of Detail Model, Building Reconstruction, Area Segmentation, Automatic Generation, Primitive Extraction
Citation format: WANG Lin, CHEN Jianlong, LIU Zhaoliang, et al. Automatic Generation Method of Building LOD2 Model Based on Energy Function Primitive Extraction[J]. Bulletin of Surveying and Mapping, 2024(5): 103-107,114.doi: 10.13474/j.cnki.11-2246.2024.0518
Abstract:In recent years, with the rapid development of real 3D China and digital China, 3D model reconstruction of buildings has become an important part of smart city construction. This paper proposes a fully automatic building LOD2 model generation framework, which aims to solve the problem of building LOD2 model generation in city-level real-world 3D scenes. In order to automatically extract the building objects in the large scene, firstly, the scene orthophoto was used to obtain the boundary information of the building through image segmentation. Then, the improved enhanced planar area segmentation method is used to extract the high-quality segmentation plane, and the hybrid linear model is used to optimize the segmentation results, which can better deal with the problems of local damage and reconstruction errors of urban buildings. The experimental results show that the building LOD2 model generated by the proposed algorithm is of higher quality and efficiency, which can cope with the complexity of the urban building model and obtain more robust reconstruction results.
About authorAbout author:WANG Lin (1972—), male, senior engineer, mainly engaged in surveying, mapping and geographic information production, technical research and management. E-mail:[email protected]
First instance: Yang Ruifang review: Song Qifan
Final Judge: Jin Jun
information