基于无人机航测数据快速构建城市建筑物三维模型

Rapid Construction of 3D Models of Urban Buildings Based onUAV Photogrammetric Data

传统的建筑物手工建模周期长、成本高,损失了地理空间信息。倾斜摄影以及三维激光扫描技术虽然能快速构建城市建筑物三维模型,且地理空间信息丰富,但是资源占用高、单体化程度低,不易进行编辑和管理。对此,研究探讨了一种基于无人机航测数据,通过U-Net语义分割网络对正射影像进行建筑物提取,然后利用DSM获取建筑物高度信息,最后编写CGA规则批量构建城市建筑物三维模型的建模流程,并通过实例分析验证了该建模流程的有效性。

The traditional manual modeling of buildings has a long cycle,high cost,and loss of geospatial information.Although ob-lique photography and 3D laser scanning technology can quickly build 3D models of urban buildings and have rich geospatial informa-tion,they are difficult to edit and manage due to high resource occupation and low degree of individualization.In this regard,based on UAV photogrammetric data,this paper presents a new method to extract buildings from orthophoto by using U-Net semantic segmenta-tion network,building height information is obtained by DSM,and CGA rules are compiled to construct 3D models of urban buildings in batch.The effectiveness of the modeling process is verified by example analysis.