基于多源数据融合的城市建筑物三维可视化研究  

Research on 3D Visualisation of Urban Buildings with Multi-Source Data Fusion

作　　者：吴健 廖燕丹 李登朝 陈尔东 冯晓敏 WU Jian;LIAO Yan-dan;LI Deng-chao;CHEN Er-dong;FENG Xiao-min(The First Geological Brigade of Hubei Geological Bureau,Huangshi Hubei 435000,China;Huangshi Natural Resources and Planning Bureau,Huangshi Hubei 435000,China)

机构地区：[1]湖北省地质局第一地质大队,湖北黄石435000 [2]黄石市自然资源和规划局,湖北黄石435000

出　　处：《地矿测绘》2024年第2期7-12,共6页Surveying and Mapping of Geology and Mineral Resources

基　　金：湖北省自然资源科技项目“基于多源数据融合的三维可视化应用研究”(ZRZY2023KJ34)。

摘　　要：针对倾斜摄影测量技术在建筑物模型生成中所存在的盲区、数据缺失和质量不达标等问题,介绍了融合倾斜摄影测量与三维激光扫描技术来构建精细三维模型的方法。采用无人机和三维激光扫描仪作为实验设备,获取到的数据经过配准、去噪、几何建模及纹理映射等处理,并运用ICP算法完成了点云的精细匹配,实现了多源点云数据的融合。与传统的单一建模方法相比,从模型的精度和纹理结构方面全面评估了融合模型的质量。结果显示,融合模型呈现出清晰的纹理和完整的几何结构,模型误差明显低于倾斜摄影模型,且精度接近三维激光扫描模型。该方法最终实现了对城市复杂建筑物的三维模型重建,可为同类项目提供借鉴。In view of the issues of blind spots,data loss,and substandard quality in building model generation with the oblique photogrammetry technology,this article introduces a method of integrating oblique photogrammetry with 3D laser scanning technology to create detailed 3D models.The UAV and 3D laser scanner are used as experimental equipment,and the acquired data are processed by registration,denoising,geometric modeling and texture mapping.Fine point cloud matching is achieved by use of the ICP algorithm.Thus the fusion of multi-source point cloud data are realized.Compared with traditional single modeling methods,the quality of the fusion model is comprehensively evaluated in terms of model accuracy and texture structure.The results show that the fusion model presents clear textures and complete geometric structures,with significantly lower model errors than the oblique photography model,and accuracy close to that of the 3D laser scanning model.The research method adopted in this article ultimately achieves the 3D model reconstruction of complex urban buildings,which can provide reference for similar projects.

关 键 词：倾斜摄影测量 三维激光扫描 ICP算法 融合建模 数字城市 

分 类 号：P237[天文地球—摄影测量与遥感]