摘要
针对存在涵洞、桥梁的复杂陆路交通场景,交通设施遮挡造成的观测及建模困难,在使用空天地协同观测和外内业技术联动的基础上,提出基于同名几何一致性的多源融合建模方法,结合点云配准与多视角投影实现激光点云与影像间的匹配,通过同名一致的光束平差约束融合机载LiDAR数据、SLAM数据、地面三维激光数据、全球导航卫星系统定位数据、无人机倾斜影像和地面相机影像,最终构建准确、完整、稳健的复杂交通场景三维模型。通过沪渝高速恩施段的技术验证,证明本方法的建模成果在完整度及细节纹理方面优于两种常见方法,无人机倾斜摄影测量配合地面三维激光扫描的采集策略最便捷高效,若配合机载LiDAR或地面相机则建模效果更佳。成果为交通领域的勘测、设计、施工、运营等提供有力支持。
Facing the complex land traffic scene with culverts and bridges,aiming at the difficulties of observation and modeling caused by traffic facilities occlusion,a multi-source fusion modeling method based on homonymous geometric consistency is proposed on the basis of using space-air-ground collaborative observation and field and office technology linkage.In this study,we combining point cloud alignment and multi-view projection to realize the matching between laser point cloud and image,and fuse airborne light detection and ranging(LiDAR)data,simultaneous localization and mapping(SLAM)data,ground-based three-dimensional(3D)laser data,global navigation satellite system(GNSS)positioning data,unmanned aerial vehicle tilted image,and ground-based camera image through the homonymous beam leveling constraints,and finally constructing an accurate,complete and robust 3D model of the complex traffic scene.Through the technical validation of the Enshi section of the Huyu Expressway,it is proved that the modeling results of this method are superior to the two common methods in terms of integrity and detail texture.At the same time,the acquisition strategy of unmanned aerial vehicle oblique photogrammetry combined with ground three-dimensional laser scanning is the most convenient and efficient.If it is combined with Airborne LiDAR or ground camera,the modeling effect is better.The results provide strong support for survey,design,construction and operation in the field of transportation.
作者
丛铭
展熠
陈斯亮
崔建军
费亮
彭琛
CONG Ming;ZHAN Yi;CHEN Siliang;CUI Jianjun;FEI Liang;PENG Often(School of Geological Engineering and Mapping,Chang’an University,Xi’an 710054,China;School of Architecture,Chang’an University,Xi’an 710061,China;China Railway SIYUAN Survey and Design Group Co.,Ltd.,Wuhan 430063,China)
出处
《测绘科学》
CSCD
北大核心
2023年第11期117-124,共8页
Science of Surveying and Mapping
基金
国家重点研发计划项目(2021YFB2600400)
陕西省教育厅服务地方专项计划项目(23JE002)
关键词
复杂交通场景
多源融合建模
空天地协同
三维实景模型
几何一致性
complex traffic scenarios
multi-source fusion modeling
space-air-ground collaborative
3D realistic model
geometric consistency
