倾斜摄影与地面激光点云融合精细化建模研究
|
摘要
倾斜摄影是一种高效、低成本的三维建模技术,被越来越广泛地应用于城市三维重建当中。由于其通过空中作业获取地面影像数据,不可避免地会产生局部遮挡现象,使得在空三加密过程中出现点云漏洞,导致三维模型在相应部位出现拉花、变形现象。为解决倾斜摄影技术的遮挡问题,从而实现模型的精细化,提出倾斜摄影与地面激光点云融合建模的方法。研究表明:①基于倾斜摄影的三维模型在有遮挡区域会出现明显的扭曲变形;②地面激光点云可以修补缺失的影像点云;③倾斜摄影与地面激光点云融合建模可以明显地提高模型的精细度。
Tilt photography is a high efficient and low-cost 3D modeling technology that is increasingly used in urban 3D reconstruction.Because it acquires ground image data through air operations,local occlusion phenomenon will inevitably occur,which causes a point cloud vulnerabilities in the process of air three encryption,resulting in three-dimensional model appearing in the corresponding parts.In order to solve the occlusion problem of tilt photography,and to achieve the refinement of the model,a method of fusion modeling between tilt photography and ground laser point cloud is proposed.The research shows that:①The 3D model based on tilt photography will have obvious distortion in the occluded area.②The ground laser point cloud can repair the missing image point cloud.③The tilt photography and ground laser point cloud fusion modeling can significantly improve the model.accuracy.
Tilt photography is a high efficient and low-cost 3D modeling technology that is increasingly used in urban 3D reconstruction.Because it acquires ground image data through air operations,local occlusion phenomenon will inevitably occur,which causes a point cloud vulnerabilities in the process of air three encryption,resulting in three-dimensional model appearing in the corresponding parts.In order to solve the occlusion problem of tilt photography,and to achieve the refinement of the model,a method of fusion modeling between tilt photography and ground laser point cloud is proposed.The research shows that:①The 3D model based on tilt photography will have obvious distortion in the occluded area.②The ground laser point cloud can repair the missing image point cloud.③The tilt photography and ground laser point cloud fusion modeling can significantly improve the model.accuracy.
引文
[1] 王伟,黄雯雯,镇姣.Pictometry倾斜摄影技术及其在三维城市建模中的应用[J].测绘与空间地理信息,2011,34(3):181-183.
[2] 邓非,张祖勋,张剑清.一种激光扫描数据与数码相片的配准方法[J].武汉大学学报(信息科学版) ,2007(4):290-292.
[3] 邓非,杨海关,李丽英.基于互信息的LIDAR与光学影像配准方法[J].测绘科学,2009(6):51-52.
[4] 谭仁春,李鹏鹏,文琳,等.无人机倾斜摄影的城市三维建模方法优化[J].测绘通报,2016(11):39-42.
[5] 严剑锋.地面LIDAR点云数据配准与影像融合方法研究[D].徐州:中国矿业大学,2014.
[6] 李怡静,胡翔云,张剑清,等.影像与LiDAR数据信息融合复杂场景下的道路自动提取[J].测绘学报,2012(6):870-876.
[7] 赵煦.基于地面激光扫描点云数据的三维重建方法研究[D].武汉:武汉大学,2010.
[8] 张帆,黄先锋,李德仁.激光扫描与光学影像数据配准的研究进展[J].测绘通报,2008(2):7-10.
[9] 丁延辉.地面三维激光数据配准研究[J].测绘通报,2009(2):57-59.
[10] 张剑清,潘励,王树根.摄影测量学[M].武汉:武汉大学出版社,2002.
[11] 甘晓川,赫明判,叶孝佑.数字近景摄影测量中相机内参数校准[J].计量学报,2012,33(4):391-394.
[12] 李鸿祥.三维城市规划辅助决策支持系统的决策与实现[D].厦门:厦门大学,2013.
[13] 陈天凡,高诚辉,何炳蔚.基于表面贴合的不同视点传感器点云拼合方法研究[J].中国机械工程,2012(24):2968-2971.
[14] 鲁艺玲.POS辅助空中三角测量中的定权研究[D].成都:西南交通大学,2013.
[15] 范冲,王学.三维城市建筑物的纹理映射综述[J].测绘与空间地理信息,2014,37(7):1-4.
[2] 邓非,张祖勋,张剑清.一种激光扫描数据与数码相片的配准方法[J].武汉大学学报(信息科学版) ,2007(4):290-292.
[3] 邓非,杨海关,李丽英.基于互信息的LIDAR与光学影像配准方法[J].测绘科学,2009(6):51-52.
[4] 谭仁春,李鹏鹏,文琳,等.无人机倾斜摄影的城市三维建模方法优化[J].测绘通报,2016(11):39-42.
[5] 严剑锋.地面LIDAR点云数据配准与影像融合方法研究[D].徐州:中国矿业大学,2014.
[6] 李怡静,胡翔云,张剑清,等.影像与LiDAR数据信息融合复杂场景下的道路自动提取[J].测绘学报,2012(6):870-876.
[7] 赵煦.基于地面激光扫描点云数据的三维重建方法研究[D].武汉:武汉大学,2010.
[8] 张帆,黄先锋,李德仁.激光扫描与光学影像数据配准的研究进展[J].测绘通报,2008(2):7-10.
[9] 丁延辉.地面三维激光数据配准研究[J].测绘通报,2009(2):57-59.
[10] 张剑清,潘励,王树根.摄影测量学[M].武汉:武汉大学出版社,2002.
[11] 甘晓川,赫明判,叶孝佑.数字近景摄影测量中相机内参数校准[J].计量学报,2012,33(4):391-394.
[12] 李鸿祥.三维城市规划辅助决策支持系统的决策与实现[D].厦门:厦门大学,2013.
[13] 陈天凡,高诚辉,何炳蔚.基于表面贴合的不同视点传感器点云拼合方法研究[J].中国机械工程,2012(24):2968-2971.
[14] 鲁艺玲.POS辅助空中三角测量中的定权研究[D].成都:西南交通大学,2013.
[15] 范冲,王学.三维城市建筑物的纹理映射综述[J].测绘与空间地理信息,2014,37(7):1-4.