倾斜摄影技术是目前三维建模的重要方式,具有效率高、成本低且灵活快捷的优点,然而,其也同样存在模型结构不完整、表面拉花的问题,这种情况在构建筑物部分尤为严重.为解决倾斜摄影建模精度不足问题,提出一种视频帧影像与倾斜摄影融合精细建模方法.首先,通过无人机采集测区的倾斜影像,并通过内业处理生成倾斜点云数据;其次,使用运动相机获取测区构建筑物的视频帧影像,利用增量式运动恢复结构(Structure From Motion,SFM)算法和多视图立体视觉(Multi-View Stereo,MVS)算法进行三维重构,生成点云数据;最后,在点云配准的基础上,将两部分点云数据统一到相同坐标系统,再对两部分点云数据进行融合建模,实现三维模型的精细化呈现.结果表明:该方法能够快速、精确地融合视频帧影像和倾斜摄影影像,生成高质量的三维模型,为不同数据的融合精细建模提供一种新的思路.
Research on Intergrating Fine Modeling Method of Point Cloud of Video Frame Image and Oblique Photography
Oblique Photography technology is an important way of three-dimensional modeling at present,with the advantages of high efficiency,low cost and flexible and fast,but its model structure is not complete,and surface scratches also exists,which is particularly serious in the construction of building models.In order to solve the prob-lem of insufficient modeling accuracy in oblique photography,this paper proposes a fine modeling method of video frame image fusion with oblique photography.Firstly,the oblique image of the survey area is collected by UAV and the obliquepoint cloud data is generated by internal processing;Secondly,the motion camera is used to obtain the video frame image of the buildings in the survey area,and the incremental Structure From Motion(SFM)algorithm and Multi-View Stereo(MVS)algorithm are used to reconstruct the three-dimensional point cloud data;Finally,on the basis of point cloud registration,the two parts of point cloud data are unified into the same coordinate sys-tem,and then the two parts of point cloud data are fused to achieve the fine presentation of the three-dimensional model.The results show that this method can fuse video frame image and oblique photographic image quickly and accurately,and generate high quality 3D model,providing a new idea for fine modeling of different data fusion.
万方数据
维普
