Abstract
In this paper, a method of slope surface model reconstruction based on smart phone is proposed for the problems of point layout limitation, expensive measuring equipment and easy operation of monitoring personnel. By explaining the principle and steps of the SfM-MVS algorithm, the slope surface model is reconstructed based on the slope image taken by smart phone. In this paper, the principle of partial reconstruction is described, which involves the principle of polar geometry and projection error, and the corresponding description of motion recovery structure algorithm and dense reconstruction algorithm steps. The concrete steps of slope 3D reconstruction are as follows: first, the mobile phone camera is calibrated by Zhang Zhengyou camera and the slope image is enhanced for subsequent processing. Finally, the SfM-MVS algorithm is used for sparse reconstruction and dense reconstruction to obtain point cloud data, and the slope surface model is reconstructed by triangulation and texture mapping. The slope surface model obtained by smartphone image lays the foundation of slope shape change monitoring and coordinate calculation, and has the characteristics of perfect slope overall information. The model reconstruction based on mobile terminal image acquisition can reduce the equipment cost of slope monitoring and the maneuverability of monitoring personnel, and has certain application value.
Highlights
SfM-MVS 算法获取边坡影像特征点的点云数据再使用 Delaunay三角剖分进行重建并纹理映射获取具有纹理特 征的边坡表面模型。此表面模型能够为后续坐标推算以及 边坡形态变化监测提供有利条件,不同时间段的重建模型 [13] Peppa M V, Mills J P, Moore P, et al Automated co-registration and calibration in SfM photogrammetry for landslide change detection: Automated SfM co-registration for landslide change detection [J]
a method of slope surface model reconstruction based on smart phone is proposed for the problems
the slope surface model is reconstructed based on the slope image taken by smart phone
Summary
SfM-MVS 算法获取边坡影像特征点的点云数据再使用 Delaunay三角剖分进行重建并纹理映射获取具有纹理特 征的边坡表面模型。此表面模型能够为后续坐标推算以及 边坡形态变化监测提供有利条件,不同时间段的重建模型 [13] Peppa M V, Mills J P, Moore P, et al Automated co-registration and calibration in SfM photogrammetry for landslide change detection: Automated SfM co-registration for landslide change detection [J]. 传统边坡监测是以水准仪、全站仪、GNSS、GPS、 雷达等[1,2,3]方法获取数据信息,同时由于边坡种类繁多、 监测点布置的局限性、监测设备贵重等原因,传统监测方 法无法全面反映边坡整体状况。随着计算机视觉、影像匹 配算法、立体视觉重建等技术的不断更新,近景摄影测量 [4]和无人机倾斜摄影测量[5,6]已经能够被用于边坡监测 领域,但量测相机、非量测相机[7]以及摄测无人机都存在 操作和价格上的限制,而高像素智能移动终端的普及则将 基于立体视觉三维重建的边坡监测变为可能[8,9]。本研究 利用智能手机对边坡进行多视角拍摄,结合影像匹配、 SfM-MVS算法以及三角网格剖分和纹理映射,实现边坡 的单目多视角三维重建[10],为边坡稳定性评价奠定基础。 以上述方法得到的匹配对因仅依据特征点外观特征 判断而存在大量误匹配,故采用几何约束来验证匹配,不 同图像间的相同点存在的映射关系可用单应性矩阵H来表 示。图5对极几何中基础矩阵F反映点p在立体像对中像点 极线映射关系,即对极几何代数表示。本质矩阵E反映点p 在立体像对中获取两张图像的相机位姿之间的关联,其不 涉及相机内参。 性 , 重建场景精度高 。 增 量 式 SFM 重建过程中使用 RANSAC不断过滤外点从而可以不断优化场景结构。因此 增量式SfM契合边坡工程的要求[13,14]。其算法流程如下: (1) 对应点搜索 图像匹配中主要关注点在于特征点和特征描述子。特 征点部分包含方向、尺度信息。描述子通常作为向量表示, 用其描述关键点周围像素的信息。在任意一幅图像Ii中, 第 NFi 个像素点xj有特征点: 为避免求解基础矩阵F受到噪声干扰,在获取初始匹 配关系后选用RANSAC算法进行迭代优化,从而得到满足 对极几何一致性的正确匹配点对。RANSAC选择出可以估
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
