小型无人机航测系统大比例尺测图方法研究

发布时间：2018-11-03 11:40

【摘要】：无人机航测系统作为大比例尺地图的手段,已经得到众多领域的认可。随着国产轻小型无人机平台和数码相机的飞速发展,技术的日益成熟,生产成本逐步下降,无人机摄影测量软件的自动化和智能化,无人机将会快步走入大比例尺地形图成图的领域,已经有了取代传统野外实测成图的趋势。摄影测量系统不仅能自动的进行严密的空中三角测量生成数字高程模型(DEM)、数字正射影像(DOM)和数字表面模型(DSM),还可以智能的通过影像匹配等多个步骤进行三维重建得到密集点云信息。目前数字线划图(DLG)的制作方法是基于传统的“双目视觉”的人工立体测图。本文结合工程实例,探讨基于密集点云与正射影像相结合的小型无人机航测系统大比例尺测图方法。该方法首先基于PIX4DMAPPER软件生成研究区数字正射影像图(DOM)、数字表面模型(DSM)、密集点云数据;在此基础上,采用MAPMATRIX“双目视觉”人工立体测图、ARCGIS配合CASS成图、PIX4DMAPPER点云编辑配合CASS成图三种数字线划图成图方法,进行DLG的制作;最后,将该方法与传统基于“双目视觉”的人工立体测图方法进行精度和效率的对比分析。结果表明,本文方法避免了人眼判读的误差,成图精度可靠;对成图人员的要求降低,无需进行长时间的训练就可进行成图工作;生产效率得到了较大提高。
[Abstract]:As a means of large scale map, UAV aerial survey system has been recognized in many fields. With the rapid development of domestic light and small UAV platform and digital camera, the technology is maturing day by day, the production cost is decreasing step by step, the UAV photogrammetry software is automatic and intelligent. UAVs will move quickly into the field of mapping large scale topographic maps, and there is a tendency to replace the traditional field mapping. Photogrammetry system can not only automatically carry out strict aerial triangulation to generate digital elevation model (DEM), digital orthophoto (DOM) and digital surface model (DSM), It can also be intelligently reconstructed by image matching and other steps to obtain dense point cloud information. At present, the method of making digital line drawing (DLG) is based on the traditional "binocular vision" artificial stereo mapping. In this paper, based on the combination of dense point cloud and orthophoto image, a large scale mapping method for small UAV aerial survey system is discussed. Firstly, the method is based on PIX4DMAPPER software to generate (DSM), dense point cloud data of (DOM), digital surface model. On this basis, three digital line mapping methods, MAPMATRIX "binocular vision" artificial stereo mapping, ARCGIS and CASS mapping, PIX4DMAPPER point cloud editing and CASS mapping are used to make DLG. Finally, the accuracy and efficiency of this method are compared with the traditional stereo mapping method based on binocular vision. The results show that the method avoids the error of human eye interpretation and the mapping accuracy is reliable. The requirements of the cartographers are reduced and the mapping work can be carried out without long training. The production efficiency has been greatly improved.
【学位授予单位】：成都理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：P231

【参考文献】