基于无人机航测技术的农村建筑物矢量化方法研究  

Research on vectorization method of rural buildings based on UAV aerial survey technology

作　　者：任德旺 REN Dewang(Guangdong Cartographic Institute,Guangzhou,Guangdong 510000,China)

机构地区：[1]广东省地图院,广东广州510000

出　　处：《测绘技术装备》2024年第3期36-39,共4页Geomatics Technology and Equipment

摘　　要：本文以P村为例,研究如何快速将农村建筑物进行矢量化。首先,运用无人机航测技术和实景三维建模技术建立数字化模型;其次,通过Matlab软件调用反向传播(BP)神经网络模型对实景三维模型进行边界矢量化迭代,实现对分布分散、布局混乱的农村建筑物的快速测绘和边界快速提取、优化与识别的矢量化过程。误差分析结果表明,本文矢量化方法成果精度相对较高,具有一定的研究意义与应用前景。This paper studies the method of quickly vectoring rural buildings by taking P village in Guangdong Province as an example.It firstly establishes a digital model by using UAV aerial survey technology and real-time 3D modeling technology.Then,it quickly surveys the scattered and disordered rural buildings,and implements the vectorization process of automatic boundary extraction,optimization,and recognition by performing boundary vectorization iteration through involking real-time 3D model and BP neural network model in Matlab.It is concluded from error analyzing that the accuracy from this method is relatively good,it has some research significance and application prospects.

关 键 词：无人机 实景三维建模 BP神经网络模型 建筑物矢量化 

分 类 号：P231[天文地球—摄影测量与遥感]