关键词: 无人机, 运动恢复结构(SfM), 多视立体视觉(MVS), 数字表面模型, 真正射影像

Abstract:

In recent years,unmanned aerial vehicle(UAV) low-altitude remote sensing technology continues to develop,true digital orthophoto maps(TDOM) generated by UAV images still need to get improved in mapping accuracy and production process.In this paper,images were collected using a fixed-wing UAV and professional camera,and ground control points were laid out.A structure from motion (SfM) and multi view stereo (MVS) workflow was proposed to generate a high-precision digital surface model (DSM) and digital orthophoto map(DOM).We performed shadow tests to determine areas of tilt,perform DSM revision,and compensate for multi-view textures to produce a TDOM.Finally,we used random distributed checkpoints on the TDOM to verify its precision.The horizontal accuracy was 3.3 cm and the vertical accuracy was 7.5 cm.Tilt and shadow parts of the DOM were eliminated so that buildings maintained a vertical viewing angle.The generated TDOM meets precision requirements at the 1:500 scale. Therefore,it can be applied to rural homesteads to define land rights,land use planning,and other factors.

Key words: UAV, struct from motion, multi-view stereo, digital surface model, true digital orthophoto map