Analysis of accuracy factor for 3D reconstruction using 2D image obtained from unmanned aerial vehicle (UAV)

Abstract

Recently UAV has been used in various fields such as construction and civil engineering, disaster, terrain analysis, transportation, etc. In addition, many companies from all over the world are focusing on developing UAV technology. The market of personal UAV has been growing and attracting people's attention year after year due to the low cost of UAV. Especially, in construction industries where people work on large-scale fields, there are many methods to obtain space information and those methods can be done at low cost. However, the existing studies are mostly about utilization of UAV such as flight plan and acquisition 3d image methods. Therefore, it is necessary to study result values according to use of variable value. This study focuses on the variables which affect the result of 3D reconstruction using 2D image obtained by UAV. First, we fixed values of flight planning of UAV and distortion correction of picture. After that, we took pictures of building and obtained point cloud that allowed us to make 3D model of the building. Second, we analyzed results of the 3D model according to variable of illumination. Then, we found number of point cloud from the previous results. Finally, we identified optimal illumination result to build an effective 3D Modeling. We present about optimal value of illuminance variation to improve current method of developing 3d model. This study can be a more efficient and accurate method to obtain images by UAV.