Due to technological developments, unmanned aerial vehicles (UAVs) have become a valuable tool in numerous applications including 3D spatial information extraction. UAV imagery can be employed efficiently to generate dense point clouds, digital surface models and orthoimages in short period of time. The aim of this research is to assess the accuracy of point clouds and orthoimages resulted from UAV photogrammetric mapping. A block of 70 UAV images captured with 80 % overlap and 80 % side lap at altitude of nearly 174m above ground is used. Many variables influencing the accuracy have been tested and evaluated which include percentage of overlap, georeferencing technique, configuration of ground control points, image resolution, as well as matching variables. The results showed
that higher values of image overlap and image resolution improved positioning accuracy of the generated point cloud. Good accuracy results are yielded by using direct and indirect georeferencing techniques. Also, the limits of key/tie points have a significant impact on positioning accuracy and processing time. The positioning accuracy of a few centimeters was achieved for the generated point cloud. It was found also that generated orthomosic has high
accuracy with RMSE values of 0.009m and 0.017m in the east and north directions, respectively, by using indirect georeferencing. The equivalent RMSE values were 0.020m and 0.019m considering direct georeferencing.