Mühendislik ölçmelerinde İHA fotogrametrisi ve SFM'nın doğruluk analizi ve değerlendirilmesi

Highly detailed topographic mapping at minimal cost and effort has always been one of the developing areas of scientific interest. Image-based remote sensing solutions using Unmanned Aerial Systems (UAS) and Structure from Motion (SfM) with Multi-View Stereo (MVS) photogrammetry are the latest automation and advancement in surveying engineering. The aim of this research was to evaluate the accuracy of UAS SfM-MVS as a rapid and low-cost alternative to conventional survey methods. For this purpose, two flights at 100 and 170 m heights Above Ground Level (AGL) for nadir and one flight at 16 m AGL for oblique image acquisition were conducted using fixed-wing and rotary-wing UAS equipped with non-metric cameras. For validating the UAS-SfM products, field measurements were performed using Real-Time Kinematic Global Navigation Satellite System (RTK GNSS), terrestrial laser scanner, and total station. The collected images were processed using three SfM software packages. By applying the SfM-MVS workflow, 22 photogrammetric projects were processed, and as a result, point clouds, triangulated 3D models, Digital Surface Models (DSMs), Digital Terrain Models (DTMs), and orthomosaics were generated. Finally, absolute and relative accuracy assessments were performed. The spatial resolution of orthomosaics and DSMs derived from the UAS images of 16, 100, and 170 m AGL, was 0.005, 0.025, and 0.04 m, respectively. The highest accuracy was obtained from the results derived from 16 m AGL oblique images, which the horizontal and vertical RMSE values were 0.025 and 0.02 m, respectively. From the nadir images of 100 and 170 m AGL, RMSEXY of 0.03 m and RMSEZ of 0.04 and 0.06 m were achieved, respectively. The achieved accuracies of orthomosaics and DSMs reconstructed from 16, and 100 m AGL images are sufficient for most applications requiring terrain analysis, including civil engineering projects. Keywords: UAS Photogrammetry, Structure from motion (SfM), Digital elevation model, UAV, Terrestrial laser scanning