Abstract:
This research examined the effect of flight altitude and atmospheric weather condition on
unmanned aerial photogrammetry survey accuracy, thus its modeling to calculate and correct such errors. The research made use of Dji Phantom 3 Pro for image acquisistion at flight altitude ranging from (50 m – 150 m) at varying atmospheric conditions as monitored by the Automatic Weather Station (AWS). Prior to data acquisition, Ground Control Points (GCPs) and Checkpoints (CPs) were established and measured on the study area using GPS and were used for geo-referencing of pictures and post-photogrammetry accuracy assessment respectively. ImageJ Photogrammetry software was used for the photogrammetric processing of the 2D
images. To evaluate the photogrammetry positional accuracy, Horizontal coordinates of the CPs extracted from photogrammetry images were compared to those Predetermined by GPS. The positional accuracy of the CPs was determined by calculating the Root Mean Square Error (RMSE) of the coordinate difference between the photogrammetry and the GPS measured points in N and E. Analysis from this research shows that Total RMSE of unmanned photogrammetry survey has a direct relationship with flight altitude, i.e. with increasing flight altitude, the RMSE of survey increases. CPs of survey taken at altitudes 50 m, 60 m, 70 m, 80 m, 90 m, 110 m, 120 m, 130 m, 140 m, and 150 m have a RMSE value of 0.0138 m, 0.0185 m, 0.0220 m, 0.0247 m, 0.0281 m, 0.0360 m, 0.0391 m, 0.04157 m, 0.0574 m, 0.0611 m and 0.0652 m. The research also
established that wind speed and pressure are the most prominent atmospheric weather condition parameters that affect photogrammetry survey. It was revealed that air temperature affects positional accuracy due to contrast degradation of image background. The research recommend a maximum flight altitude of 95 meters at atmospheric temperature of 21oC -25oC, wind speed ≤ 8m/s and pressure ≥ 95.8 kpa for photogrammetric work to be done at 2.5cm resolution.
