PEMANFAATAN DIGITAL TERRAIN MODEL HASIL FOTOGRAMETRI UNMANED AERIAL VEHICLE UNTUK ESTIMASI KETINGGIAN GENANGAN AIR BANJIR HASIL DETEKSI CITRA SYNTHETIC APPERTURE RADAR

Abstract

Utilization of The Unmanned Aerial Vehicle (UAV) or Drone has brought revolution in digital photogrammetry. The feature matching on surface reconstruction or Digital Surface Model (DSM) are quickly finished. However, DSM doesn’t represent itself as a part of topography, that is why DSM should be converted into Digital Terrain Model (DTM). This research is to investigate the accuracy of UAV photogrammetry’s DTM  for hydraulic modeling purpose. This study has produced 4 sets of DTMs; 2 sets of DTMs with different grid resolution which are 2 cm & 40 cm, also the 2 other sets of DTM with extra fine nature algorithm and set of filtering parameters adjustment; bulge, offset, spike and standard deviation. Every DTM are validated by Ground Control Point (GCP) from Real Time Kinematic-Different Global Positioning System (RTK-DGPS) measurement. According to the validation, the adjustment of filtering parameters is the most accurate method with Root Mean Square Error (RMSE) of 6,17 cm for 2 cm resolution; and 5,22 cm for 40 cm resolution. Next, DTM UAV is used to estimate the flood water level from Synthetic Aperture Radar (SAR) Image detection with 46 flood images on Glane and Losser area, east part of Overijssel, The Netherlands, since October 2014 to December 2017, then validated with the insitu water level measurement and resulted RMSE 6,72 cm for set of UAV DTM’s 40 cm resolution with the filtering parameters adjustment. Therefore, this DTM UAV can be used as a topography parameter in hydraulic modeling, especially at the similar flat-surface terrain where this research have been conducted.  

Keywords: UAV photogrammetry, SAR detection, DTM production