Consistency Analysis of RTK and Non-RTK UAV DSMs in Vegetated Areas

Abstract

Lately unmanned aerial vehicles (UAVs) evolve into one of the most popular remote sensing technologies by providing rapid and periodical acquisition of high-resolution, accurate, and low-cost aerial data. Progressively, the technology of UAVs rises with the addition of advanced equipment such as multispectral (MS) digital cameras and global navigation satellite system (GNSS) receivers with real-time kinematic (RTK) positioning ability. While MS cameras aid to achieve higher land use and land cover classification accuracy, RTK GNSS receivers provide real-time positioning without the need for ground control points (GCPs), reducing fieldwork prior to UAV flights. In this study, the consistency of digital surface models (DSM) produced using RTK and non-RTK UAVs' simultaneously acquired data was evaluated with model-based visual and statistical comparison analysis in a densely vegetated study area. The analysis was done for six different vegetation classes utilizing non-RTK UAV DSM which is GCP-supported and has higher spatial resolution, as the reference model. To determine the effect of terrain inclination, the whole area and uninclined areas (arctan-10.1 similar to 6 degrees) were analyzed separately. The visual and statistical outcomes show that the absolute vertical consistency of the DSMs depending on the restricted positioning potential of on-board RTK GNSS receivers in the UAVs is insufficient for particularly high vegetation areas and GCP-supported absolute orientation is required.