GROUND CONTROL POINTS AND THEIR INFLUENCES ON THE PRECISION OF GENERATING A DIGITAL SURFACE MODEL USING AN UNMANNED AERIAL VEHICLE

Abstract

Purpose. To determine the optimal number and spatial distribution of Ground Control Points (GCPs) required to achieve high-precision georeferencing in unmanned aerial vehicle (UAV) imagery, particularly for applications requiring 1:1,000 scale mapping. Methodology. Seven experimental scenarios were conducted, varying the number of GCPs from 4 to 30. For each scenario, GCPs were arranged in seven different spatial configurations, including central, corner, edge, and evenly distributed placements. The Root Mean Squared Error (RMSE) was calculated for each configuration to assess georeferencing accuracy. Findings. The results showed that using only 4 GCPs produced the highest RMSE, indicating the lowest accuracy. RMSE values decreased as the number of GCPs increased, with minimal improvement beyond 20 GCPs. Among all distribution patterns, placing GCPs at the corners consistently resulted in the highest RMSE. The most accurate results were achieved with 20 evenly distributed GCPs. Originality. This study provides a systematic evaluation of both the quantity and spatial arrangement of GCPs in UAV photogrammetry, offering empirical evidence to support optimal GCP deployment strategies. Practical value. The findings offer practical guidance for UAV mapping professionals, suggesting that 20 evenly distributed GCPs are sufficient to meet the accuracy standards for 1:1,000 scale maps. This helps optimize fieldwork efficiency while ensuring data quality.