Large-scale topographic mapping of vegetation areas based on UAV and GNSS technology

Purpose. To develop a method for generating accurate large-scale topographic maps in densely vegetated areas using UAV (Unmanned Aerial Vehicle) and GNSS (Global Navigation Satellite System) technologies. The focus is on correcting elevation data from UAV imagery through a polynomial model based on CORS checkpoints. Methodology. The research was conducted in Mai Pha commune, Lang Son province, Vietnam. A DJI Phantom 3 Pro UAV was used to capture aerial images, and GNSS-RTK (CORS) technology was employed to collect ground control points. A polynomial surface fitting method (1st to 3rd degree) was applied to model vegetation thickness and correct the digital surface model (DSM) to obtain a digital elevation model (DEM). The DSM was smoothed using various radii (0; 20; 50; 70; 100 m), and the accuracy was assessed using different numbers of checkpoints (150; 350; 1,000). A software module was developed to automate the correction process. Findings. The accuracy of the DEM improved with an increased number of checkpoints and an appropriate smoothing radius. The best results were achieved with 1,000 checkpoints and a smoothing radius of 50 m. The corrected DEM achieved elevation accuracy within 1–2 meters, meeting regulatory standards for large-scale topographic maps (1:2,000–1:5,000). The method proved effective even in areas with dense vegetation, where traditional mapping methods face limitations. Originality. This study introduces a novel approach that integrates UAV photogrammetry with GNSS-RTK data and polynomial surface modeling to correct elevation data in vegetated areas. The developed software module automates the correction process, enhancing efficiency and consistency. Practical value. The proposed method enables the creation of accurate, large-scale digital topographic maps in challenging environments with dense vegetation. It offers a cost-effective and efficient alternative to traditional surveying methods, with potential applications in forestry, land management, and infrastructure planning.