Evaluation of Land Surface Temperature Orthophoto Generation using Thermal Infrared Video Stream from Unmanned Aerial Vehicles; 무인 항공기의 열적외선 영상 스트림을 이용한 지표면 온도 정사영상 생성 평가

초록

This study developed LST (Land Surface Temperature) orthophotos using a TIR (Thermal Infrared) sensor mounted on an UAV (Unmanned Aerial Vehicle) and evaluated the accuracy based on ground data measured with a laser thermometer. Unlike traditional single-image methods, continuous data were collected through TIR video recording, enabling a more precise reflection of land surface temperature. Results from July and August 2023 showed an average temperature difference of 1.21℃ and 0.49℃ between the TIR sensor and the laser thermometer, respectively, which falls within the Zenmuse XT2’s accuracy range of ±5℃ or ±5%. Materials that absorb significant heat, such as asphalt and urethane, exhibited larger temperature differences, while marble and vegetation showed relatively smaller differences. Compared to traditional photogrammetry, TIR video recording reduced battery consumption, improving UAV flight time and data collection efficiency.

키워드