都市地區應用無人機攝影產生高解析度正射熱影像的方法

A Method to Produce High-Resolution Thermal Orthoimage by Applying UAV Photography Techniques in Urban Areas

Landsat系列衛星的地表溫度（LST）解析度僅100*100 m，對於都市地區的熱環境解析而言過於粗糙。本研究開發了一種新穎的攝影方法，利用無人機在100 m的高度拍攝產生每個像素分辨率高達0.2*0.2 m的正射地表溫度地圖。運用中國文化大學校園為實驗空間，使用無人機（DJI Mavic 2）進行資料收集。內建兩個鏡頭，包含RGB相機和熱感相機，並且同時記錄具有精確GPS資訊的實景及熱影像。利用Context Capture實景建模軟體對影像進行處理，成功製作出解析度為0.2*0.2 m精度的熱正射影像。在這個尺度上，可以清楚地識別單一建築物和其他景觀單元的熱特徵，提供一種更實用和有效的方法來分析小尺度的熱環境。

The resolution of land surface temperature (LST) derived from Landsat satellite series is only 100*100 m, which is too rough for observations of detailed heat in urban areas. The present research developed a novel photography method to generate a thermal map with a resolution as high as 0.2*0.2 m for every pixel by applying unmanned aerial vehicles (UAVs) at a height of 100 m. A campus in Taipei was taken as the experimental space. The latest UVA (DJI Mavic 2) was used for data collection. The two built-in cameras, one RGB camera and one thermal camera, recorded images with precise GPS information simultaneously. Processing the images using Context Capture reality modelling software, which was produced successfully with a resolution of thermal orthoimage with 20*20 cm precision. At this scale, the heat features of individual buildings and other landscape units can be clearly recognized. This practice presents a method with increased utility and effectiveness to study thermal conditions in the small scale.