基於無人機航拍技術的蓬勃發展,目前市面上已有許多特別針對UAS影像進行後續解算處理並產製相關應用產品(正射影像、DEM/DSM、等高線、密點雲模型)的商業軟體,是以空三方位解算成果的精度深刻影響著後端產品的精度。國內在傳統航測作業上對於控制點之佈設位置與數量已有明確的規範加以訂定(如:內政部基本圖測製規範),但針對UAS作業卻無相關規範可供遵循,且各商業軟體中也幾乎未對控制點的需求有詳細說明。 本研究試圖以實際資料驗證控制點佈設密度對成果解算精度之影響,並考量不同商業軟體之解算模式對控制點的需求及敏感程度可能會有所差異,茲以目前市面上基於攝影測量或電腦視覺不同解算模式原理之商業軟體(基於作業資源考量,選定EnsoMOSAIC、ORIMA、Pix4D、APS)進行實驗及分析,期能在實務作業上提供最佳的控制配置方案。相關結論與建議乃基於本次實驗之成果。
Benefited from the newly developed UAS photogrammetry technology, commercial softwares converting images into photogrammetry productions such as orthophotos, DEM/DSM, contour lines, dense matching models are available. The quality of photogrammetric productions is directly related to the results of the aerotriangulation adjustment of UAS photogrammetry. Regulations about arrangement of ground control points (GCPs) in traditional photogrammetry are announced by National Land Surveying and Mapping Center, Ministry of the Interior (NLSC) for years. With regard to UAS, none of related regulations allows users to follow. In this paper, influence of different arrangement in GCPs with different commercial softwares are presented. All the execution models of the softwares are based on bundle adjustment with self-calibration or/and computer vision. The experiment results show the most appropriate methodology of different commercial software. Finally, conclusions and suggestions are illustrated based on the experiment results.