衛照影像可提供大範圍的圖資資訊,當利用衛照影像進行三維物點定位,其品質主要受到物像對應參數、像點量測及交會幾何影響,其中,物像對應參數攸關物像對應射線(從像到物空間)在空間的正確性,考量衛照影像成像時的長物距,物像對應參數誤差所導致的射線偏轉(尤其是與姿態有關之參數)對物點定位影響不容忽視。在一般作業考量下,高品質的物像對應參數可由大量高精度及分佈良好的地面控制點來解算,或者在已具相對精度良好的物像對應參數條件下,利用少數控制點進行參數精化。本研究目的在探討如何將衛照影像結合物空間較穩固的約制,來解算衛照影像物像對應參數。 研究內容在物空間的約制上包含以下幾種形態:控制點、來自已具方位參數的航照影像共軛光線(控制射線)及前述兩種混合下的約制方式探討及效能分析;在物像對應解算模式上涵蓋動態共線式模式及有理函數模式。本研究的具體工作成果並非意欲取代常態作業模式,而是提供整合衛照及航照影像圖資的作業可能性及其增益。
The 3-D geo-positioning errors of satellite imagery are mainly affected by the quality of exterior orientation parameters, image point measurements and intersection geometry. Among them, pose-related parameters are vital to the geo-positioning accuracy considering the long distance from a satellite to the ground while taking imagery. In general, high quality exterior orientation parameters can be calculated by utilizing a great amount of as well as well distributed control points, or refined by only a few control points if relatively precise exterior orientation parameters are provided. The goal of this research is to investigate how to acquire or to promote high quality exterior orientation parameters of satellite imagery by employing strongly constrained rays coming from aerial imagery. The author consider three forms of constraints, control points, control rays and a hybrid constraint supported by aerial imagery and use rigorous sensor model and rational function model for analyzing object to image correspondence of the satellite imagery. The proposed working scheme is found to be effective and can be only regarded as an alternative when fusing aerial and satellite imagery is applicable.