控制影像於多時期影像方位解算之分析

傳統空中三角測量大都是處理同一時期之重疊影像，影像內容之幾何及輻射條件相似，以光束法平差進行物像對應解算已是常規性的作法。相較於此，多時期影像之空中三角測量作業，多了時間變因，因而衍生出特殊的幾何及輻射變化情況，這些特異現象在同時期影像空中三角測量之作業中並未加以考量。因此，如何在多時期影像之空中三角測量作業，建構有效的影像特徵共軛對應，並將物空間控制引入求解影像外方位參數，為本研究的目的。本研究主要探討多時期影像方位參數解算方式，工作中藉由已知方位參數影像，透過共軛特徵解算在重疊區中之多時期待解算影像方位參數。由於作業之特殊性，本工作將已知方位參數之影像稱為控制影像，據此，引出對應點、線特徵觀測量之控制射線及控制判斷面，提供解算的部分或唯一控制資料。本研究針對前述觀測量安排適當的平差函數及隨機模式，並定性分析控制射線及控制判點面的功能，透過模擬資料分析本任務中各類影響因子之作用。而在實際資料處理上，取用兩不同地形特徵之局部區域，進行多時期影像共軛特徵量測及方位解算。根據作業方法及成果分析歸結本研究藉由控制影像所推演之物空間控制確實能在多時期影像方位解算任務上發揮具體功效。

關鍵字

控制影像；控制射線；控制判斷面；前交點；前交線；點特徵；線特徵

並列摘要

The targets of traditional aerial triangulation are mostly the same period of overlapping photos. Because geometric and radiation conditions of photos are similar, solving aerial triangulation with bundle adjustment is easy to get photo orientations. Compared with this, aerial triangulation of multi-temporal photos contains the special geometry and radiation which derive from time factor. The specific condition didn't be considered in the same period of photos. Therefore, how to construct the effective conjugate corresponding at the multi-temporal photos aerial triangulation task, and introduction the control information from object space to solve orientation parameters is the purpose of the study. It investigates the orientation of multi-temporal photos solver method, using known orientation parameters, and it's called control photos, through conjugate features to solve the orientation of multi-temporal photos at the overlap region. Due to the special characteristics of the work, Using control photos to provide control rays and control interpretation planes. This study will use a mathematical model including the control photos, according the reliability of the control photos and link features type to give appropriate weight, to solve the corresponding between image and object space. Using control photos to solve the orientation of multi-temporal photos. According foregoing observations to arrange appropriate adjustment function and random model, and analysis the control rays and control interpretation qualitative function, through simulating date to analysis the influencing factors. In the actual data processing, accessing two different topographical features of the local area, to solve the multi-temporal photos orientation, methods and results analyzed according to the study using control photos can really apply specific effects in the multi-temporal photos orientation solving tasks.

並列關鍵字

Control photos ； Control rays ； Control interpretation planes ； Forward intersection point ； Forward intersection line ； Point features ； Line features

參考文獻

Habib, A., A. Asmamaw, D. Kelley, and M. May, 2000. Linear Features in Photogrammetry, Report No. 450, Department of Civil & Env. Eng. & Geodetic Science, The Ohio State University, Columbus, OH, USA, 75p.

Redweik, P., Roque, D., Marques, A., Matildes, R., Marques, F.,2010. Triangulating the Past– Recovering. Portugal’s Aerial Images Repository, Photogrammetric Engineering & Remote Sensing,Vol.76,No.9, pp.1007–1018.

Schenk, T., 2004. From point-based to feature-based aerial triangulation, ISPRS Journal of Photogrammetry and Remote Sensing, 58(5-6):315-329.

周尚弘，2005。GPS 與 INS 結合同軸數位量測相機之外方位精度分析，國立成功大學地球科學系專班學位論文，121p。

邱依屏，2009。航測網形分析，國立臺灣大學工學院土木工程學系研究所碩士論文，82p。

國際替代計量

控制影像於多時期影像方位解算之分析

全文下載

主題瀏覽