自空氣往水中之攝影測量相對方位解算

Relative Orientation in Air-to-water Photogrammetry

黃聖日(Sheng-Jih Huang)

指導教授 ： 趙鍵哲

國立臺灣大學/工學院/土木工程學研究所/碩士(2014年)

https://doi.org/10.6342/NTU.2014.02094

摘要

在單介質的環境中，攝影測量基於中心透視投影進行物像對應。然而，當光線成像的過程中穿過不同介質時，受到折射效應影響，像點必須經由改正以符共線特性，這種特殊的攝影測量課題稱為雙介質攝影測量。
在雙介質的環境中，可能無法或不易接觸待測物而無法取得控制資訊，此時無須倚賴控制資訊的相對方位即嶄露其效用。另外，考量到水與空氣交界的場景是測量任務中較常碰觸的課題，因此，本文探討在無控制資料下，自空氣往水中攝影(文中即稱雙介質攝影)之相對方位解算。主要工作內容包含：建立折射光線行經模型空間及至像空間的軌跡，並建構物像對應方程式。再據此，針對相對方位解算品質以及求解條件幾何進行定性及定量分析。
模擬資料及實際資料的實驗成果顯示當介面參數及折射係數資訊皆為未知時，雙介質攝影之物像對應相較於單介質攝影之物像對應為一幾何較弱之結構，解算系統的穩定性及參數的解算品質易受微量之觀測量誤差影響，而造成解算系統不穩定，使相對方位無法解算或解算成果品質不符合應用需求。除此之外，相對方位解算品質取決於成像幾何、觀測量品質及介面資訊。當這些影響因子品質不佳或資訊不足時會使相對方位解算系統條件不好，進而造成無法進行解算或獲取品質不佳之結果。因此，為提升相對方位解算品質，宜以額外介面資訊或觀測量作為約制條件或以多像進行解算，如此能有效穩固介面之網型幾何，使求解精度大幅提升。

關鍵字

雙介質 ； 折射 ； 相對方位 ； 自空氣往水中

並列摘要

The effect of refraction would bend the imaging ray when passing through varied media. To analyze perspective projection under air-to-water imaging situation, the conventional collineaity-based approach must be modified to cope with refraction distortion. Relative orientation is crucial to the mission of 3D reconstruction under air-to-water environment, when control information is not available or simply irrelevant to the tasks. Therefore, the authors in this study aim to derive the collinearity-based approach and analyze how the quality of the relative orientation when dealing with air-to-water imaging geometry is affected, especially taking the factors of uncertainty of refraction indices, interface and imaging geometry into main consideration. Results obtained from both simulation and experimental tests quantify the performance of relative orientation in air-to-water photogrammetry and acknowledge that the multi-media relative orientation quality strongly depends on how much the knowledge of interface and media can be accurately captured and how images are taken.

並列關鍵字

Two-media ； Refraction ； Relative Orientation ； Air-to-Water

參考文獻

Ke, X., M.A. Sutton, S.M. Lessner, and M. Yost, 2008. Robust stereo vision and calibration methodology for accurate three-dimensional digital image correlation measurements on submerged objects. The Journal of Strain Analysis for Engineering Design, 43(8):689-704.

Kotowski, R., 1988. Phototriangulation in Multi-media Photogrammetry, International Archives of Photogrammetry and Remote Sensing, ISPRS 1988 Congress, Vol. XXXVII, Part 5, pp. 324-334.

Lavest, J.M., G. Rives, and J.T. Laprest’e, 2003. Dry camera calibration for underwater applications, Machine Vision and Applications, 13:245-253.

Li, R., H. Li, W. Zou, R.G. Smith, and T.A. Curran, 1997. Quantitative photogrammetric analysis of digital underwater video imagery, IEEE Journal of Oceanic Engineering, 22(2):364-375.

Shan, J., 1994. Relative Orientation for Two-media Photogrammetry. The Photogrammetric Record, 84(14):993-999.