三維空間數值地形模型(Digital Terrain Model)資料之使用在近年來有逐漸增加之趨勢，不論是在國防軍事、遙測影像、生態保育上都有其相關的應用。但由於要獲取地貌之三維立體座標遠比二維平面座標困難，因此，如何藉由影像處理與電腦視覺技術自動獲取地形的三維座標，已成為近年來電腦視覺研究領域中受重視之研究方向之一。

在一般三維地形模型重構的研究當中，其所使用的技術與方法常常需要仰賴特殊且昂貴的設備來協助，或是利用一些特殊的影像資料。此外，還需經由許多複雜而煩瑣人工設定與校準之後，才能獲得相關的資訊來產生真實的三維地形模型。然而，在本研究中，我們選擇利用一台一般的數位攝影機來拍攝我們所要重構地形場景的連續影像，再利用此一連續影像序列做特徵點的萃取與追踨。再藉由這些特徵點的對應關係來計算相對應的投影幾何關係，進而利用影像間投影幾何的關係來對攝影機做自我校正。最後，可自動地重構出整個真實場景在三維空間中的地形模型。故雖然我們的主要目的同樣是重構出真實場景的三維地形模型，但我們的系統十分簡易、方便，並不需要特殊的設備、儀器，或是其它額外的資訊，也不需要經過煩瑣的相機校正(calibration)及太多的人工介入，而同樣也能夠達到自動建立三維地形模型的目標。

Recently, the three-dimensional (3D) digital terrain model (DTM) has become very popular and widely used in various fields, including military, navigation, remote sensing, and ecological protection. To obtain 3D coordinates from a sequence of 2D terrain images is not a trivial task, and is nowadays an important topic in computer vision research. It is generally intended to recover the 3D coordinates of the terrain model automatically by using the image processing and computer vision techniques.
In most 3D terrain reconstruction research, expensive equipment or specialized image data structure are usually employed to acquire the 3D information. Furthermore, lots of manual operations, calibration, complex setup and regulation are also inevitable. In this thesis, we use a general-purposed digital video recorder to acquire a series of 2D images. By extracting and tracking the feature points of these images, the correlation among corresponding feature points in the consecutive images can be used to calculate the geometry structure. The camera parameters of the digital video recorder can be obtained after the self-calibration based on geometry structure. Finally, the 3D world coordinates of the model can be reconstructed by triangulation with these parameters automatically. In comparison with the existing methods, our system is simpler and more convenient without special equipment or extra information. Neither the complex camera calibration nor lots of artificial interference is necessary. Our proposed system is an easier alternative to reconstruct the 3D terrain model automatically.

[1] Z. Zhang, “Determining the Epipolar Geometry and its Uncertainty: A Review”, Technical Report RR-2927, INRIA, 1996.
[2] H.C. Longuet-Higgnis, “A Computer Algorithm for Reconstructing a Scene From Two Projections”, Nature, Vol. 293, pp. 133-135, Sept 1981.
[3] R.I. Hartley, “Projective reconstruction and invariants from multiple images”, IEEE Trans. on Pattern Analysis and Machine Intelligence, October 1994.
[4] Q.-T. Luong, R. Deriche, O.D. Faugeras, and T. Papadopoulo, “On Determining the Fundamental Matrix: Analysis of Different Methods and Experimental Results,” Technical Report RR-1894, INRIA, 1993.
[5] R.I. Hartley, “In Defense of the Eight-Point Algorithm”, IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 19, No. 6, June 1997.
[6] http://vision.stanford.edu/~birch/klt/
[7] Z. Zhang, R. Deriche, O. Faugeras, Q.T. Luong, “A robust technique for matching two uncalibrated images through the recovery of the unknown epipolar geometry.” Artificial Intelligence Journal, 78:87-119, 1995.
[8] P.H.S. Torr, D. W. Murray, “Outlier detection and motion segmentation.” In Proc SPIE Senor Fusion VI, pages 432-443, September, 1993.
[9] P.H.S. Torr, “Motion segment and outlier detection.” PhD thesis, Dept. of Engineering Science, University of Oxford, 1995.
[10] P.H.S. Torr, A. Zisserman, D. W. Murray, Motion clustering using the trilinear constraint over the three views.” Europe-China Workshop on Geometrical Modeling and Invariants for Computer Vision, Xidan University Press, 1995.
[11] P.H.S. Torr, A. Zisserman, “Robust parameterization and computation of the trifocal tensor.” Image and Vision Computing, 15:591-605, 1977.
[12] P.H.S. Torr, A. Zisserman, “Robust computation and parameterization of multiple view relations.” In Proc. 6th International Conference on Computer Vision, Bombay, India, January 1998.
[13] P.H.S. Torr, A. Zisserman, A. W. Fitzgibbon, “The Problem of degeneracy in structure and motion recovery from uncalibrated image sequences.” International Journal of Computer Vision, 32(1):27-44, August 1999.
[14] M. Fischler, R. Bolles, “Random sample consensus: A paradigm for model fitting with applications to image analysis and automated cartography.” Communications of the ACM, 24:381-385, 1981.
[15] Bruce D. Lucas and Takeo Kanade. “An Iterative Image Registration Technique with an Application to Stereo Vision.” International Joint Conference on Artificial Intelligence, pages 674-679, 1981.
[16] Carlo Tomasi and Takeo Kanade. “Detection and Tracking of Point Features.” Carnegie Mellon University Technical Report CMU-CS-91-132, April 1991.
[17] Jianbo Shi and Carlo Tomasi. “Good Features to Track.” IEEE Conference on Computer Vision and Pattern Recognition, pages 593-600, 1994.
[18] P. Beardsley, P. Torr, A. Zisserman, “3D Model Acquisition from Extended image Sequences.” Proc. ECCV. , Cambridge, UK, vol.2, pp.683-695, 1996.
[19] M. Pollefeys, R. Koch, M. Vergauwen and L. Van Gool, “Metric 3D Surface Reconstruction from Uncalibrated Image Sequences”, Proc. SMILE Workshop (post-ECCV'98), LNCS 1506, pp.138-153, Springer-Verlag, 1998.
[20] Schultz, H., Hanson, A., Holmes, C., Riseman, E., Slaymaker, D., Stolle, F., “Integrating Small Format Aerial Photography, Videography, and a Laser Profiler for Environmental Monitoring”, ISPRS WG III/1 Workshop on Integrated Sensor Calibration and Orientation, Portland, Maine, USA, June 16-17, 1999.
[21] Schultz, H., Terrain Reconstruction from Widely Separated Images,Proc. SPIE, Volume 2486, pp. 113-123, Orlando, FL, April, 1995.
[22] Zhu, Z., Hanson, A R., Schultz, H., Stolle, F., Riseman, E. M., Stereo Mosaics from a Moving Video Camera for Environmental Monitoring, First International Workshop on Digital and Computational Video, pp. 45-54, Tampa, Florida, Dec 10, 1999.
[23] Schultz, H., Terrain Reconstruction from Oblique Views, Proc. DARPA Image Understanding Workshop, Monterey, CA, November 1994, pp. 1001-1008.
[24] Schultz, H., Hanson, A., Riseman, E., Stolle, F., Zhu. Z., A System for Real-time Generation of Geo-referenced Terrain Models, SPIE Symposium on Enabling Technologies for Law Enforcement, Boston MA, Nov 5-8, 2000.
[25] C. Baillard, H. Maitre, “3D Reconstruction of Urban Scenes from Aerial Stereo Imagery: A Focusing Strategy.” Computer Vision and Image Understanding, Vol. 76, No. 3, December, pp244-258, 1999.
[26] M. Vergauwen, M. Pollefeys, R. Moreas, F. Xu, G. Visentin, L. Van Gool and H. Van Brussel. “Calibration, Terrain Reconstruction and Path Planning for a Planetary Exploration System”, Proc. i-SAIRAS 2001.
[27] Z. Zhang, “A Stereovision System for a Planetary Rover: Calibvration, Correlation, Registration, and Fusion.” Proc. IEEE Workshop on Planetary Rover Technology and Systems, April, 1996, USA.
[28] R. Bolter, F. Leberl, “Detection and Reconstruction of Human Scale Features from High Resolution Interferometric SAR Data.” Proceedings of ICPR 2000.
[29] B. Triggs, “The Absolute Quadric”, Proc. 1997 Conference on Computer Vision and Pattern Recognition, IEEE Computer Soc. Press, pp. 609-614, 1997.
[30] J.G. Semple and G.T. Kneebone, Algebraic Projective Geometry, Oxford University Press, 1952.
[31] M. Pollefeys, Obtaining 3D Models with a Hand-Held Camera, Course notes, Siggraph 2001, Sunday 12 august 2000.
[32] M. Pollefeys, R. Koch and L. Van Gool. Self-Calibration and Metric Reconstruction in spite of Varying and Unknown Internal Camera Parameters, International Journal of Computer Vision, 32(1), 7-25, 1999.
[33] R.Y. Tsai, “A Versatile Camera Calibration Technique for High-Accuracy 3D Machine Vision Metrology Using Off-the-Shelf TV Cameras and Lenses,” IEEE J. Robotics and Automation, vol. 3, no. 4, pp. 323-344, Aug.1987.
[34] Z. Zhang, “A Flexible New Technique for camera Calibration.” IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 22, No. 11, November 2000.
[35] A. Durer, “Underweysung der Messung.” Albaris Books, New York. Original edition 1525, Nuremberg.
[36] Olivier Faugeras, Quang-Tuan Luong, “The geometry of multiple images : the laws that govern the formation of multiple images of a scene and some of their applications.”, Cambridge, Mass. :/MIT Press, 2001.
[37] R. Hartley, A. Zisserman, “Multiple view geometry in computer vision”, Cambridge, UK ; Cambridge University Press,2000,New York.