隧道是台灣重要的民生基礎建設的一部分，而隧道崩塌可能造成人員傷亡並
造成嚴重的災害，因此需要定期檢查與維護。在定期檢查中，已經有方便的裝置
減輕檢查人員的負擔，如使用手持行動裝置進行拍照、錄影或是填寫電子表單，
但是仍然缺乏隧道中自動定位的技術。
前人建立隧道影像定位研究平台，使用深度相機系統透過立體影像進行隧道
內定位。然而該平台在特徵點匹配與追蹤正確度不高的情況下，系統定位結果準
確度不佳。本研究透過在隧道數值模型中添加特徵點資訊，在渲染影像時計算特
徵點影像座標，使其特徵點有亞像素精度的影像位置，並能進行完美的特徵點匹
配與追蹤求得精準的定位結果，里程誤差不到 0.0001%、位移誤差均低於 1.4%、
旋轉誤差維持大約 0.4deg/m。
藉由分析模擬的特徵點影像位置數值，量化精度損失造成的特徵點誤差，探
討特徵點的誤差對於定位結果的影響，並與 SIFT、SURF、AKAZE、ORB 等演
算法的定位結果比較。

Tunnels are a part of important civil infrastructure in Taiwan. The collapse of the tunnel may
cause devastating disasters and cost human lives, so regular inspections and maintenance are
necessary. In regular inspections, there are already convenient devices to reduce the burden on
inspectors, such as taking photos with mobile devices, recording videos or filling in electronic
forms, etc. But the technology of automatic localization in tunnels is still lacking.
Former studies have developed a research platform for using computer vision technique of
localization in tunnels. However, under the condition that the accuracy of feature point
matching and tracking is unsatisfactory, the accuracy of the localization is not good. In this
study, by adding feature point information to the numerical model of the tunnel, the image
position of the feature points can be calculated when rendering the image. So that the feature
points have subpixel-accurate image positions, and perfect feature point matching and tracking
can be used to obtain accurate positioning. As a result, the mileage error is less than 0.0001%,
the translation error is less than 1.4%, and the rotation error is maintained at about 0.4deg / m.
By analyzing the value of simulated feature points' image position, quantifying the feature
points error caused by the loss of accuracy, exploring the influence of the feature point error on
the positioning result, and comparing with the positioning results of SIFT, SURF, AKAZE,
ORB algorithms.

參考文獻
[1] 廖育慶, "應用電腦視覺於隧道內定位之研究," 碩士, 營建工程系, 國立臺灣科技大學, 2014.
[2] 翁琪婷, "應用電腦視覺於隧道內定位之特徵點匹配過濾分析," 碩士, 營建工程系, 國立臺灣科技大學, 2018.
[3] 陳頎夫, "應用電腦視覺於隧道內定位之初探," 碩士, 營建工程系, 國立臺灣科技大學, 2012.
[4] H. Hirschmuller, P. R. Innocent, and J. M. Garibaldi, "Fast, unconstrained camera motion estimation from stereo without tracking and robust statistics," in 7th International Conference on Control, Automation, Robotics and Vision, 2002. vol. 2, pp. 1099-1104 , doi: 10.1109/ICARCV.2002.1238577.
[5] F. Zeng, A. Jacobson, D. Smith, N. Boswell, T. Peynot, and M. Milford, "TIMTAM: Tunnel-Image Texturally Accorded Mosaic for Location Refinement of Underground Vehicles With a Single Camera," IEEE Robotics and Automation Letters, vol. 4, no. 4, pp. 4362-4369, 2019, doi: 10.1109/LRA.2019.2932579.
[6] H. P. Moravec, "Obstacle avoidance and navigation in the real world by a seeing robot rover," Stanford Univ CA Dept of Computer Science, 1980.
[7] D. Nister, O. Naroditsky, and J. Bergen, "Visual odometry," in Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. vol. 1, pp. I-I, doi: 10.1109/CVPR.2004.1315094.
[8] H. Durrant-Whyte and T. Bailey, "Simultaneous localization and mapping: part I," IEEE Robotics & Automation Magazine, vol. 13, no. 2, pp. 99-110, 2006, doi: 10.1109/MRA.2006.1638022.
[9] S. Ullman, "The interpretation of structure from motion," Proceedings of the Royal Society of London. Series B. Biological Sciences, vol. 203, no. 1153, pp. 405-426, 1979.
[10] D. W. Eggert, A. Lorusso, and R. B. Fisher, "Estimating 3-D rigid body transformations: a comparison of four major algorithms," Machine vision and applications, vol. 9, no. 5-6, pp. 272-290, 1997.
[11] K. S. Arun, T. S. Huang, and S. D. Blostein, "Least-squares fitting of two 3-D point sets," IEEE Transactions on pattern analysis and machine intelligence, vol. PAMI-9, no. 5, pp. 698-700, 1987, doi: 10.1109/TPAMI.1987.4767965.
[12] S. Umeyama, "Least-squares estimation of transformation parameters between two point patterns," IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 13, no. 4, pp. 376-380, 1991, doi: 10.1109/34.88573.
[13] K. Kanatani, "Analysis of 3-D rotation fitting," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 16, no. 5, pp. 543-549, 1994, doi: 10.1109/34.291441.
[14] B. K. Horn, H. M. Hilden, and S. Negahdaripour, "Closed-form solution of absolute orientation using orthonormal matrices," Journal of the Optical Society of America, vol. 5, no. 7, pp. 1127-1135, 1988.
[15] B. K. Horn, "Closed-form solution of absolute orientation using unit quaternions," Journal of the Optical Society of America, vol. 4, no. 4, pp. 629-642, 1987.
[16] M. W. Walker, L. Shao, and R. A. Volz, "Estimating 3-D location parameters using dual number quaternions," CVGIP: image understanding, vol. 54, no. 3, pp. 358-367, 1991.
[17] Wikipedia. "VTK." Retrieved January 17, 2020, from https://en.wikipedia.org/w/index.php?title=VTK&oldid=923011689
[18] Kitware, "VTK - The Visualization Toolkit," Retrieved January 17, 2020, from https://vtk.org/.
[19] Kitware, VTK User’s Guide. Retrieved January 17, 2020, from https://www.kitware.com/products/books/VTKUsersGuide.pdf
[20] W. contributors. "OpenCV." Retrieved January 17, 2020, from https://en.wikipedia.org/w/index.php?title=OpenCV&oldid=933245332
[21] intel, "OpenCV," Retrieved January 17, 2020, from https://opencv.org/.
[22] D. G. Lowe, "Distinctive Image Features from Scale-Invariant Keypoints," International Journal of Computer Vision, vol. 60, no. 2, pp. 91-110, 2004/11/01, doi: 10.1023/B:VISI.0000029664.99615.94.
[23] G. Guennebaud. "Eigen." Retrieved January 17, 2020, from http://eigen.tuxfamily.org/index.php?title=Main_Page
[24] 中華民國交通部, 公路路線設計規範. 108修訂完整版.