臺灣博碩士論文加值系統

根據水保局最新的統計資料，全臺灣有1705條土石流潛勢溪流，遍佈159個鄉鎮，做好土石流防災預警工作有著非常重要的意義。現有的土石流監測站均安裝有固定式的室外監控攝像機，但這些攝影機只能回傳影像供學者後續研究使用，還不具備監測預警功能(例如自動判別土石流和計算流速等)。本研究旨在探索一種適合土石流現場流速計算的方法，研究重點在於使用現有的監控攝像機恢復圖像實際的三維資訊，後續再結合LSPIV技術實現土石流監測預警。本文主要的研究方法是使用光學變焦相機，通過變化兩個焦距恢復圖像的實際三維座標，再利用粒子追踪測速技術計算流體的表面流速。
研究的內容包含以下幾點：第一，比較基於點和基於面積的變焦測距實驗，發現基於面積的變焦測距方式結果更加穩定，適合實際應用；第二，利用變焦技術恢復了圖像的中幾個參考點之後，結合影像處理的方法計算的實驗水槽中流體的表面流速，平均誤差在10%之內；第三，設計一套適合現場使用的參數檢定實驗，獲得相機的內外參數矩陣，計算結果較為可靠。
本研究的意義在于土石流發生時，河床高層變化顯著，原有的參考點坐標被破壞，變焦的方式可以實時更新現場目標的三維資訊。通過以上的研究內容，初步驗證變焦測距應用於表面流速測量具有一定的實用性，但是本研究還未能真正應用在土石流監測現場，這也是未來需要努力的方向。

According to the latest statistics of the Soil and Water Conservation Bureau, there are 1705 potential debris flows in Taiwan, all over 159 villages and towns, so it is very important to do a good job in disaster prevention and relief work. The existing earthwork monitoring stations are equipped with fixed outdoor surveillance cameras, but these cameras can only return images for follow-up study of the use of experts, do not have the ability to monitor and warning (such as automatic identification of debris flow and the flow rate, etc.). The purpose of this study is to explore a method for calculating the flow velocity. The research focus is to use the existing zoom cameras to restore the physical three-dimensional information of the image, and then follow the LSPIV technology to realize the monitoring of debris flow. The main method of this paper is to use the optical zoom camera to restore the physical three-dimensional coordinates of the image by changing two focal lengths, and then use the particle tracking velocity technique to calculate the surface velocity of the fluid.
The results of the study include the following points: First, comparing the point-based and area-based depth measurement experiments, it is found that the area-based method is more stable and suitable for practical application. Second, using zoom technology to restore the three-dimensional information of the image, the surface velocity of the fluid in the experimental tank calculated by the method of image processing and the average error is within 10%, Third, a set of parameter calibration experiment is designed for calculating the camera''s internal and external parameters matrix, and the results are reliable.
The significance of this study is that when the debris flow occurs, the elevation of the riverbed changes significantly, then the original reference point coordinates will be destroyed, and the method of changing focus can update the three-dimensional information on the scene real-timely. Through the research, it is proved that applying depth estimation of the zoom images on surface velocity calculation is practical, but the research has not been applied to the monitoring site of the earth flow yet. This is also the future research direction.

第一章 緒論 1
1.1 研究背景與目的 1
1.2 文章主要內容與結構安排 2
第二章 文獻回顧 3
第三章 基本影像處理技術 6
3.1圖像前處理 6
3.1.1 灰階變化 6
3.1.2 圖像增強 8
3.1.3 圖像降噪 11
3.2圖像分割 13
3.3運動目標偵測 16
第四章 變焦相機的成像模型與標定 18
4.1 引言 18
4.2變焦相機的成像模型 18
4.2.1針孔成像模型 18
4.2.2厚透鏡模型 22
4.3基於Zhang的平面移動模板標定法 23
4.3.1標定理論 23
4.3.2標定實驗過程與結果 27
第五章 變焦圖像的深度估計 31
5.1 引言 31
5.2基於點的變焦測距 32
5.2.1基於點的測距理論 32
5.2.2基於點的變焦圖像深度估計實驗 33
5.3基於面積的變焦測距 38
5.3.1基於面積的測距理論 38
5.3.2基於面積的測距實驗 42
5.4兩種方法誤差敏感度分析 47
第六章 基於變焦方法的流速推估 50
6.1 引言 50
6.2 實驗流程架構 50
6.3影像處理過程 52
6.4表面流速估計 54
6.5實驗結果與誤差分析 56
6.5.1測距結果與誤差分析 56
6.5.2流速結果誤差分析 60
第七章 現場的參數檢定流程 62
7.1：引言 62
7.2：現場首次內外參數檢定方法 62
7.2.1 理論推導 62
7.2.2 用數值方法求解內部參數 64
7.2.3 室外的參數檢定實驗 69
7.3：現場參數檢定流程初步設計與後續修正 74
7.3.1現場參數檢定流程初步設計 74
7.3.2 相機參數的後續修正 74
第八章 結論與建議 76
參考文獻 77
附錄 矩陣中係數的具體表達 A

[1]Aldo, L. (1994). “The visual hull concept for silhouette-based image understanding.” IEEE Transactions on Pattern Analysis and Machine Intelligence, 16(2), 150-162.
[2]Arattano, M., and Grattoni, P. (2000). “Using a fixed video camera to measure debris flow velocity”, the 2nd International Conference on Debris Flow Hazard Mitigation
[3]Baroni, C. et al., (2003). “Complete grain-size analyses on debris-flow source area in the Carrara marble basins”, Apuane Alps, Italy, Proceedings 3rd International DFHM Conference, Davos, Swizerland, September 10-12, pp. 809-820.
[4]Creutin, J. D., Muste, M., Bradley, A. A., Kim, S. C., and Kruger, A.(2003). “River gauging using PIV techniques: A proof of concept experiment on the Iowa river.” Journal of Hydrology, Vol. 277,Nos. 3-4, pp. 182-194.
[5]Chan, S.Y., and Lin, C.P. (2007) “Debris flow detection using image processing techniques”, the 4th International Conference on Debris Flow Hazard Mitigation
[6]Ettema, R., Fujita, I., Muste, M., and Kruger, A. (1997). “Particle-image velocimetry for whole-field measurement of ice velocities.” Cold Regions Science and Technology, Vol. 26, No. 2, pp. 97-112.
[7]Fayman J F,Sudarsky O ,Rivlin E,et al. ZoomTracking and Its Applications{J}. Machine Vision and Applications,2001,13(1):25-37
[8]Fujita, I., Komura, S. (1994). “Application of video image analysis for measurements of river-surface flows.” Proc. of Hydraulic Engineering, JSCE, Vol. 38, No. 2, pp. 733-738.
[9]Fujita, I., Aya, S., and Deguchi, T. (1997). “Surface velocity measurement of river flow using video images of an oblique angle.” Proc., 27th IAHR Congress, Theme B, San Francisco.
[10]Fujita, I. and Hino, T. (2003). “Unseeded and seeded PIV measurements of river flows videotaped from a helicopter.” Journal of Visualization,Vol. 6, No. 3, pp. 245-252.
[11]Fujita, I., Watanaba, H., and Tsubaki, R. (2007). “Development of a nonintrusive and efficient flow monitoring technique: The space time image velocimetry (STIV).” International Journal of River Basin Management, Vol. 5, No. 2, pp. 105-114.
[12]Fujita, I. and Kunita, Y. (2011). “Application of aerial LSPIV to the 2002 flood of the Yodo River using a helicopter mounted high density video camera.” Journal of Hydro-environment Research,Vol. 5, pp. 323-31, DOI: 10.1016/j.jher.2011.05.003
[13]Hartley R.I. (1994) Self-calibration from multiple views with a rotating camera. In: Eklundh JO. (eds) Computer Vision — ECCV ''94. ECCV 1994. Lecture Notes in Computer Science, Vol. 800. Springer, Berlin, Heidelberg
[14]Hauet, A., Kruger, A., Krajewski, W. F., Bradley, A., Muste, M., and Creutin, J. D. (2008). “Experimental system for real-time discharge estimation using an image-based method.” Journal of Hydrological Engineering, Vol. 13, No. 2, pp. 105-110.
[15]Kantoush, S. A., Schleiss, A. J., Sumi, T., and Murasaki, M. (2011).“LSPIV implementation for environmental flow in various laboratory and field cases.” Journal of Hydro-environment Research, Vol. 5,No. 4, pp. 263-276.
[16]Lavest J.M.,Rives G, Dhome M. Three-dimensional reconstruction by zooming. IEEE Transactions on Robotics and Automation,199319(2):196-207
[17]Ma J，Olsen S I. Depth from Zooming[J]. Journal of the Optical Society of America, 1990 , 7(10):1883-1890
[18]Muste, M., Fujita, I., and Hauet, A. (2008). “Large-scale particle image velocimetry for measurements in riverine environments.” Water Resources Research, Vol. 44, No. 4, W00D19, DOI: 10.1029/2008WR006950.
[19]Muste, M., Ho, H-C., and Kim, D. (2011). “Considerations on direct stream flow measurements using video imagery: Outlook and research needs.” Journal of Hydro-environment Research, Vol. 5, No. 4, pp. 289-300.
[20]Patalano, A., García, C. M., Brevis, W., Bleninger, T., Guillen, N., Moreno, L., and Rodriguez, A. (2015). “Recent advances in Eulerian and Lagragian large-scale particle image velocimetry.” E-proceedings of the 36th IAHR World Congress (The Hague, the Netherlands).
[21]Rafael C. Gonzalez, Richard E. Woods (2007) Digital Image Processing (3rd Edition)
[22]Richard Szeliski (2010),Computer Vision:Algorithms and Application
[23]Shrivasthava P., Vundavilli P.R., Pratihad D.K., An approach for 3D reconstruction of environment using stereo-vision system. IEEE Region 10 and The Third International Conference on Industrial and Information Systems, 2008, 1-7
[24]Tsai R.Y., A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf TV camera and lenses. IEEE Trans on Robotics and Automation,1987,3(4):323-344
[25]Tsubaki, R., Fujita, I., and Tsutsumi, S. (2011). “Measurement of the flood discharge of a small-sized river using an existing digital video recording system.” Journal of Hydro-environment Research, Vol. 5, No. 4, pp. 313-321.
[26]Yang, M.D., Chao, C.F., and Huang, K.S. (2013). “3D model construction on imagery shot by mobile devices.” Journal of Aeronautics, Astronautics and Aviation, 45(3), 219-224.
[27]Zhang Zhengyou. A flexible new technique for camera calibration. IEEE Transaction on Pattern Analysis and Machine Intelligence,2000,22(11):1330-1334
[28]Zhang, S., and Chen, J. (2003). “Measurement of debris-flow surface characteristics through close-range photogrammetry”, the 3rd International Conference on Debris Flow Hazard Mitigation
[29]白絜成、劉益誠、蕭宇伸、連惠邦、林秉賢 (2015)，「無人飛行載具掛載消費型攝影機應用於防災可行性研究」，中華水土保持學報，46 (3)，142-149
[30]呂珍謀 (2008)，「影像分析方法應用於土石流表面速度計算之研究」，農業工程學報
[31]張守陽、黃榮堂、李璟芳 (2005)，「機械視覺應用於土石流監測之研究」，中華水土保持學報
[32]楊明德、陳韋廷、黃凱翔 (2015)，「應用UAV 影像建置現地堰塞壩三維模型」，中華水土保持學報，46 (2)，88-95。(Yang, M.D., Chen, W.T., and Huang, K.S. (2015)
[33]賴進松, et al. "UAV 影像技術應用於河道洪水位及流場之模擬分析." Journal of the Chinese Institute of Civil and Hydraulic Engineering 27.3 (2015): 231-240.
[34]蕭震洋、謝寶珊、冀樹勇 (2011)，「應用非常規攝影量測評估國道3 號3.1 公里崩塌事件之土方量」，中華水土保持學報，42(2)，120-130