簡易檢索 / 詳目顯示

回結果列表
研究生： 翁琪婷
Chi-Ting Weng
論文名稱： 應用電腦視覺於隧道內定位之特徵點匹配過濾分析
Feature Point Matching Filter Analysis of Localization in Tunnels Using Computer Vision
指導教授： 謝佑明
Yo-Ming Hsieh
口試委員: 陳鴻銘
Hung-Ming Chen
楊元森
Yuan-Sen Yang
莊子毅
Tzu-Yi Chuang
謝佑明
Yo-Ming Hsieh
學位類別： 碩士
Master
系所名稱： 工程學院 - 營建工程系
Department of Civil and Construction Engineering
論文出版年： 2018
畢業學年度： 106
語文別： 中文
論文頁數： 98
中文關鍵詞： 電腦視覺 、隧道 、定位 、特徵點 、匹配
外文關鍵詞： computer vision, localization, tunnel, feature point
相關次數： 點閱：125  下載：11 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報
  •  上一筆

    • 　　無論是鐵路、公路、捷運、自行車道、行人專用道或是水利相關設施，因為台灣地形關係，擁有許多隧道。若隧道結構損毀，將會造成許多的問題，甚至是對生命安全的危害。因此，隧道需要定期的維護檢查與紀錄。在定期檢查上，使用行動裝置進行例如照相、錄音、錄影、表單勾選等電子化的記錄之技術已被開發出來，以減輕巡檢人員的負擔。然而，目前仍缺乏在隧道內自動取得位置資訊的技術。
    　　前人建立基於電腦視覺技術於隧道內定位之研究平台，其以電腦視覺利用視差計算隧道內特徵點之空間座標，並根據景物的移動來計算自身之相對移動距離，以達到隧道內定位之目的。本研究對此研究平台進行函式庫更新、功能的擴充及計算方法的改善。
    　　更新後之平台，可以自行搭配組合不定數量之影像處理流程與參數，亦可選擇使用多種特徵點處理之方法與參數，以研究精準的隧道定位演算法。本研究亦開發新的匹配過濾之方法，以濾除不正確的特徵點匹配，藉以減少定位誤差。

    　　Tunnels are part of important civil infrastructure in Taiwan. The consequence of tunnel collapse is devastating and can cost human lives. Tunnels have been regulated to have regular inspections and maintenance. For regular inspections, mobile-device based operations such as taking photos, recording videos, filling electronic forms, etc. have been developed to help reduce burden to inspectors. However, automated localization in tunnels is still lacking.
    Former studies have developed a research platform for using computer vision technique for localization in tunnels. The localization involves stereo imaging, feature point identification and tracking. This study updates the research platform by upgrading the used computer vision library, expanding its capabilities, and improving the algorithms.
    　　The updated research platform can customize image processing process with different algorithms and parameters. Feature points can be identified by one of many available feature point extraction algorithms with customizable parameters. The platform can be used to develop precise tunnel localization algorithms. New match-filtering algorithms are also developed in this research in order to remove incorrect stereo pairs to reduce localization errors.

    論文摘要 I ABSTRACT II 誌謝 III 目錄 V 圖目錄 IX 表目錄 XI 第一章　緒論 1 1.1 研究動機與目的 1 1.2 研究流程 2 1.3 論文架構 3 第二章　文獻回顧 5 2.1 電腦視覺 5 2.2 應用電腦視覺於隧道內定位之研究 6 2.2.1 電腦視覺定位研究平台 6 2.3 光流 8 第三章　研究方法與工具 11 3.1 OpenCV函式庫 11 3.2 影像處理的方法 12 3.2.1 雙邊濾波器 13 3.2.2 影像灰階化 14 3.2.3 直方圖均衡化 15 3.2.4 Canny邊緣檢測器 16 3.2.5 Harris角點檢測器 17 3.2.6 Shi – Tomasi角點檢測器 18 3.3 特徵點處理的方法 19 3.3.1 特徵點探測演算法與描述符演算法 19 3.3.1.1 GFTT 20 3.3.1.2 FAST 20 3.3.1.3 AGAST 20 3.3.1.4 SIFT 20 3.3.1.5 SURF 21 3.3.1.6 BRISK 21 3.3.1.7 ORB 21 3.3.1.8 MSER 21 3.3.1.9 KAZE 22 3.3.1.10 AKAZE 22 3.3.2 特徵點匹配 22 第四章　系統分析與設計 23 4.1 整體流程 23 4.2 程式類別與架構 24 4.3 純文字格式設定檔 - Config.yml 28 4.4 影像處理 40 4.5 造成匹配錯誤的因素探討 41 4.5.1 三角過濾法 41 4.5.2 四格過濾法 44 第五章　參數研究與誤差探討 47 5.1 影像處理分析 47 5.1.1 雙邊濾波 47 5.1.2 直方圖均衡化 48 5.1.3 Canny邊緣檢測 48 5.1.4 Harris角點檢測 49 5.1.5 Shi – Tomasi角點檢測器 50 5.1.6 小結 50 5.2 特徵點處理分析 51 5.2.1 SIFT 51 5.2.2 SURF 52 5.2.3 BRISK 52 5.2.4 ORB 53 5.2.5 KAZE 53 5.2.6 AKAZE 54 5.2.7 FAST 54 5.2.8 AGAST 55 5.2.9 GFTT 55 5.2.10 MSER 56 5.2.11 小結 56 5.3 匹配過濾分析 57 第六章　系統展示與驗證 59 6.1 系統展示 59 6.1.1 系統座標 59 6.1.2 左右匹配結果 60 6.1.3 前後匹配結果 60 6.1.4 內群過濾結果 61 6.1.5 特徵點追蹤 61 6.1.6 特徵點分布 62 6.1.7 移動追蹤 63 6.1.8 移動路徑 64 6.1.9 命令視窗 65 6.2 系統驗證 66 6.2.1 靜止測試 66 6.2.2 直線線型隧道 69 6.2.3 曲線線型隧道 72 6.2.4 小結 75 第七章　結論與建議 77 7.1 結論 77 7.2 建議 78 參考文獻 79

    1. 廖育慶, 應用電腦視覺於隧道內定位之研究, in 營建工程系. 2014, 國立臺灣科技大學: 台北市. p. 111.
    2. 羅聖學, 混合車流及高密度下之車輛影像辨識及軌跡追蹤, in 運輸與物流管理學系. 2017, 國立交通大學: 新竹市. p. 58.
    3. 邱彥霖, 應用電腦視覺處理混合車流之交通衝突分析, in 土木工程學研究所. 2016, 國立臺灣大學: 台北市. p. 57.
    4. Singh, A.K., et al., Vision based rail track extraction and monitoring through drone imagery. ICT Express, 2017.
    5. Santur, Y., M. KARAKÖSE, and E. Akin, An adaptive fault diagnosis approach using pipeline implementation for railway inspection. Turkish Journal of Electrical Engineering & Computer Sciences, 2018. 26(2): p. 987-998.
    6. 梁文愷, 應用電腦視覺變位量測於邊坡滑動之研究, in 土木工程學系所. 2017, 國立中興大學: 台中市. p. 84.
    7. 陳漢倫, 3D電腦視覺變位觀測系統用於集水井之變形觀測, in 土木工程學系所. 2017, 國立中興大學: 台中市. p. 95.
    8. Rahimi-Ajdadi, F., et al., Development of a novel machine vision procedure for rapid and non-contact measurement of soil moisture content. Measurement, 2018. 121: p. 179-189.
    9. 張宇樑, 應用電腦視覺於排水廊道變形監測之研究, in 土木工程學系所. 2017, 國立中興大學: 台中市. p. 79.
    10. Alimuin, R., E. Calilung, and E. Dadios. Implementation of machine vision as an input to a fuzzy water level control system. in Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment and Management (HNICEM), 2017 IEEE 9th International Conference on. 2017. IEEE.
    11. Iglesias, C., J. Martínez, and J. Taboada, Automated vision system for quality inspection of slate slabs. Computers in Industry, 2018. 99: p. 119-129.
    12. 李穎亮, 影像處理技術於混凝土結構裂縫自動化偵測, in 土木與工程管理學系碩士班. 2012, 國立金門大學: 金門縣. p. 79.
    13. Faula, Y., S. Bres, and V. Eglin. FLASH: a new key structure extraction used for line or crack detection and image matching. in International Conference on Computer Vision Theory and Applications (Visapp 2018). 2018.
    14. Xinman, Z., et al. Steel bars counting method based on image and video processing. in Intelligent Signal Processing and Communication Systems (ISPACS), 2017 International Symposium on. 2017. IEEE.
    15. Hamledari, H., et al. UAV-Enabled Site-to-BIM Automation: Aerial Robotic-and Computer Vision-Based Development of As-Built/As-Is BIMs and Quality Control. in Construction Research Congress 2018. 2017.
    16. 陳頎夫, 應用電腦視覺於隧道內定位之初探, in 營建工程系. 2012, 國立臺灣科技大學: 台北市. p. 84.
    17. 趙智凡, 應用多視立體及運動回復結構技術之三維場景重構與擴增實境, in 土木工程學系所. 2012, 國立中興大學: 台中市. p. 86.
    18. 劉廣德, 應用電腦視覺技術實作以2D影像資訊建立3D建物模型於iOS平台之研究, in 土木工程學系. 2012, 國立交通大學: 新竹市. p. 106.
    19. Reinoso, J., et al., Cartography for Civil Engineering Projects: Photogrammetry Supported by Unmanned Aerial Vehicles. Iranian Journal of Science and Technology, Transactions of Civil Engineering, 2018. 42(1): p. 91-96.
    20. 陳佶宏, 應用電腦視覺於道路標線品質評鑑之研究, in 土木與生態工程學系. 2014, 義守大學: 高雄市. p. 102.
    21. 賴譯民, 貼磚整齊度自動評估系統之研究, in 土木工程系所. 2015, 國立交通大學: 新竹市. p. 140.
    22. Mneymneh, B.E., M. Abbas, and H. Khoury, Automated hardhat detection for construction safety applications. Procedia engineering, 2017. 196: p. 895-902.
    23. Fang, Q., et al., Computer vision aided inspection on falling prevention measures for steeplejacks in an aerial environment. Automation in Construction, 2018. 93: p. 148-164.
    24. Kaehler, A., G. Bradski, and 賴屹民, OpenCV 3 學習手冊. 2017.
    25. Lucas, B.D. and T. Kanade, An iterative image registration technique with an application to stereo vision. 1981.
    26. Kaehler, A. and G. Bradski, Learning OpenCV 3: computer vision in C++ with the OpenCV library. 2016: " O'Reilly Media, Inc.".
    27. OpenCV Library. Available from: https://opencv.org/.
    28. Tomasi, C. and R. Manduchi. Bilateral filtering for gray and color images. in Computer Vision, 1998. Sixth International Conference on. 1998. IEEE.
    29. Canny, J., A computational approach to edge detection. IEEE Transactions on pattern analysis and machine intelligence, 1986(6): p. 679-698.
    30. Harris, C., A combined corner and edge detector. Proc. 4th Alvey Vision Conference, 1988: p. 147-151.
    31. Jianbo, S. and C. Tomasi. Good features to track. in 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. 1994.
    32. Rosten, E. and T. Drummond, Machine learning for high-speed corner detection, in Proceedings of the 9th European conference on Computer Vision - Volume Part I. 2006, Springer-Verlag: Graz, Austria. p. 430-443.
    33. Mair, E., et al. Adaptive and generic corner detection based on the accelerated segment test. in European conference on Computer vision. 2010. Springer.
    34. Lowe, D.G., Distinctive Image Features from Scale-Invariant Keypoints. International Journal of Computer Vision, 2004. 60(2): p. 91-110.
    35. Bay, H., T. Tuytelaars, and L. Van Gool. SURF: Speeded Up Robust Features. in Computer Vision – ECCV 2006. 2006. Berlin, Heidelberg: Springer Berlin Heidelberg.
    36. Leutenegger, S., M. Chli, and R.Y. Siegwart. BRISK: Binary Robust invariant scalable keypoints. in 2011 International Conference on Computer Vision. 2011.
    37. Rublee, E., et al. ORB: An efficient alternative to SIFT or SURF. in Computer Vision (ICCV), 2011 IEEE international conference on. 2011. IEEE.
    38. Nistér, D. and H. Stewénius. Linear time maximally stable extremal regions. in European Conference on Computer Vision. 2008. Springer.
    39. Alcantarilla, P.F., A. Bartoli, and A.J. Davison. KAZE features. in European Conference on Computer Vision. 2012. Springer.
    40. Alcantarilla, P.F., Fast explicit diffusion for accelerated features in nonlinear scale spaces. 2011.
    41. Calonder, M., et al. Brief: Binary robust independent elementary features. in European conference on computer vision. 2010. Springer.
    42. Feature Matching. Available from: https://docs.opencv.org/3.0-beta/doc/py_tutorials/py_feature2d/py_matcher/py_matcher.html#flann-based-matcher.

    QR CODE