臺灣博碩士論文加值系統

本文主要的研究目標是利用電腦視覺系統與影像處理的 技術來辨識模具表面點群之曲面特徵，並依據幾何的推 理重建模型CAD資料。研究的內容包括了立體視覺量測技術，曲面的區域分割，曲面的辨識及曲面CAD 資料格式 的重建。利用影像編碼技術，可將三維的空間點群轉換 成二維的影像，利用影像的邊界搜尋、區域分割與邊界 迴路追跡，各個區域曲面與邊界上的點群資料可以被獨 立分割出來。根據區域主軸之兩筆截痕點群的曲線擬合 結果，各個區域曲面可以被辨識為平面、旋轉面或自由 曲面三種曲面類型，並依據不同類型的曲面進行重建。 在重建的過程中，可藉由辨識鄰接曲面的相交條件判斷 公用邊界在空間中是否存在一垂直壁面。並依據幾何完 整性的推理建立合理的曲面邊界條件，最後將重建的曲 面以IGES 格式輸出。

The aim of this research is to recognize the geometrical features of a mold surface via the computer vision and image process techniques. Based on the geometrical reasoning, 3D CAD model of the molds are reconstructed. This study also includes stereo vision depth measurement system. By employing the image encoding skill, we can transform the 3D point cloud into a 2D image format. The point clouds in each individual surface and its corresponding surface edges are separated by using boundary search, image segmentation and loop tracing in the image format. The surface is recognized by curve fitting of two sectional depth data through the major axes of each regional surface. And the surface can then be recognized as plane, surface of revolution, and freeform surface according to the feature of recognized curves. In the process of reconstruction, the boundaries of surfaces can be judged whether there exist a wall surface according to the intersection of the neighboring surface. The completion of surface model is done by geometric reasoning from the boundary of surface. The reconstructed surface is outputted as IGES format for further usages.

摘要................................................................................................................... I
ABSTRACT..................................................................................................... II
誌謝.................................................................................................................III
目錄.................................................................................................................IV
圖目錄..........................................................................................................VIII
表目錄..........................................................................................................XIII
第一章 序論.....................................................................................................1
1.1 研究動機與目的........................................................................................1
1.2 文獻回顧....................................................................................................2
1.3 論文綱要....................................................................................................4
第二章 量測系統.............................................................................................5
2.1 立體視覺系統............................................................................................5
2.1.1 硬體介紹.................................................................................................6
2.1.2 系統架設.................................................................................................7
2.2 主動式視覺系統模型與校正....................................................................9
2.2.1 主動式視覺系統模型...........................................................................10
2.2.2 系統參數校正.......................................................................................12
2.3 空間深度計算..........................................................................................13
2.4 立體視覺系統性能..................................................................................16
第三章 單一曲面的類型辨識.......................................................................20
3.1 影像編碼..................................................................................................20
3.2 影像邊界搜尋與區域分割......................................................................22
3.2.1 影像邊界搜尋.......................................................................................22
3.2.2 影像區域分割.......................................................................................24
3.2.3 區域迴路追跡.......................................................................................27
3.3 影像迴路分割..........................................................................................27
3.4 曲線擬合與辨識......................................................................................31
3.4.1 曲線擬合...............................................................................................31
3.4.1.1 直線擬合............................................................................................31
3.4.1.2 圓弧擬合............................................................................................32
3.4.1.3 二次曲線擬合....................................................................................32
3.4.1.4 自由曲線擬合....................................................................................35
3.4.2 曲線資料結構.......................................................................................37
3.4.2.1 直線資料結構....................................................................................37
3.4.2.2 圓弧資料結構....................................................................................38
3.4.2.3 二次曲線資料結構............................................................................39
3.4.2.4 B-spline 曲線資料結構.....................................................................40
3.4.3 截痕曲線之辨識...................................................................................41
3.5 曲面辨識..................................................................................................45
第四章 表面邊界的幾何推理.......................................................................48
4.1 計算及修正平面方程式..........................................................................48
4.2 公用邊界的推理......................................................................................51
4.2.1 平面和旋轉面.......................................................................................52
4.2.1.1 旋轉面之建立....................................................................................53
4.2.1.2 修正生成線與外迴路之方程式........................................................55
4.2.2 平面和平面...........................................................................................58
4.3 表面完整性的建立..................................................................................63
4.3.1 合併連續邊界.......................................................................................63
4.3.2 統一公用邊界方程式...........................................................................66
4.3.3 邊界端點的計算...................................................................................67
4.3.3.1 直線和直線........................................................................................68
4.3.3.2 直線和圓弧........................................................................................68
4.4 壁面的推理..............................................................................................71
4.4.1 平面.......................................................................................................72
4.4.2 圓柱面...................................................................................................77
第五章 曲面重建與資料格式輸出...............................................................78
5.1 自由曲面的建構......................................................................................78
5.2 資料格式輸出..........................................................................................81
5.2.1 平面資料結構.......................................................................................82
5.2.2 旋轉面資料結構...................................................................................85
5.2.3 自由曲面資料結構...............................................................................88
5.3 模具重建結果..........................................................................................90
5.3.1 No.1 模具..............................................................................................90
5.3.2 No.2 模具..............................................................................................91
5.3.3 No.3 模具..............................................................................................92
5.3.4 No.4 模具..............................................................................................93
5.4 模具重建誤差分析..................................................................................96
5.4.1 No.1 模具..............................................................................................98
5.4.2 No.2 模具..............................................................................................99
5.4.3 No.3 模具............................................................................................100
第六章 結論與建議.....................................................................................101
6.1 研究成果................................................................................................101
6.2 討論與建議............................................................................................102
參考文獻.......................................................................................................105
APPENDIXA IGES 資料結構定義.............................................................109
自述...............................................................................................................115

[1] D. G. Hakala, R. C. Hillyard, P. F., Malraison, and B. F. Nource,“Natural Quadrics in Mechanical Design,” SIGGRAPH/81, Seminar on Solid modeling, Dallas, Texas, 1981.
[2] P. J. Besl, N. D. McKay, “A method for registration of 3-D shapes,” IEEE Transactions on Pattern Analysis and MachineIntelligence, Vol. 14, No. 2, pp. 239-256,1992.
[3] T. J. Fan, G. Medioni, and R. Nevatia, 1987, “Segmented Description of 3- D Surface,” IEEE Journal of Robotics and Automation, Vol. RA-3, No.6, Dec. pp. 527-538.
[4] P. J. Besl, and R. C. Jain, 1986, “Invariant Surface Characteristics for 3D Object Recognition in Range Images,” Computer Vision, Graphics and Image Processing, Vol. 33, pp. 33-80.
[5] 姚宏宗，邱顯智，陳信全，1996，“逆向工程-點資料前置處理與曲面重建”，中國機械工程學會第十三屆全國學術研討會論文集（設計），第119~126 頁。
[6] 許來興，游星宇，1997，“逆向工程曲面重建過程中點資料區域分割之方法”，中國 機械工程學會第十四屆全國學術研討會論文集（製造與生產），第121~127 頁。
[7] C. Bradley, G. W. Vickers, and M. Milroy, 1994, “Reverse Engineering of Quadric Surfaces Employing Three-Dimensional Laser Scanning,” Journal of Engineering Manufacture, Vol. 208, pp.21-28.
[8] M. J. Tsai, J. H. Hwung, T. F. Lu, and H. Y. Hsu, 2006,“Recognition of Quadratic Surface of Revolution Using a Robotic Vision System,” International Journal of Robotics and Computer Integrated Manufacturing, Vol. 22, Apr. 2006, pp.134-143.
[9] 黃佳豪，“以特徵為基礎的交談式電腦視覺模具幾何辨識系統”，國立成功大學機械 工程研究所碩士論文，中華民國89 年。
[10] W. Wan, J. A. Ventura, “Segmentation of planar curves into straight-line segments andelliptical arcs,” Graphical Models and Image Processing, Vol. 59, No. 6, pp. 484-494, 1997.
[11] A. Carmona-Poyato, N. L. Fernandez-Garcia, R. Medina-Carnicer, F. J. Madrid-Cuevas, “Dominant point detection: A new proposal,”Image and Vision Computing, Vol. 23, Issue 14, pp. 1264-1276, 2005.
[12] S. H. Bae, B. K. Choi, “NURBS surface fitting using orthogonal coordinate transform rapid product development,” Computer-Aided Design, Vol. 34, No. 10, pp. 683-690, 2002.
[13] J. Y. Lai, W. D. Ueng, “Reconstruction of surfaces of revolution
from measured points,” Computers in Industry Vol. 41 , pp.147–161, 2000.
[14] H. Park, K. Kim, “Smooth surface approximation to serial cross- sections,” Computer-Aided Design, Vol. 28, No. 12, pp.995-1005, 1996.
[15] G. P. Stein, “Lens Distortion Calibration Using Point Correspondences,” IEEE Proceedings of Computer Society Conference on Computer Vision and Pattern Recognition, 1997.
[16] 洪全成，“光學式三維量測系統之建立及量測效能評估法則之研究”，國立成功大學機械工程研究所博士論文，中華民國94年。
[17] J. Gühring, 2000, “Dense 3-D Surface acquisition by Structured Light Using off-the-shelf Components,” Report of University of Stuttgart, Germany.
[18] A. M. Bronstein, M. M. Bronstein, E. Gordon, R. Kimmel,“High-Resolution Structured Light Range Scanner with Automatic Calibration,” Technion - Computer Science Department –Technical Report CIS-2003-06 – 2003.
[19] 蔡明俊，“影像編碼壓縮方法及具人體特徵的資料結構”，中華民國專利，發明第 195369 號，2004。
[20] 蔡明俊，韋志昇，魏緒強，“應用影像處理技術於模具邊界之幾何特徵辨識”，中華民國自動化科技學會第十屆全國自動化科技研討會，論文集第二冊，第801-808 頁，中華 民國86 年。
[21] N. J. Naccache, R. Shinghal, “SPTA: A proposed algorithm for thinning binary patterns,” IEEE Transactions on System, Man and Cybernetics, Vol. SMC- 14, No. 3, pp. 409-418, 1984.
[22] A. C. Ugural, S. K. Fenster, 1995, “Advanced Strength and Applied Elasticity,” Prentice-Hall, New Jersey.
[23] K. Sugimoto, F. Tomita, “Boundary segmentation by detection of corner, inflection and transition points,” IEEE Workshop on Visualization and Machine Vision, Proceedings, pp. 13-17, 1994.
[24] 安乃駿，“三維模型幾何特徵辨識與重建之研究”，國立成功大學機械工程研究所碩士論文，中華民國93年。
[25] A. Fitzgibbon, M. Pilu, R. B. Fisher, “Direct least square fitting of
ellipses,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 21, No. 5, pp. 476-480, 1991.
[26] L. Piegl, W. Tiller, “The NURBS Book, 2nd Edition,”Springer-Verlag, 1997.
[27] K. Reed, 1991, “Initial Graphic Exchange Specification, Version 5.1,” National Institute of Standard and Technology.
[28] R. C. Gonzalez, R. E. Woods, 1992, “Digital Image Processing, Addison- Wesley,” U. S. A.
[29] Spiegel, R. Murray, J. Liu., Mathematical Handbook of Formulas and Tables, 2nd Edition, McGRAW-HILL, 1999.
[30] Jones, M. Richard, Introduction to MFC programming with Visual C++, Prentice Hall PRT, 2000.