跳到主要內容

臺灣博碩士論文加值系統

(44.201.99.222) 您好!臺灣時間:2022/12/03 14:38
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:賴泰華
研究生(外文):Tai-Hua Lai
論文名稱:不完整三維點群之特徵擷取與結構差補
論文名稱(外文):Feature Extraction and Structure Interpolation of Incomplete 3D Scattered Data
指導教授:王中行王中行引用關係
指導教授(外文):Chung-Shing Wang
口試委員:蕭世文王偉華莊漢東周重石
口試委員(外文):Shih-Wen HsiaoWei-Hua WangHan-Tung ChuangChorng-Shyr Jou 
學位類別:碩士
校院名稱:東海大學
系所名稱:工業設計學系
學門:設計學門
學類:產品設計學類
論文種類:學術論文
論文出版年:2010
畢業學年度:98
語文別:中文
論文頁數:89
中文關鍵詞:三次元數位量測破面修補核算法自組織映射網路隱函曲面
外文關鍵詞:3D Digital MeasurementHole-FillingKernel MethodSelf-Organizing MapsImplicit Surface
相關次數:
  • 被引用被引用:0
  • 點閱點閱:231
  • 評分評分:
  • 下載下載:25
  • 收藏至我的研究室書目清單書目收藏:1
逆向工程技術是近年來快速發展的一項設計模式,其技術流程是先採用探針、雷射或光學方式,將物體表面資料座標化,藉此分析物體形狀幾何,並透過CAD系統重現曲面,最後將資料送至RP設備做快速成型。然而,進行三次元數位量測過程,會因物體表面材質及形狀因素影響,造成量測資料的不完整,破面結構因此產生。由於CAD系統其演算法限制,不完整的點群模型將導致後續資料處理上困難。因此,本研究提出一個以特徵幾何為基礎的核算內差方式,透過隱函曲面 (Implicit Surface) 技術來重建資料模型,達到結構差補的效果。
研究上,藉由結合自組織映射網路 (Self-Organizing Maps, SOM) 及核算理論 (Kernel Methods),以三維點群資料的幾何重建為方向,從事深入的理論探討及程式模擬,研擬的課題包括 「SOM網路之特徵擷取」 及 「核算法於隱函曲面重建」 等子題。文中所提出的幾何重建模式,除了能有效縮減系統於模型存取的資料量外,並同時確保重建模型的幾何完整性,使得後續於模型的處理上,更能順利的進行。本論文之具體研究成果與貢獻:
(1) 驗證SOM網路於完整及不完整三維點群資料之特徵擷取效果。
(2) 驗證核算法於特徵點群資料之隱函曲面重建效果。
(3) 比較薄板樣條及隱函曲面的重建方式於破面函數模型之結構差補效果。
(4) 比較雙諧波、三諧波及多元二次型核函數於隱函曲面重建的效果。
(5) 透過Golf及Kitty兩組實際案例的重建,由實例驗證學理。

The technique of reverse engineering (RE) is one of the design methods which develop rapidly in recent year. RE process is to get the surface coordinates of objects through probe, laser or optical mechanism first. It obtains geometric points cloud data of object shape, and applies CAD system to reconstruct the curved surface. Afterwards, it transfers coordinate data into STL triangular mesh and builds up physical model by rapid prototyping (RP) machine. However, during the process of 3D laser scanning, the incomplete scan data causes holes due to the factors of different material, shape or ambient light. Because of the limitation of reverse CAD system algorithm, the holes in STL meshes structure will cause difficulties in data reuse or inaccurate processing result.
For this purpose, an interpolation mechanism with kernel method based on geometric features is presented. More specifically, SOM (Self-Organizing Maps) network is used to fetch the geometric features of the model, and kernel method is introduced to reconstruct the 3D surface represented by an implicit function. We therefore crystallize our research goal that aimed at an in-depth investigation of several related domestic and international research in the scope of implicit surface, with SOM network and kernel method, both in theory and experiment. The method presents at this research can result in significant data compression and speed improvement at the reconstruction procedure. Results and contributions in this paper are shown in the following:
(1) To investigate the validity of using SOM network to extract features at complete and incomplete 3D scattered data surface.
(2) To verify the validity of using kernel method to construct implicit surface based on feature data.
(3) To compare thin-plate approach with implicit surface approach in feature data interpolation at incomplete model.
(4) To compare three types of kernel function, biharmonic, triharmonic and multiquadric, to reconstruct the implicit surface.
(5) Two case studies in golf and kitty models are implemented to verify by Matlab program simulation

中文摘要 I
英文摘要 II
誌謝 III
目錄 IV
圖目錄 VI
表目錄 X
第一章 緒論 1
1.1研究背景 1
1.2文獻回顧 2
1.2.1幾何結構重建 3
1.2.2隱函曲面 5
1.2.3 類神經網路 7
1.3研究目的 9
1.4研究方法 11
1.5論文架構 13
第二章 特徵擷取 15
2.1 類神經網路的學習與記憶 15
2.2 SOM網路架構 17
2.3 SOM演算法 22
2.4 點群特徵擷取 24
2.4.1 網路初始化 26
2.4.2 迪氏三角網格 28
第三章 結構差補 30
3.1 隱函曲面 30
3.2 函數近似 33
3.3 事例基學習 36
3.4 核算法 39
3.5 點群結構差補 41
3.5.1 頂點法向量 42
3.5.2 零值約束及法向量約束 43
3.5.3 建立核矩陣 45
3.5.4 求解隱函曲面方程式 46
第四章 系統架構 48
4.1 點群資料的特徵擷取 48
4.2 點群資料的結構差補 61
4.3 分析與討論 69
第五章 實例驗證 72
5.1 Golf模型的驗證 72
5.2 Kitty模型的驗證 76
第六章 結論與建議 80
參考文獻 85

[1] Anderson, J. A. (1972). A simple neural network generating an interactive memory. Mathematical Biosciences (14), pp. 197-220.
[2] Barequeta, G., & Sharir, M. (1995). Filling gaps in the boundary of a polyhedron. Computer Aided Geometric Design , 12, pp. 207-229.
[3] Barhak, J., & Fischer, A. (2002). Adaptive reconstruction of freeform objects with 3D SOM neural network grids. Computers & Graphics (26), pp. 745–751.
[4] Barhak, J., & Fischer, A. (2001). Parameterization for reconstruction of 3D freeform objects from laser-scanned data based on a PDE method. The Visual Computer (17), pp. 353-369.
[5] Berg, M., Cheong, O., Kreveld, M., & Overmars, M. (2008). Computational Geometry: Algorithms and Applications (3 ed.). Springer.
[6] Carr, J. C., Beatson, R. K., Cherrie, J. B., Mitchell, T. J., Fright, W. R., McCallum, B. C., et al. (2001). Reconstruction and representation of 3D objects with radial basis functions. ACM SIGGRAPH 2001 , pp. 12-17.
[7] Carr, J. C., Fright, W. R., & Beatson, R. K. (1997). Surface interpolation with radial basis functions for medical imaging. IEEE Trans. Medical Imaging , 16 (1), pp. 96-107.
[8] Chang, D. Y., & Chang, Y. M. (2002). A freeform surface modelling system based on laser scan data for reverse engineering. Int J Adv Manuf Technol , 20 (1), pp. 9-19.
[9] Chapman, C. D., Saitou, K., & Jakiela, M. J. (1994). Genetic algorithms as an approach to configuration and topology design. Journal of Mechanical Design (116), pp. 1005-1012.
[10] Chen, Y. H., & Wang, Y. Z. (1999). Genetic algorithms for optimized re-triangulation in the context of reverse engineering. Computer Aided Design (31), pp. 261-271.
[11] Chen, Y. H., Ng, C. T., & Wang, Y. Z. (1999). Generation of an STL file from 3D measurement data with user-controlled data reduction. The International Journal of Advanced Manufacturing Technology (15), pp. 127-131.
[12] Chua, C. K., Leong, K. F., & Lim, C. S. (2003). Rapid Prototyping: Principles and Applications in Manufacturing (2 ed.). World Scientific.
[13] Fritzke, B. (1995). A growing neural gas network learns topologies. Adavances in Neural Information Processing System (pp. 625-632). Cambridge, MA: MIT Press.
[14] Furrens, K. K. (1999). Standards for the rapid prototyping industry. Rapid Prototyping Journal (5), pp. 169-178.
[15] Gelas, A., Valette, S., Prost, T., & Nowinski, W. L. (2009). Variational implicit surface meshing. Computers & Graphics , 33 (3), pp. 312-320.
[16] Gieng, T. S., Joy, K. I., Schussman, G. L., & Trotts, I. J. (1998). Constructing hierarchies for triangle meshes. IEEE transactions on Visualization and Computer Graphics (4), pp. 145–161.
[17] Haykin, S. (1999). Neural Network: A Comprehensive Foundation (2 ed.). New Jersey: Prentice Hall.
[18] Hebb, D. (1949). The Organization of Behavior. New York: Wiley.
[19] Holdstein, Y., & Fischer, A. (2008). Three-dimensional surface reconstruction using meshing growing neural gas (MGNG). Visual Comput (24), pp. 295-302.
[20] Hopfield, J. J. (1982). Neural networks and physical systems with emergent collective computational properties. Proceedings of the National Academy of Sciences (79), pp. 2554-2558.
[21] Hur, S. M., Kim, H. C., & Lee, S. H. (2002). STL file generation with data reduction by the delaunay triangulation. International Journal of Advanced Manufacturing Technology, (19), pp. 669-678.
[22] Kalvin, A. D., & Taylor, ,. R. (1996). Superfaces : Polygonal mesh simplification with bounded error. IEEE Computer Graphics and Applications , 3 (16), pp. 64-77.
[23] Kohonen, T. (1972). Correlation matrix memories. IEEE Transaction on Computers (21), pp. 353-359.
[24] Kohonen, T. (1990). The self-organizing map. Proc. of IEEE , 9 (78), pp. 1464-1480.
[25] Kotsiantis, S. B. (2007). Supervised machine learning: A review of classification techniques. Informatica , 31, pp. 249-268.
[26] Kumar, G. S., Kalra, P. K., & Dhande, S. G. (2004). Curve and surface reconstruction from points an approach based on self-organizing maps. Applied Soft Computing (5), pp. 55-66.
[27] Kumar, S. (2005). Neural Networks : A Classroom Approach. McGraw-Hill.
[28] Kumar, V., & Dutta, D. (1997). An assessment of data formats for. Advances in Engineering Software (28), pp. 151-164.
[29] Lee, K. H., Woo, H., & Suk, T. (2001). Data reduction methods for reverse engineering. Int J Adv Manuf Technol , pp. 735-743.
[30] Lorensen, W. E., & Cline, H. E. (1987). Marching cubes: A high resolution 3D surface construction algorithm. Computer Graphics , 21 (4), pp. 163-169.
[31] Mark, E., & Steve, M. (2001). Triangle-mesh simplification using error polyhedra. Proceedings of EuroGraphics UK 2001 (pp. 3-5). London: University College London.
[32] Martin, R. R., Stroud, I. A., & Marshall, A. D. (1997). Data reduction for reverse engineering. (Goodman, Ed.) The Mathematics of Surface VII .
[33] Mitchell, T. M. (1997). Machine Learning. New York: McGraw-Hill.
[34] Nakanishi, Y. (2001). Application of homology theory to topology optimization of three-dimensional structures using genetic algorithm. Comput. Methods Appl. (190), pp. 3849-3863.
[35] Negnevitsky, M. (2001). Artificial Intelligence: A Guide to Intelligent Systems (2 ed.). Boston: Addison-Wesley.
[36] Pham, D. T., & Gault, R. S. (1998). A comparison of rapid prototyping technologies. International Journal of Machine chine Tools and Manufacture (38), 1257-1287.
[37] Rogers, D. F., & Adams, J. A. (1990). Mathematical elements for computer graphics. McGraw-Hill.
[38] Rosenblatt, F. (1958). The perceptron : A probabilistic model for information storage and organization in the brain. Psychological Review (65), pp. 386-408.
[39] Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986). Learning representations by back-propagating error. Nature (323), pp. 533-536.
[40] Savchenko, V. V., Pasko, A. A., Okunev, O. G., & Kunii, T. L. (1995). Function respresentation of solids reconstructed from scattered surface points and contours. Computer Graphics Forum , 14 (4).
[41] Shawe-Taylor, J., & Cristianini, N. (2004). Kernel Methods for Pattern Analysis. New York: Cambridge University Press.
[42] Szilvasi-Nagy, M., & Matyasi, G. (2003). Analysis of STL files. Mathematical and Computer Modeling , 38, pp. 945-960.
[43] Tekumalla, L. S., & Cohen, E. (2003). A hole-filling algorithm for triangular meshes. School of Computing. USA: University of Utah.
[44] Tian, X. D., Wang, Y. X., Zhou, X. H., & Ruan, X. (2002). Mesh simplification based on super-face and genetic algorithm in reverse engineering. International Journal of Advanced Manufacturing Technology (20), pp. 303-312.
[45] Tobor, I., Reuter, P., & Schlick, C. (2004). Efficient reconstruction of large scattered geometric datasets using the partition of unity and radial basis functions. Journal of WSCG , 12 (1-3), pp. 467-474.
[46] Turk, G., & O'Brien, J. (2002). Modelling with implicit surfaces that interpolate. ACM Transactions on Graphics , 21 (4), pp. 855-973.
[47] Turk, G., & O'Brien, J. (1999). Shape transformation using variational implicit functions. SIGGRAPH 99 , pp. 335-342.
[48] Turk, G., Dinh, H. Q., O'Brien, J., & Yngve, G. (2001). Implicit surfaces that interpolate. Shape Modelling International 2001, (pp. 62-71.). Genova,Italy.
[49] Vapnik, V. N. (1999). The Nature of Statistical Learning Theory (2 ed.). New York: Springer.
[50] Wang, C. S., & Chang, T. R. (2007). Re-triangulation in STL meshes for rapid prototyping. International Journal of Advanced Manufacturing Technology (37), pp. 770-781.
[51] Yngve, G., & Turk, G. (1999). Greating smooth implicit surfaces from polygonal meshs. Georgia Institute of Technology.
[52] 吳宗憲 (2002),以多重RBF函數建構精簡3D模型表示法與動態點繪圖演算法,碩士論文,國立成功大學資訊工程研究所。
[53] 楊偉倫 (2007),三角網格錯誤偵測及特徵線線萃取,碩士論文,國立台灣大學機械工程研究所。
[54] 陳俊彥 (2000),利用3D 多重掃瞄資料建構多面體架構之實體模型,碩士論文,國立中正大學機械工程研究所。
[55] 黃偉銓 (2006),以放射基礎函數及漸進式物件壓縮建立精簡三維物件模型,碩士論文,國立中山大學資訊工程研究所。


連結至畢業學校之論文網頁點我開啟連結
註: 此連結為研究生畢業學校所提供,不一定有電子全文可供下載,若連結有誤,請點選上方之〝勘誤回報〞功能,我們會盡快修正,謝謝!
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top