(3.238.186.43) 您好!臺灣時間:2021/03/01 09:12
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:::

我願授權國圖
: 
twitterline
研究生:賴岳宏
研究生(外文):Yueh-Hung Lai
論文名稱:利用演化式計算做最佳化之研究
論文名稱(外文):Research on Optimization Using Evolutionary Computation
指導教授:唐政元
指導教授(外文):Cheng-Yuan Tang
學位類別:碩士
校院名稱:華梵大學
系所名稱:資訊管理學系碩士班
學門:電算機學門
學類:電算機一般學類
論文種類:學術論文
論文出版年:2005
畢業學年度:93
語文別:中文
論文頁數:55
中文關鍵詞:演化式計算基因演算法田口方法基礎矩陣
外文關鍵詞:Evolutionary ComputationGenetic algorithm (GA)Taguchi MethodFundamental Matrix
相關次數:
  • 被引用被引用:6
  • 點閱點閱:239
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:35
  • 收藏至我的研究室書目清單書目收藏:0
演化式計算(Evolutionary Computation)包括演化式規劃(Evolutionary Programming)、演化式策略(Evolution Strategy)、基因演算法(Genetic Algorithm)、基因規劃(Genetic Programming)等,已經被廣泛的使用來解決最佳化的問題。
本論文中,我們針對一場景進行拍攝,在未校正相機情況下,估測包含影像間投影幾何資訊的基礎矩陣(Fundamental matrix)。本論文的研究工作,是利用演化式計算的快速、精確、穩定等特性來計算複雜的基礎矩陣公式以尋得最佳化後的基礎矩陣。
我們引用田口方法(Taguchi Method)與基因演算法(Genetic Algorithm)來估測基礎矩陣,並改良了HTGA(Hybrid Taguchi-Genetic Algorithms)與IMOEA(Intelligent Multiobjective Evolutionary Algorithms)使其在運算基礎矩陣時可以更強健與可靠。為了得到更好的結果,我們設計了前置實驗並測試出使用Bucket母體、Headless Chicken Test以及中位數選擇法這三個組合能有效排出Outlier並尋找到較佳的基礎矩陣,而我們還與GA、HTGA以及我們改良後的HTGA做比較,而實驗結果我們改良後的HTGA能夠找到較佳的基礎矩陣以及有效排除Outlier。而IMOEA主要是針對多個目標方程式(幾何距離與代數距離)來進行基礎矩陣的最佳化,在實驗結果上,雖然能取得較佳的基礎矩陣,但其演化速度太慢不符合實驗成本。最後我們採取這兩種演算法來提供較正確的基礎矩陣,以作為重建3D的重要依據。
Evolutionary computation, such as evolutionary programming, evolution strategy, genetic algorithm, genetic programming etc., has already been widely used for solving the problem of the optimization.
In this paper, we take a scene in uncalibrated camera cases, estimated the fundamental matrix which includes the geometry information of the projection among the images. Research work of this paper is using characteristics of evolutionary computation such as fast, accurate and steady to calculate complicated fundamental matrix in order to look for the fundamental matrix after the optimization.
In this paper, we try to apply both the Taguchi method (orthogonal array) and GA (Genetic Algorithm) to estimate the fundamental matrix in computer vision. Modification of HTGA (Hybrid-Taguchi genetic algorithm) and IMOEA (Intelligent multiobjective evolutionary algorithm) are used to obtain robust and reliable solution. In order to obtain better results, we design a pre-test experiment in order to obtain an optimal combination: bucket population + headless chicken test + evaluation using median. In estimating the fundamental matrix, we use two methods, such as using an objective function and multiobjective functions. For experiments using an objective function, we compare three algorithms (modification of HTGA, HTGA and GA), and obtain that our proposed method (modification of HTGA) performs best. For experiments using multiobjective function (geometric distance and algebraic distance), experimental results reveal that using IMOEA can lead to correct solutions. However, it needs 8.5 hours to achieve it. In the future, we need to analyze the phenomenon and performance of IOMEA. Furthermore, we need to find some methods to speed up its computation power.
目錄


致 謝...............................................................................................................................I
摘 要.............................................................................................................................II
Abstract....................................................................................................................... IV
目 錄............................................................................................................................VI
圖 目 錄...................................................................................................................VIII
表 目 錄......................................................................................................................XI
一、簡介....................................................................................................................... 1
二、演化式計算........................................................................................................... 3
三、Taguchi Method..................................................................................................... 5
3.1直交表.......................................................................................................5
3.2訊號雜音比...............................................................................................7
3.3 SNR的公式分類......................................................................................7
3.3.1 望目特性......................................................................................7
3.3.2 望小特性......................................................................................7
3.3.3 望大特性......................................................................................8
四、Hybrid Taguchi-Genetic Algorithm....................................................................... 9
五、Intelligent Multiobjective Evolutionary Algorithms………………....................14
5.1 IGC.......................................................................................................14
5.2 GPSIFF.................................................................................................16
5.3 IMOEA演化流程.................................................................................16
六、基礎矩陣與八點演算法..................................................................................... 18
6.1 基礎矩陣..............................................................................................18
6.2 八點演算法..........................................................................................19
七、演化流程設計...................................................................................................... 21
八、實驗結果與分析.................................................................................................. 25
8.1 前置實驗...................................................................................................... 25
8.2 後置實驗...................................................................................................... 29
8.2.1 An objective function......................................................................... 29
8.2.1 Multiobjective function...................................................................... 36
九、結論..................................................................................................................... 40
參考文獻..................................................................................................................... 41
參考文獻


[1]M. Z. Brown, D. Burschka, and G. D. Hager, “Advances in Computational Stereo,” IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 25, No. 8, Aug. 2003, pp. 993-1008.
[2]M. Carcassoni, and E. R. Hancock, “Correspondence Matching with Modal Clusters,” IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 25, No. 12, Dec. 2003, pp. 1609-1615.
[3]J. Chai and S. D. Ma, “Robust Epipolar Geometry Estimation Using Genetic Algorithm,” Pattern Recognition Letters, Vol. 19, 1998, pp. 829-838.
[4]L. Davis, Handbook of Genetic Algorithms, Van Nostrand Reinhold, 1996.
[5]D.A. Forsyth, Jean Ponce, Computer Vision A Modern Approach, Prentice Hall, 2003.
[6]B. Georgescu, and P. Meer, “Point Matching under Large Image Deformations and Illumination Changes,” IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 26, No. 6, Jun. 2004, pp. 674-688.
[7]S.Y. Ho, L.S. Shu, and J.H. Chen, “Intelligent Evolutionary Algorithms for Large Parameter Optimization Problems,” IEEE Trans. Evolutionary Computation, Vol.8, No.6, December 2004, pp. 522-541.
[8]T. Jones, Evolutionary Algorithms, Fitness Landscapes and Search, Ph.D Dissertation, University of New Mexico, May 1995.
[9]Y.W. Leung and Y. Wang, “An orthogonal genetic algorithm with quantization for global numerical optimization,” IEEE Trans. Evolutionary Computation, Vol. 5, Feb. 2001, pp. 41–53
[10]J. M. Renders and S. P. Flasse, “Hybrid Methods Using Genetic Algorithms for Global Optimization,” IEEE Trans. Systems, Man, and Cybernetics-part B: Cybernetics, Vol. 26, No. 2, April 1996, pp. 243-258.
[11]C. T. Tang, H. L. Chou, Y. H. Yeh, M. W. Lin, S. T. Tzou, and Y. H. Ding, “Feature Selection for Estimating Fundamental Matrices using Disparity Constraint,” Proc. of CVGIP, 2004.
[12]J. T. Tsai, T. K. Liu, and J. H. Chou, “Hybrid Taguchi-Genetic Algorithm for Global Numerical Optimization,” IEEE Trans. Evolutionary Computation, Vol. 8, No. 4, Aug. 2004, pp. 365-377.
[13]Z. Tu and Y. Lu, “A Robust Stochastic Genetic Algorithm (StGA) for Global Numerical Optimization,” IEEE Trans. Evolutionary Computation, Vol. 8, No. 5, October 2004, pp. 456-470.
[14]W. Zhong, J. Liu, M. Xue, and L. Jiao, “A Multiagent Genetic Algorithm for Global Numerical Optimization,” IEEE Trans. Systems, Man, and Cybernetics—part B: Cybernetics, vol. 34, no. 2, April 2004, pp. 1128-1141.
[15]E. Zitzler, M. Laumanns, and S. Bleuler, “A Tutorial on Evolutionary Multiobjective Optimization,” Metaheuristics for Multiobjective Optimization, Lecture Notes in Economics and Mathematical Systems, Springer, vol. 535, 2004, pp. 3–37.
[16]李輝煌,田口方法-品質設計的原理與實務,高利圖書有限公司。
[17]孫春在,『演化式計算』http://cindy.cis.nctu.edu.tw/EC/main/ ec_ind.htm, Access time 2005/3/10.
[18]蘇朝墩,『田口式品質工程』 http://www.qrc.nthu.edu.tw/QRC/ studyQ/pdf/田口式品質工程簡介.pdf, Access time 2005/3/10.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔