進階搜尋


 
系統識別號 U0026-0707201111395800
論文名稱(中文) 應用於小鋼胚面部及角部缺陷檢測之電腦視覺自動化系統
論文名稱(英文) Computer Vision System for Corner and Planar Billet Defect Inspection
校院名稱 成功大學
系所名稱(中) 資訊工程學系碩博士班
系所名稱(英) Institute of Computer Science and Information Engineering
學年度 99
學期 2
出版年 100
研究生(中文) 蘇育興
研究生(英文) Yu-Hsing Su
學號 p76984283
學位類別 碩士
語文別 中文
論文頁數 112頁
口試委員 指導教授-孫永年
口試委員-李同益
口試委員-朱銘祥
口試委員-陳中明
口試委員-柳金章
中文關鍵字 工業表面缺陷檢測  影像處理  自動檢測  模組化系統  模糊類神經網路 
英文關鍵字 surface defect inspection technology  image processing  automated inspection  modified system  fuzzy neural 
學科別分類
中文摘要 本篇論文提出一套針對小鋼胚角部影像,包含表面之角部部份之缺陷自動偵測與分析系統。本系統主要針對現場即時所拍攝的小鋼胚影像偵測缺陷,並分辨其缺陷種類。系統主要包含了三個主要步驟:影像處理、缺陷偵測、缺陷特徵擷取以及缺陷分類。本系統包含了兩個目標設定。第一個目標是為了協助現場人員能夠即時的監控與調整鋼胚,所以在影像處理、缺陷偵測及缺陷特徵的擷取部份皆採用計算複雜度較低的演算法,同時為確保線上偵測速率,也加入了一些演算法加速的機制。在缺陷分類之前,也依據缺陷的特性及各攝影機所拍攝到的影像特性作特徵擷取,以確保特徵的缺陷類別區分能力。因此,我們分別針對採用所有特徵與經過特徵篩選機制所獲得的特徵組合進行比較。在特徵篩選的部份,我們採用了禁忌搜尋法(Tabu Search)配合KNN分類器來找出該特徵群的最佳組合。另一個目標則是為了確保高準確率及符合現場技術人員的缺陷判斷機制,採用模糊分類器來進行分類,並從中建構出描述各項缺陷特徵的規則作為描述。
實驗過程中,我們請現場專家判定影像中缺陷的位置及種類,接著和系統檢測的結果做比較。實驗結果顯示我們的系統能夠快速且準確地檢測出缺陷的位置與種類。且在缺陷分類器所建立的模糊規則也與專家判斷缺陷的規則相當吻合。
英文摘要 In this thesis we proposed an automatic defect detection and analysis system for steel billet. The proposed system detects the defects in both corner and planar billet images and classifies the defect types. The system consists of three modules: (1) Image processing, (2) defect detection, (3) feature extraction and Fuzzy Neural network classification. There are two major goals for the proposed system. The first goal is to provide factory staff with real-time monitoring and adjustment for billets. Therefore, the selected algorithms for image processing, defect detection and classification are with certain acceleration mechanisms to assure computational efficiency for online inspection. Based on defect characteristics and image properties, some defect features are designed and selected for defect classification. The classification results by using all features are compared with the ones using only selected features. The feature selection module which combines the Tabu Search with the k-nearest neighbor classifier is used to obtain the best set of features. The second goal is to follow the manual concepts of defect judgment. Thus, we use the fuzzy classifiers which are similar to human judgment to construct the detection rules for different kinds of billet corner defects.
In the experiments, some expert detection results were used to validate the correctness of proposed system. Our system was found capable of detecting defects correctly and classifying defect type accurately. The adopted fuzzy rules are also close to the ways that human experts classify the defect types.
論文目次 第一章、序論 1
1.1研究之背景與目的 1
1.2國內外相關研究 2
1.3論文組織架構 7
1.4論文貢獻 8
第二章、小鋼胚影像與系統概述 10
2.1小鋼胚缺陷定義 10
2.2缺陷檢測概述 16
第三章、影像處理與缺陷偵測 22
3.1影像前處理 22
3.2缺陷偵測 32
第四章、特徵擷取與特徵篩選 44
4.1缺陷特徵擷取 44
4.2缺陷特徵篩選 53
第五章、模糊規則與缺陷分類 66
5.1模糊規則分類器 66
5.2倒傳遞類神經網路分類器 73
5.3模糊類神經網路概述 75
5.4模糊類神經網路詳細流程 76
第六章、實驗結果與討論 84
6.1樣本收集與統計 84
6.2特徵篩選組合實驗 85
6.3模糊分類器之結果 88
6.4模糊規則與類神經網路分類器比較 91
6.5模糊類神經網路建立之規則討論 99
6.6缺陷偵測處理速度 101
第七章、結論與未來展望 104
參考文獻 108
參考文獻 [1] Giuseppe Acciani, Gioacchino Brunetti, and Girolamo Fornarelli, “Application of Neural Networks in Optical Inspection and Classification of Solder Joints in Surface Mount Technology”, IEEE Trans.on Industrial Informatics, vol. 2, no. 3, August 2006.
[2] FreÂde ric Jurie and Michel Dhome, “Hyperplane Approximation for Template Matching”, IEEE Trans.on Pattern Analysis And Machine Intelligence, vol. 24, no. 7, pp. 996-1000 , July 2002.
[3] HenryY.T. Ngan, Grantham K.H. Pang, S.P.Yung and Michael K. Ng, “Wavelet based methods on patterned fabric defect detection”, Pattern Recognition, vol. 38, pp. 559 – 576 , 2005.
[4] T. Masters, “Practical neural network recipes in C++”, Academic press, 1993.
[5] Henry Y. T. Ngan and Grantham K. H. Pang, “Regularity Analysis for Patterned Texture Inspection”, IEEE Trans.on Automation Science and Engineering, vol. 6, no. 1, pp.131-144, January 2009.
[6] Otsu N., "A threshold selection method from gray-level histogram", IEEE Trans. Syst. Man Cybern. Vol. 9, pp. 62-66, 1979.
[7] J. R. Parker, “Algorithms for image processing and computer vision”, John Wiley and Sons , 1997.
[8] Jong Pil Yun, Sung Hoo Choi, Jong-Wook Kim, Sang Woo Kim, “Automatic detection of cracks in raw steel block using Gabor filter optimized by univariate dynamic encoding algorithm for searches (uDEAS)”, NDT&E International, vol. 42, pp. 389–397, 2009.
[9] Kim JW, Kim SW, "Parameter identification of induction motors using dynamic encoding algorithm for searches(DEAS)", IEEE Trans Energy Convers, vol. 20, no.1, pp. 16–24, 2005.
[10] Kim JW, Kim SW, “PID control design with exhaustive dynamic encoding algorithm for searches(eDEAS)”, Int J Control Autom Syst, vol. 5, no. 6, pp. 691–700, 2007.
[11] J. R. Parker, “Algorithms for image processing and computer vision”, John Wiley and Sons , 1997.
[12] Rafael Vilar, Juan Zapata, and Ramon Ruiz, “An automatic system of classification of weld defects in radiographic images”, NDT&E International, vol. 42, pp.467–476, 2009.
[13] R.A. Johnson and W.W. Dean, “Principal component analysis, 2nd edition”, Springer-Verlag, 2002.
[14] T.M. Cover and P.E. Hart, “Nearest Neighbor Pattern Classification”, IEEE Trans. Inform. Theory, vol. 13, no. 1, pp. 21–27, 1967.
[15] T. Masters, “Practical neural network recipes in C++”, Academic press, 1993.
[16] R.O. Duda, P.E. Hart, and D.G. Stork, “Pattern Classification”, 2nd edition, John Wiley & Sons, 2001.
[17] T.Kohonen, “An Introduction to Neural Computing”, Neural Networks, vol. 1, p. 3-16, 1988.
[18] S. Haykin, “Neural Networks, a comprehensive foundation”, 2nd edition, Prentice Hall, 1994.
[19] Hornik, K., M. Stinchcombe and H., White, "Multilayer feed forward Networks are Universal Approximators", Neural Networks, 2 (5): 35g, 366, 1989.
[20] D Zipser and R A Andersen, "A back-propagation programmed network that simulates response properties of a subset of posterior parietal neurons", Nature, vol. 331, pp. 679 - 684, Feb, 1988.
[21] Dong Ho Shin, Do Hyun Kim and Man Suk Song, "The Steepest Descent Method and The Conjugate Gradient Method for slightly non-symmetric, Positive Definite Matrices", Comm. Korean Math. Soc., No.2, pp. 439-448, 1994.
[22] D. P. Muni, N. R. Pal, and J. Das, “Genetic Programming for Simultaneous Feature Selection and Classifier Design”, IEEE Trans. Syst., Man, Cybern. B, vol. 36, pp.106–117, 2006.
[23] T. Nakashimaa, G. Schaeferb, Y. Yokotaa, and H. Ishibuchia “A weighted fuzzy classifier and its application to image processing tasks”, Fuzzy Sets and Systems, vol. 158, no. 3, pp. 284-294, 2007.
[24] T. Nakashimaa, G. Schaeferb, Y. Yokotaa, and H. Ishibuchia “A weighted fuzzy classifier and its application to image processing tasks”, Fuzzy Sets and Systems, vol. 158, no. 3, pp. 284-294, 2007.
[25] Y Freund, “Boosting a weak learning algorithm by majority”, Information and Computation, vol. 121, No. 2, pp. 256-285, 1995.
[26] V. Lashkia and S. Aleshin, “Test Feature Classifiers: Performance and Applications”, IEEE Trans. Syst., Man, Cybern. B, vol. 31, vol. 4, pp.643–650, 2001.
[27] Yeong Min Kim, Chee Kyeong Kim and Sung Gul Hong, “Fuzzy based state assessment for reinforced concrete building structures”, Engineering Structures , vol. 28, pp. 1286–1297. 2006.
[28] Vincent Bombardier and Emmanuel Schmitt, “Fuzzy rule classifier: Capability for generalization in wood color recognition”, Engineering Applications of Artificial Intelligence, vol. 23, pp. 978–988, 2010.
[29] George Tsekouras, Haralambos Sarimveis and George Bafas, "A simple algorithm for training fuzzy systems using input–output data", Advances in Engineering Software, vol. 34, pp. 247–259, 2003.
[30] Li-Xin Wang and Jerry M. Mendel, "Fuzzy Basis Functions, Universal Approximation, and Orthogonal Least-Squares Learning", IEEE Trans. On Neural Networks, VOL. 3, NO. 5, Sep 1992.
[31] Chin-Teng Lin, "A neural fuzzy control system with structure and parameter learning", Fuzzy Sets and Systems, vol. 70, pp. 183 212, 1995.
[32] Chunshien Li and Chun-Yi Lee, "Self-Organizing Neuro-Fuzzy System for Control of Unknown Plants", IEEE Trans. On Fuzzy Systems, Vol. 11, No. 1, Feb 2003.
[33] Chi-Feng Wua, Cheng-Jian Lin and Chi-Yung Lee, "A functional neural fuzzy network for classification applications", Expert Systems with Applications, vol. 38, pp. 6202–6208, 2011.
[34] 許紋誠,2008,《小鋼胚缺陷偵測與分析系統》,國立成功大學資訊工程學系碩士論文。
論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2014-08-30起公開。
  • 同意授權校外瀏覽/列印電子全文服務,於2014-08-30起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw