臺灣博碩士論文加值系統

本研究主要期望研發ㄧ個有效的影像切割方法，將模糊物件形貌特徵，從亮度不均勻的影像背景，正確的切割出來。2-D自動化光學檢測技術，在工業製造過程裡扮演著一個重要絕對的角色；影像切割技術是一項決定性的判斷，將物件從相鄰背景裡切割為結果。由於在切割模糊物件形貌特徵上，仍有許多複雜度和技術處理上的挑戰，因此仍需去研發ㄧ個強健的方法和探討。
ㄧ般影像切割於模糊物件上，仍然存在的主要問題：(1)昏暗不明的物件和影像背景，導致模糊物件形貌難判斷。(2)亮度不均勻的影像背景，導致影像切割物件形貌錯誤。(3)鄰近於模糊物件形貌特徵的雜訊，將會被錯誤的切割於結果。
本論文針對這三大問題發展一個有效的物件切割方法，優於SVD和DCT切割法上，無法解決影像背景亮度變化大時產生的切割錯誤。在上述問題模擬的惡劣影像下，本論文的切割結果可保留物件特徵高於DCT方法14%以上。並經由驗證工業上影像切割物件，本研究方法足以有效且正確的切割出物件形貌特徵。

This article presents an innovative image segmentation method to extract an object underlying defect detection from its background image. 2-D automatic optical inspection (AOI) technology for defect detection and classification has played a vital role for in-situ manufacturing industrial sectors nowadays. Image segmentation is a crucial step to extract component information from its neighboring background. Due to potential complexity in such an image processing operation, considerable challenges are crucially encountered in establishing a robust approach.
In general, three major factors play a significant influence on the result of the segmented image objects: (1) brightness distribution of the background image; (2) degree of unbalanced brightness of the background image; (3) noise level near the object feature to be detected.
The research addresses these important factors and develops an effective segmentation method. Exclusive advantage of the method is to overcome the current limitations of the existing SVD (singular value decomposition) or DCT (discrete Cosine transfer) methods. The segmentation performance of the developed method is up to 14% better than the DCT method, in terms of accuracy of segmentation. From the test results on some real industrial cases, it is verified that the method is capable of extracting the tested object desirably.

中文摘要 i
英文摘要 ii
誌 謝 iii
目錄 iv
表目錄 vi
圖目錄 vii
第一章 緒 論 1
1.1 研究背景與動機 1
1.2 研究範圍與目的 2
1.3 研究方法簡介 4
1.4 論文架構 7
第二章 文獻探討 8
2.1 AOI影像應用介紹 8
2.2 ㄧ般工業製程實例簡介 10
2.3文獻回顧 12
2.3.1 閥值切割法 13
2.3.1.1 群聚型態閥值法(Clustering-based threshold) 14
2.3.1.2 熵值型態閥值法(Entropy-based threshold) 16
2.3.1.3 形狀統計閥值法(Histogram shape- based threshold) 22
2.3.1.4 區域型態閥值法(Local threshold) 23
2.3.2 基於背景評估的影像差異切割法 27
2.3.2.1 傅立葉轉換(Fourier Transform, FT) 27
2.3.2.2 離散餘弦轉換(Discrete Fourier Transform, DCT) 28
2.3.3 以背景評估影像差異切割為主的Mura檢測方法 28
2.4 文獻探討結論 30
第三章 研究方法 32
3.1 收斂式沉澱揮發法強化模糊特徵和背景對比 34
3.1.1影像灰度強化與模糊特徵之相關性 34
3.1.2沉澱揮發法與模糊特徵之相關性 37
3.1.3影像灰階均值化與模糊特徵之相關性 39
3.1.4收斂式沉澱揮發法強化灰度特徵細部流程 39
3.2 方向性影像亮度補償解決背景亮度分布不均 41
3.3 DCT背景評估法問題解決討論 43
第四章 實驗結果與討論 51
4.1檢測系統架構及實驗參數說明 51
4.1.1 檢測系統架構說明 51
4.1.2 實驗參數說明 52
4.2 實驗結果 52
4.2.1 系統實驗流程圖說明 52
4.2.2 系統實驗結果 54
4.2.3 系統實驗結果與DCT實驗結果比較 65
4.2.4 系統實驗特徵於不同影像背景亮度的切割影響 72
4.2.5 系統實驗Mura缺陷切割結果 73
4.3 系統效能評估 76
4.3.1 系統檢測能力評估設計 76
4.3.2 系統檢測能力評估模擬流程和結果 78
4.3.3 系統檢測能力各參數極限驗證 78
4.3.4 系統檢測能力討論 83
4.4 系統檢測定量差異分析 84
4.5 系統檢測速度分析 85
4.6 系統檢測於多模糊物件的影像探討 86
第五章 結論與展望 88
5.1 結論 88
5.2 未來展望 91
參考文獻 93
附錄 95
附錄 A. 影像處理技術 95
A.1均值濾波(Smoothing filter) 95
A.2影像灰度強化 96
A.3影像亮度補償 98
附錄 B. 參考文獻相關流程圖 102
B.1影像傅利葉轉換 102
B.2影像離散餘弦轉換 103
B.3基於背景評估的影像差異法 106

[1] 張振堶，自動光學檢測(AOI)市場及技術發展趨勢調查，新竹：工研院IEK系統能源組，2002。
[2] http://en.wikipedia.org/wiki/Computer_tomography
[3] http://www.tbcl.com.tw/Product/Xavis/TBCL-SX-090A-SMT.pdf
[4] http://www.yamato-china.cn/english/products/tdm1000/tdm1000.htm
[5] M. Sezign and B. Sankur, "Survey over image thresholding techniques and quantitative performance evaluation," 2003.
[6] N. Otsu, "A Tlreshold Selection Method from Gray-Level Histograms," IEEE, 1979.
[7] C. I. Chang, Y. Du, J.Wang and Thouin. "Survey and comparative analysis of entropy and relative entropy thresholding techniques," 2006.
[8] T. Pun, "A new method for grey-level picture thresholding using the entropy of the histogram," Signal process, vol. 2(3), 1980, pp. 223-237.
[9] T. Pun, "Entropic thresholding: a new approach," Compute. Graph. Image Process, vol. 16, 1981, pp. 210-239.
[10] J. N. Kapur, P. K. Sahoo and A. K. C. Wong, "A new method for gray-level picture thresholding using the entropy of the histogram," Computer Vision, Graphics, Image Processing, vol. 29(3), 1985, pp. 273-285.
[11] T. Cover and J. Thomas, Elements of information theory, John Wiley & Sons, 1991.
[12] R. Gonzalez and J. Woods, Digital image processing, 1992.
[13] A. S. Abutaleb, "Automatic thresholding of grey-level pictures using two-dimensional entropy," Comput. Vis. Graph. Image Process, vol. 47, 1989, pp. 22-32.
[14] N. R. Pal and S. K. Pan, "Entropic thresholding", Signal process, vol. 16, 1989, pp. 97-108.
[15] J. Kittler and J. Illingworth, "Minimun error thresholding," 1986.
[16] C. H. Li and C. K. Lee, "Minimum cross entropy thresholding," 1993.
[17] C. I. Chang, K. Chen, J. Wang and M. L. G. Althouse, "A relative entropy-based approach to image thresholding," 1994.
[18] S. S. Lee, S. J. Horng and H. R. Tsai. "Entropy thresholding and its parallel algorithm on the reconfigurable array of processors with wide," 1999.
[19] K. Pratt, S. Sawkar and O.Reilly, "Automatic blemish detection in liquid crystal flat panel displays," SPIE, vol. 3306, 1998, pp. 2-13.
[20] 郭家成，薄膜電晶體液晶顯示器Mura瑕疵檢測技術之研發，碩士論文，國立臺北科技大學自動化科技研究所，台北，2006。
[21] Niblack, "An Introduction to Image Processing," 1986, pp. 115–116.
[22] O. Stepan, M. Jiri, "Image Retrieval Using Local Compact DCT-based Representation," Pattern Recognition, DAGM''03,25th, Germany, 2003.
[23] R. C. Gonzalez and R. E. Woods, Digital Image Processing, 2nd Edition, Prentic-Hall, NJ, 2002.
[24] J. M. S. Prewitt and M. L. Mendelsohn, "The analysis of cell images," in Ann. New York Acad. Sci., vol. 128, 1998, pp. 1035-1053.
[25] A. Rosenfeld and P. De La Torre, "Histogram concavity analysis as an aid in threshold selection," IEEE Trans. System Man Cybernet., vol. 13, 1983.
[26] W. Doyle, "Operation useful for similarity-invariant pattern recognition," J.Assoc.Comput.Mach, vol. 9, 1962, pp. 259-267.
[27] A. Khashman and B. Sekeroglu, "A Novel Thresholding Method for Text Separation and Document Enhancement," 11th Panhellenic Conference in Informatics, 2006, pp. 323-330.
[28] http://en.wikipedia.org/wiki/Fourier_transform
[29] http://en.wikipedia.org/wiki/Cooley-Tukey_FFT_algorithm
[30] J. G. Campbell, C. Fraley, D. Stanford, F. Murtagh and A. E. Raftery, Model-Based Mesed Methods for Textile Fault Detection, John Wiley & Sons, Inc., vol. 10, 1999, pp. 339-346.
[31] S. Golam and A.Ajith, "DCT based texture classification using soft computing approach," Malaysian Journal of Computer Science, 2004.
[32] J. Y. Lee and S. I. Yoo, "Automatic Detection of Region-Mura Defect in TFT-LCD," IEICE TRANS. INF. &SYST., vol. E87-D (10), 2004, pp. 2371-2378.
[33] K. N. Choi, J. Y. Lee and S. I. Yoo, "Area-Mura Detection in TFT-LCD Panel," The Interational Society for Optical Engineering 5300, 2004, pp. 151-158.
[34] 尹志瑋，應用二維離散餘弦轉換於雙穩態膽固醇液晶顯示器之表面瑕疵檢測，碩士論文，私立中原大學機械研究所，桃園，2008。
[35] 謝承恩，以累積差異及多重解析度背景相減偵測LCD Mura瑕疵，碩士論文，國立中央大學資訊工程研究所，桃園，2008。
[36] 楊舜霖，基於奇異值分解法運用於薄膜電晶體液晶顯示器Mura瑕疵之自動化檢測技術，碩士論文，國立臺北科技大學自動化科技研究所，台北，2007。
[37] C. J. Lu and D. M. Tsai, “Automatic defect inspection for LCDs using singular value decomposition,” Int. J. Adv. Manuf. Technol., 2005, pp. 53–61.
[38] 薛昭能，應用奇異值分解法(SVD)於雙穩態膽固醇液晶顯示面板之表面瑕疵檢測，碩士論文，私立中原大學機械研究所，桃園，2008。