臺灣博碩士論文加值系統

玻璃為一種透明、不透氣且具一定硬度的物料，主要成分為二氧化矽（SiO2，即石英，砂的主要成分）。因其化學性質穩定與價格低廉，使得玻璃瓶於日常生活環境中使用量相當龐大，然而玻璃瓶的製造過程中容易產生破泡、雜質、裂痕、質變等缺陷，造成原物料的浪費、成本提升。傳統檢測方法為僱用「瓶」檢員，對玻璃瓶進行生產線上逐一檢測，但玻璃瓶的生產量數以百萬計，檢測過程枯燥，由於檢測時需面對強光，使目檢員無法持續且快速的判斷瑕疵，故易導致判斷準確率隨著時間降低，而產生漏檢之情形。機器視覺 (Machine Vision) 技術為機器模仿人類視覺的光學系統，其結合工業相機、鏡頭、電腦等工業元件，可有效擷取、分析及解釋所得之檢測影像內容，並可進行線上即時全面檢測，提供高精準度以及不間斷之工作型態。台灣為玻璃瓶之主要生產國之一，對於玻璃瓶自動化檢測之需求急迫，但目前國內仍無玻璃瓶檢測系統之相關研究，故本研究目的為運用機器視覺建構一高精準度、高效率且穩定性高之自動化玻璃瓶瑕疵檢測系統。本研究系統建構方式分為硬體與軟體兩方面進行；硬體部分，利用鋁擠型、工業電腦、光源、工業相機及相關機構進行系統硬體建構；系統軟體部份為結合 Euresys eVision與Borland C++ Builder為主體進行建構。系統效能經測試，其平均整體正確率達99.68%。本研究另外進行相關影像分割演算法，分水嶺演算法、均值飄移演算法與連通元件標記法之比較，並擇優置入系統，經測試連通元件標記法較為適合本系統。

Glass bottle is one of the most important utensils for human being. Various kinds of bottle products are used in large amount in food and drink production. Since bottles may have some defects that may cause negative impact, even dangerous consequences, for production and customers, empty glass bottles requires a 100% on-line inspection after production. Currently, this kind of work is most likely performed by well-trained human inspector. Manual inspection not only increases labor cost but is very difficult to guarantee inspection quality. Therefore, machine vision-based empty bottle inspection becomes a crucial issue for glass bottle industry.
The area of empty bottle inspection requires more DIP (Digital Image Processing) and lighting knowledge, and experimental study due to the transparency feature of glass bottles. Therefore, the objective of this study is to develop an empty bottle inspection system which adopts machine vision to effectively extract and identify the bottle defects. This study is planned to develop two major parts: glass bottle lighting device module and DIP bottle defect detection algorithm. All the defective bottle examples will be collected by H Com., and then DIP inspection algorithm and lighting device module will be developed in this study.

摘　要 i
ABSTRACT iii
誌 謝 v
目 錄 vi
圖目錄 x
第一章 緒論 1
1.1 研究背景與動機 1
1.2 研究目的 3
1.3 研究範圍與限制 3
1.4 研究架構 4
第二章 文獻探討 5
2.1 玻璃瓶瑕疵之相關文獻 5
2.2 玻璃瓶檢測之相關文獻 11
2.3 影像分割演算法相關文獻 13
2.3.1 分水嶺分割演算法……………………………………………………..……………….13
2.3.2 均值飄移分割演算法…………………………………………………………………….18
2.3.3 連通物件標記演算法…………………………………………………………………….23
2.4 工業應用與市場分析 …………27
第三章 研究方法 31
3.1 影像擷取環境設計 31
3.2 影像處理與分析 32
3.2.1 一般性瑕疵之影像處理與分析….………………………………………………….33
3.2.2 低對比瑕疵之影像處理與分析……………………………………………………..35
3.2.3 螺紋之影像處理與分析……………………………………………………..………….36
第四章 實驗結果與分析 41
4.1 檢測系統硬體之建構 41
4.1.1 鏡頭焦長之遴選……………………………………………………………………………45
4.1.2 彩色相機之測試…………………………………………………………………….……..47
4.1.3 光柵背景之測試……………………………………………………..…………………….49
4.2 系統影像分割演算法之比較 50
4.3 實驗數據分析與討論 60
4.3.1 各類瑕疵個別檢測率………………………………………………………….…………61
4.3.2 整體系統之線上檢測率…………………………………………………………….….66
4.4 分水嶺與均值飄移演算法於原始影像之測試 69
4.5 實驗結果問題與討論 72
第五章 結論與未來研究方向 74
5.1 結論 74
5.2 研究貢獻 74
5.3 未來研究建議 75
參考文獻 76

[1]殷海榮與李啟甲，玻璃的成形與精密加工，北京：化學工業出版社，2009。
[2]楊昌祐，中文扭曲文件影像之還原與文字切割，碩士論文，國立台灣科技大學電子工程系，台北市，2011。
[3]網址：http://zh.wikipedia.org
[4]網址：http://jzg.kong.com.tw/
[5]Shin-Min Chao, Du-Ming Tsai , Yan-Hsin Tseng, and Yuan-Ruei Jhang, Defect detection in low-contrast glass substrates using anisotropic diffusion, Master, Department of Industrial Engineering and Management, Yuan-Ze University, Taiwan, R.O.C.,2006
[6]Wei-Chen Li, and Du-Ming Tsai, Defect Inspection in Low-contrast LCD Images Using Hough Transform-based Non-stationary Line Detection, Master, Department of Industrial Engineering and Management Yuan-Ze University, Taiwan, R.O.C. ,2009
[7]Du-Ming Tsai, and Hsin-Yang Tsai, "Low-contrast surface inspection of mura defects in liquid crystal displays using optical flow-based motion analysis," Machine Vision and Applications, 2011, pp. 629-649
[8]Nishu, and Sunil Agrawal, "Automated Inspection of Defects in Glass by proper Color space selection and Segmentation Technique of Digital Image processing," Int.J.Computer Technology & Applications, vol. 3 , 2012, pp. 1058-1063
[9]Jun Wang, and Anand K. Asundi, "A computer vision system for wineglass defect inspection via Gabor-filter-based texture features," Information Sciences 127, 2000, pp. 157-171
[10]Feng Duan, Yao-Nan Wang, Huan-Jun Liu, and Yang-Guo Li, "A machine vision inspector for beer bottle," Engineering Applications of Artificial Intelligence 20, 2007, pp. 1013–1021
[11]Feng Duan, Yao-Nan Wang, Huan-Jun Liu, and Yang-Guo Li, "A Real-time Machine Vision System for Bottle Finish Inspection," 8th International Conference on Control, Automation, Robotics and Vision Kunming, 2004
[12]Feng Duan, Yao-Nan Wang, Huan-Jun Liu, and Yang-Guo Li, "Glass Bottle Inspector Based on Machine Vision," World Academy of Science, Engineering and Technology 20, 2008
[13]Serge Beucher and Christian Lantuejoul., "Use of watersheds in contour detection," In International workshop on image processing, real-time edge and motion detection, 1979
[14]網址：http://en.wikipedia.org/wiki/Watershed_(image_processing)#cite_note-2
[15]Gonzalez and Woods , Digital Image Processing 2/e, Mass : Addison-Wesley, 1992
[16]Laurent Najman and Michel Schmitt, "Geodesic Saliency of Watershed Contours and Hierarchical Segmentation," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 18, 1996, pp. 1163 -1173
[17]Leila Shafarenko, Maria Petrou, and Josef Kittler, "Automatic Watershed Segmentation of Randomly Textured Color Images," IEEE Transactions on Image Processing, vol. 6, 1997, pp. 1530-1544
[18]Kostas Haris, Serafim N. Efstratiadis, Nicos Maglaveras, and Aggelos K. Katsaggelos, "Hybrid Image Segmentation Using Watersheds and Fast Region Merging," IEEE Transactions on Image Processing, vol. 7, 1998, pp. 1684-1699
[19]Demin Wang, "Unsupervised Video Segmentation Based on Watersheds and Temporal Tracking," IEEE Transactions on Circuits and Systems for Video Technology, vol. 8, 1998, pp. 539-546
[20]Alan P. Mangan and Ross T. Whitake, "Partitioning 3D Surface Meshes Using Watershed Segmentation," IEEE Transactions on Visualization and Computer Graphics, vol. 5, 1999, pp. 308-321
[21]V. Grau, A. U. J. Mewes, M. Alcaniz, R. Kikinis, and S. K. Warfield, "Improved Watershed Transform for Medical Image Segmentation Using Prior Information," IEEE Transactions on Image Processing, IEEE Transactions on Medical Imaging, vol. 23, 2004, pp. 447-458
[22]Keinosuke Fukunaga and Larry D. Hostetler, "The Estimation of the Gradient of a Density Function, with Applications in Pattern Recognition," IEEE Transactions on Information Theory, vol. IT-21, 1975, pp. 32-40
[23]Dorin Comaniciu and Pater Meer, "Mean shift: a robust approach toward feature space analysis," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, 2002, pp. 603-619
[24]Xiaodong Yang, Houqiang Li, and Xiaobo Zhou, "Nuclei Segmentation Using Marker-Controlled Watershed, Tracking Using Mean-Shift, and Kalman Filter in Time-Lapse Microscopy," IEEE Transactions on Circuits and Systems—I: Regular Papers, vol. 53, 2006, pp. 2406-2414
[25]Wenbing Tao, Hai Jin, and Yimin Zhang, "Color Image Segmentation Based on Mean Shift and Normalized Cuts," IEEE Transactions on Systems, Man, and Cybernetics - Part B: Cybernetics, vol. 37, 2007, pp. 1382-1389
[26]M.A. Carreira-Perpindn, "Gaussian Mean-Shift Is an EM Algorithm," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 29, 2007, pp. 767-776
[27]Arnaldo Mayer and Hayit Greenspan, "An Adaptive Mean-Shift Framework for MRI Brain Segmentation," IEEE Transactions on Medical Imaging, vol. 28, 2009, pp. 1238-1250
[28]R.M.Haralick and L.G.Shapiro, "Computer and Robot Vision Volume I," Addison-Wesley Publishing Company Inc., United States of America, 1992
[29]Massimo Maresca, Hungwen Li, and Mark Lavin, "CONNECTED COMPONENT LABELING ON POLYMORPHIC TORUS ARCHITECTURE," IEEE, 1988
[30]N. Ranganathan, R. Mehrotra, And S. Subramanian, "A High Speed Systolic Architecture for Labeling Connected Components in an Image," IEEE Transactions on Systems, Man, and Cybernetics, vol. 25, 1995, pp. 415-423
[31]Jia-Ping Wang, "Stochastic Relaxation on Partitions With Connected Components and Its Application to Image Segmentation," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 20, 1998, pp. 619-636
[32]Yebin Fan, Shengsheng Yu, and Hualong Zhao, A Novel Line Based Connected Component Labeling Algorithm, Master, Department of Computer Science Huazhong University of Science and Technology Wuhan, China, 2010
[33]Michael Klaiber, Lars Rockstroh, Zhe Wang, Yousef Baroud and Sven Simon, A Memory-Efficient Parallel Single Pass Architecture for Connected Component Labeling of Streamed Images, Master, Institute for Parallel and Distributed Systems University of Stuttgart, Stuttgart, Germany, 2012
[34]網址：http://www.tiama.eu/
[35]網址：http://www.heye-international.com
[36]網址：http://www.iris-im.com/