本篇論文是針對印刷電路板(Printed Circuit Board, PCB)上靶孔的圓洞影像進行圓心偵測(Circular hole center detection)，藉由各別靶孔的圓心資訊來校準(Alignment)印刷電路板，然而，本論文的訴求為在連續拍攝的圓洞影像，該演算法對圓洞定位誤差的穩定度。在圓形霍夫轉換演算法(Circle Hough transform)中，圓洞定位誤差的不穩定現象，導致無法順利定位靶孔，因此本論文設計一套自動定位靶孔的系統，首先利用Otsu演算法分割前景(foreground)和背(Background)，但Otsu演算法在前景和背景差異太大的情況下，並不能夠得到好的分割效果，所以本論文針對Otsu提出改進的方法，並在實驗中的結果證實本論文改進方法優於其他演算法。而後，為了濾除前景中的訊影像，並保留可能為圓洞的前景資訊，利用Connected Component Labeling來分析Otsu分割後的前景影像，其中本論文為爭取效率而採用A Run-Based One-Scan Labeling Algorithm 的方法。濾除雜訊後，最後使用Canny邊緣偵測(Edge detection)和圓形霍夫轉換演算法偵測圓洞圓心完成靶孔定位，且圓洞定位誤差的穩定度也優於先前的演算法。

This paper addresses a method for detections of a center of a circular shape of a target hole on Printed Circuit Board (PCB). Regarding positions of the target hole, alignments and arrangements of sockets with drilling on PCB for accommodation of other electronic components are made. However, this paper is aimed at robustness of localization of the target hole with proposed algorithms on consecutive images captured.
With Circle Hough Transform algorithm, positioning error existence causes bad performance, i.e. non-robust data production. Because of this, it is not feasible that a drilling process of PCB is made. Therefore, in this paper, an automated positioning system for the target hole with Otsu segmentations is developed. Nonetheless, Otsu segmentation algorithm is not able to split an image into foreground and background when the foreground and background are not obviously different. Hence, improvement of Otsu segmentation is proposed. By comparing experimental results of the proposed and current Otsu segmentation, benefits of the proposed Otsu segmentation are verified. Besides, Connected Component Labeling is proposed to filter out noisy data on the foreground image. In order to enhance efficiency of noise removal, Run-Based One-Scan Labeling Algorithm is adopted. Then, Canny Edge Detection and Circle Hough Transform are used to locate the center of the circular shape of the target hole.

[1]R. Brunelli, Template Matching Techniques in Computer Vision: Theory and Practice, Wiley, ISBN 978-0-470-51706-2, 2009
[2]M. Cheriet, J.N. Said and C.Y. Suen, "A recursive thresholding technique for image segmentation," IEEE Trans. Image Process. 7, pp.918–921, 1998
[3]C. K. Chow and T. Kaneko, "Automatic boundary detection of the left ventricle from cineangiograms," Comput. Biomed. Res., vol. 5, pp. 388- 410, 1972
[4]L.J. Dong, G. Yu, P. Qgunbona and W.Q. Li, “An efficient iterative algorithm for image thresholding,” Pattern Recognition Lett. 29, pp.1311–1316, 2008
[5]R. O. Duda and P. E. Hart, "Use of the Hough Transformation to Detect Lines and Curves in Pictures," Comm. ACM, Vol. 15, pp. 11-15, 1972
[6]Jiu-Lun Fan, Bo Lei, "A modified valley-emphasis method for automatic thresholding," Pattern Recognition Letters 33, pp.703–708, 2012
[7]J. Gong, L. Li and W. Chen, "Fast recursive algorithms for two-dimensional thresholding," Pattern Recognition 31, pp.295–300, 1998
[8]L. He, Y. Chao and K. Suzuki, "A Run-based Two-Scan Labeling Algorithm," ICIAR. LNCS, pp.131–142, 2007
[9]L. He, Y. Chao and K. Suzuki, "A run-based two-scan labeling algorithm," IEEE Transactions on Image Processing, pp.749–756, 2008
[10]Lifeng He, Yuyan Chao, Kenji Suzuki, and Hidenori Itoh, "A Run-Based One-Scan Labeling Algorithm," ICIAR, LNCS 5627, pp.93–102, 2009
[11]Z. Hou, Q. Hu and W.L. Nowinski, "On minimum variance thresholding," Pattern Recognition Lett 27, pp.1732–1743, 2006
[12]Q.M. Hu, Z.J. Hou and L.N. Wieslaw, "Supervised range-constrained thresholding," IEEE Trans. Image Process. 15, pp.228–240, 2006
[13]Fukunage K, "Introduction to Statisticul Pattern Recogniition," New York: Academic, pp.260-267, 1972
[14]T. Kurita, N. Otsu and N. Abdelmalek, "Maximum likelihood thresholding based an population mixture models," Pattern Recognition 25, pp.1231–1240, 1992
[15]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. Graph," Models Image Process 29, pp.273–285, 1985
[16]J. Kittler and J. Illingworth, "On threshold selection using clustering criteria," IEEE Trans. Systems Man Cybernet. 15, pp.652–655, 1985
[17]Yuhai Li, L. Jian, T. Jinwen and X. Honbo, "A fast rotated template matching based on point feature," SAR and Multispectral Image Processing, pp.453-459, 2005
[18]F. Morii, "An image thresholding method using a minimum weighted squareddistortion criterion," Pattern Recognition 28, pp.1063–1071, 1994
[19]Hui-Fuang Ng, "Automatic thresholding for defect detection," Pattern Recognition Letters, pp.1644–1649, 2006
[20]Nobuyuki Otsu, "A Threshold Selection Method from Gray-Level Histograms," IEEE Transaction on System Man Cybemet SNC-9 (1), pp.62-66,1979
[21]J.M.S. Prewitt and M. L. Mendelsolhn, "The analysis of cell images," Acad. Sci., vol. 128, pp.1035-1053, 1966
[22]Y. Qiao, Q.M. Hu, G.Y. Qian, S.H. Luo and W.L. Nowinski, "Thresholding based on variance and intensity contrast," Pattern Recognition 40, pp.596–608, 2007
[23]T.W. Ridler and S. Calvard, "Picture thresholding using an iterative selection method," IEEE Trans. Systems Man Cybernet. 8, pp.630–632, 1978
[24]P.L. Rosin and L. Paul, "Unimodal thresholding," Pattern Recognition 34, pp.2083-2096, 2001
[25]P.K. Sahoo, S. Soltani and A.K.C. Wong, "A survey of thresholding techniques," Comput Vision Graphics Image Process. 41, pp.233–260 , 1988
[26]J. Sauvola and M. Pietaksinen, "Adaptive document image binarization," Pattern Recognition 33, pp.225–236, 2000
[27]M. Sezgin and B. Sankur, "Survey over image thresholding techniques and quantitative performance evaluation," J. Electron. Imaging 13, pp.146–168, 2004
[28]W.H. Tsai, "Moment-preserving thresholding: A new approach," Graph. Models Image Process. 19, pp.377–393, 1985
[29]J. S. Weszka, R. N. Nagel, and A. Rosenfeld, "A threshold selection technique," IEEE Trans. Compute., vol. C-23, pp.1322 -1326, 1974
[30]S. Watanabe and CYBEST Group. "An automated apparatus for cancer prescreening: CYBEST," Comp. Graph. Imiage Process. vol. 3, pp. 350-358, 1974
[31]Xiangyang Xu , Shengzhou Xu, Lianghai Jin and Enmin Song, "Characteristic Analysis of Otsu Threshold and its Applications," Pattern Recognition Letters 32, pp.956–961, 2011
[32]Yonghong Xie, "A new efficient ellipse detection method," IEEE Trans., pp.957, 2002
[33]"Image Transforms - Hough Transform". Homepages.inf.ed.ac.uk. Retrieved 2009
[34]盛天企業, PCB生產流程簡介,
smt.order168.com/Index%20part%20html/PCB/PCB%20MP.html
[35]瞱展電機, PCB名詞解釋,
www.researchmfg.com/2011/07/pth-blind-hole-buried-hole/
[36]CODE PROJECT,
www.codeproject.com/Articles/336915/Connected-Component-Labeling-Algorithm
[37]Canny邊緣偵測演算法
docs.opencv.org/doc/tutorials/imgproc/imgtrans/canny_detector/canny_detector.html
[38]Sony攝影機, XC-ST50
www.sony.co.jp/ISP/
[39]鏡頭MML-ST Series
www.integro-tech.com/product-category/moritex-schott-3/mml-st-series/