本論文提出一種以資料點灰階權重之密度分群演算法:GWDBSCAN (Grey Weighting Density-Based Clustering Algorithm of Applications with Noise) 。本演算法以四象限資料點灰階值權重來當導引之概念，且成長方向由資料點的個數轉為資料點上之灰階值，並融入虛擬核心點來得到未分類資料點權重趨勢，快速移動且盡可能排除已分類點，達到快速資料分群。另提出演算法配合標準差倍率法分割背景並應用於LAO表面瑕疵檢測。另檢測主要項目為研磨液的殘留率。從光學顯微鏡中取得原影像計算出平均值後得出標準差。依實驗結果使用五倍標準差來決定門檻值可得到最佳顯現的殘留瑕疵。最後再使用本論文提出之GWDBSCAN對影像資料進行分群動作，由實驗結果，每張1280*1024pixels之LAO殘留瑕疵影像能有效標記範圍大小及位置，且符合所訂定的品質規範>10μm，並加入顏色與框線數字能更了解瑕疵資訊。

This thesis proposes Grey Weighting Density-Based Clustering Algorithm of Applications with Noise (GWDBSCAN). The algorithm uses data point gray value in four-quadrant for next growing concept of guidance. The growth direction is determined by the data points on the gray value, and the density-based clustering uses unclassified data points distribution direction of the trend to determine next virtual core point. The standard deviation of the image is applied to identify the background level for the LAO surface defect detection. The main inspected item is the slurry residue detection. The average standard deviation of the original image from the optical microscope is calculated. According to experimental results using five standards deviation to determine the threshold value can get best residual defects. Finally, this paper presents the GWDBSCAN clustering of image data. From the experimental results, each 1280*1024pixels of image with LAO residual defect can effectively locate position the defect and area. It also makes the grouping result more noticeable by using different color for understand any defect information. The defect conforms to the quality standard set more than 10μm.

[1] 聯合國氣候變化綱要公約(UNFCCC)，
http://unfccc.int/kyoto_protocol/status_of_ratification/items/2613.php, 2012
[2] International Energy Agency (IEA), http://www.iea.sp.gov.br/out/index.php, World Energy Outlook 2010, Paris.
[3] 英國國會科學與技術辦公室專題報導編號351， http://uk.nsc.gov.tw/ct.asp?xItem=0990427021&ctNode=785&lang=C, 2010
[4] LED產業2012年趨勢與展望報告，http://www.ledinside.com.tw/gold_member_2011,2011
[5] 蘇慧瑄，「台灣LED產業上中下游專利佈局之比較研究」，國立政治大學智慧財產研究所，2007。
[6] 陳炤彰、周炳伸、馬培盛、李俊誼、枋明輝，「可水解基板平坦化加工分析及應用」,機械月刊第三十七卷第十期，2011。
[7] MoneyDJ理財網， http://www.moneydj.com/kmdj/wiki/wikiviewer.aspx?keyid=2c2ef994-4cac-4cd5-8637-818a058f4730
[8] Chuan-Yu Chang, Chun-Hsi Li, Chin-Huang Chang and MuDer Jeng,” Learning Vector Quantization Neural Networks for LED Wafer Defect Inspection”, International Journal of Innovative Computing, Information and Control Volume 4, Number 10, October 2008.
[9] 張嘉偉，「使用自組式類神經網路在晶圓缺陷檢測系統」，國立雲林科技大學資訊工程研究所碩士班碩士論文，2005。
[10] 謝祥文，「區域分割為基礎之晶圓缺陷圖樣辨識演算法」，國立清華大學工業工程與工程管理研究所博士論文，2002。
[11] 石又尹，「以小波轉換演算法建構半導體晶圓缺陷圖樣辨識系統」，國立清華大學，工業工程與工程管理研究所碩士論文，2007。
[12] 鐘崇毓，「LED晶粒表面瑕疵之自動化視覺檢測」，朝陽科技大學工業工程與管理系碩士論文，2007。
[13] 林正偉，「賈柏轉換應用於晶圓晶片之可見瑕疵檢測」，國立臺北科技大學工業工程與管理系所碩士論文，2004。
[14] 林松濱，「裸晶粒自動化視覺檢測模組之開發」，國立台灣科技大學機械工程研究所碩士論文，2011。
[15] 王聖儒，「渦輪葉片的智慧型磁粒檢測方法」，國立中山大學機械工程研究所碩士論文，1999。
[16] 鐘國亮，「影像處理與電腦視覺」，東華書局，2006。
[17] 桂楚華、林清河，「全面品質管理與六標準差」，華泰文化，2008。
[18] 林育臣，「群聚技術之研究」，朝陽科技大學資訊管理研究所碩士論文，2002
[19] J. MacQueen, “Some Methods for Classification and Analysis of Multivariate Observations, ” In Proc. 5th Berkeley Symp. Math. Stat. and Prob., Vol. 1, pp. 281-297, 1967.
[20] Guha, Sudipto, Rajeev Rastogi, and Kyuseok Shim, “ROCK: A Robust Clustering Algorithm for Categorical Attributes,” IEEE, Conference on Data Eng., 1999.
[21] M. Ester, H.P. Kriegel, J. Sander and X. Xu, “A density-based algorithm for discovering clusters in large spatial databases with noise, ” Proceedings of the Second International Conference on Knowledge Discovery and Data Mining, Portland, Oregon, USA, pp. 226-231, 1996.
[22] W. Wang, R. Yang and R. Muntz, “STING: A Statistical Information grid Approach to Spatial Data Mining,” In Proc. 1997 Int. Conf. Very Large Data Bases(VLDB’97), Athens, Greece, pp. 186-195, August, 1997.
[23] X. Xu, M. Ester, H.P. Kriegel and J. Sander, “A distribution– based Clustering Algorithm for Mining in Large Spatial Databases,” In Proc. 14th Int. Conf. Data Engineering (ICDE'98),Orlando, Florida, USA, 1998.
[24] M. Ester, H.P. Kriegel, J. Sander, M. Wimmer and X. Xu, “Incremental Clustering for Mining in a Data Warehousing Environment,” In Proc. 24th Int. Conf. Very Large Databases (VLDB'98), New York City, USA, pages 24 - 27, August, 1998.
[25] M. Ankerst, M. Breunig, H.P. Kriegel, and J. Sander, “OPTICS: Order-ing points to identify the clustering structure,” In Proc. 1999 ACM-SIGMOD Int. Conf. Management of Data (SIGMOD’99), pages 49-60, Philadelphia, PA, June 1999.
[26] B. Borah and D.K. Bhattacharyya, “An Improved Sampling-Based DBSCAN for Large Spatial Databases,” Intelligent Sensing and Information Processing, pp. 92-96, 2004.
[27] D. Ma and A. Zhang, “An Adaptive Density-Based Clustering Algorithm for Spatial Database with Noise,” Data Mining, ICDM '04. Fourth IEEE International Conference, Brighton, United Kingdom, pp.467–470, Nov. 01-04, 2004.
[28] P. Viswanath and R. Pinkesh, “l-DBSCAN : A Fast Hybrid Density Based Clustering Method,” Pattern Recognition, 2006. ICPR 2006. 18th International Conference on Volume 1, Hong Kong, pp. 912-915, August 20-24, 2006.
[29] Y.P. Wu, J.J. Guo and X.J. Zhang, “A Linear DBSCAN Algorithm Based on LSH,” Proceedings of the Sixth International Conference on Machine Learning and Cybernetics, Hong Kong, pp. 2608-2614, August, 2007.
[30] A. Gionis, P. Indyk, and R. Motwani, “Similarity Search in High Dimensions via Hashing,” Proc. Very Large Data Base Conf. (VLDB '99), pp. 518-529, Sept, 1999.
[31] N. Beckmann, H.P. Kriegel, R. Schneider, and B. Seeger, “The R*-Tree: An Efficient and Robust Access Method for Points and Rectangles,” Proc. ACM SIGMOD Conf. Management of Data, Atlantic City, NJ, 1990.
[32] S. Berchtold, D. Keim, and H.P. Kriegel, “The X-Tree: An Index Structure for High-Dimensional Data,” Proc. 22nd Conf. Very Large Data Bases, Taj Mahal Hotel, Mumbai (Bombay), India, pp. 28-39. 1996.
[33] Y. Chen, L. Tu, “Density-Based Clustering for Real-Time Stream Data.” KDD’07, San Jose, California, USA, pp. 133-142, August 12–15, 2007.
[34] L. Duan, L. Xu, F. Guo, J. Lee, B. Yan, “A local-density based spatial clustering algorithm with noise,” Information Systems Vol.32, pp. 978-986, 2007.
[35] B. Liu, “A Fast Density-Based Clustering Algorithm for Large Databases,” Proceedings of the Fifth International Conference on Machine Learning and Cybernetics, Dalian, pp. 996-1000, August, 2006.
[36] 吳宗諭、蔡明忠、田芬寧，「以密度為基礎之改良型快速分群演算法及其於偏光板瑕疵檢測之應用」，技術學刊Vol.24，No.4，pp. 267-274，2009，12月。
[37] 蔡明忠、田芬寧、林家戊，「以密度重心導引之分群演算法及其於抗反射玻璃瑕疵檢測之應用」，Taiwan AOI Forum, Show and Contest，台灣科技大學，台北，十一月，2009。
[38] 蔡明忠、林家戊、楊明輔,「整合密度趨勢與虛擬核心點之密度分群演算法及應用於太陽能電池背面瑕疵檢測」, Taiwan AOI Forum, Show and Contest，台灣科技大學，台北，十一月，2010。