本研究為智慧型感光元件之封裝級玻璃瑕疵檢測，主要設計開發一套自動化光學檢測系統。而本研究自行設計開發之硬體架構系統含括電荷耦合元件(charge coupled device, CCD)相機、X-Y-Z軸移動平台、光源調整器以及正、背光源。而系統之光源架構為，背光源加上偏光膜搭配正光源以突顯玻璃蓋片瑕疵特徵，其中，偏光膜主要是避免擷取之影像容易產生過曝，並結合CCD相機擷取玻璃蓋片影像，並以本研究自行提出之多層影像(multi-layer image)亮度校正方法，進行亮度校正，解決亮度不均之問題。運用影像金字塔降低影像資訊量及正規化相關係數法模板比對來進行玻璃之定位，並分別對封裝級玻璃之光阻區與非光阻區進行檢測，在光阻區利用歐蘇(Otsu)門檻值決定法，將封裝級玻璃之光阻，從影像中分割出來，並計算其光阻殘缺之面積。而非光阻區則分為高對比度瑕疵(亮點、暗點以及刮傷)與低對比度瑕疵(溶液殘留)進行檢測，高對比度瑕疵使用最大熵值法來進行瑕疵影像分割，而低對比度瑕疵則是先使用直方圖等化提高影像對比度，突顯其瑕疵特徵，再搭配最大熵值法來進行瑕疵影像分割，結合支持向量機來進行非光阻區瑕疵分類辨識(亮點、暗點、刮傷以及溶液殘留)，最後，結合電腦人機控制介面、影像擷取設備、光源設備及電控系統，以達到感光元件之封裝級玻璃瑕疵檢測系統軟硬體整合之自動化檢測目的，本研究之自動化光學檢測能夠檢測出5um以下的瑕疵，且實驗結果顯示，瑕疵分類準確率98.97%，可滿足產業界自動化光學檢測系統上之需求

This study developed an intelligent automated optical inspection system for packaging level glass defects of photosensitive elements. The light source in this study is consisted of back light source, polarizing film and positive light source, for highlighting the cover glass defect characteristic. The polarizing film avoids overexposure of the captured image, and the CCD camera captures the cover glass image. The image pyramid is used to reduce the image information content and normalize template matching of the correlation coefficient method to position the glass. The photoresist area and non-photoresist area of the packaging level glass are detected. In the photoresist area, the Otsu threshold determination is used to separate the photoresist of packaging level glass from the image, and the photoresist breakage area is calculated. In the non-photoresist area, the high-contrast defects (bright spot, dark spot and scratch) and low-contrast defect (solution residue) are detected. For the high-contrast defects, the maximum entropy is used for defect image segmentation. For the low-contrast defect, the histogram equalization is used to increase the image contrast to highlight the defect characteristic, and then the maximum entropy is used for defect image segmentation. The support vector machine is used for non-photoresist area defect classification identification (bright spot, dark spot, scratch and solution residue). Finally, the computer man-machine control interface, image capture equipment, light source equipment and electric control system are combined, so as to im-plement the software and hardware integrated automatic detection of the packaging level glass defect detection system for photosensitive elements. The automated optical inspection in this study can detect defects less than 5 um. The experimental results showed that the defect classification accuracy rate is 98.97%.

參考文獻
[1] 米本和也，「CCD/CMOS 影像感測器之基礎與應用」，全華圖書股份有限公司，臺灣臺北，2006。
[2] Boyle, W. S., and Smith, G. E., “Charge-Coupled Devices - a New Approach to Mis Device Structures”, IEEE spectrum, Vol. 8, No. 7, pp. 18-27, 1971.
[3] Belbachir, A. N., Lera, M., Fanni, A., and Montisci, A., “An Automatic Optical Inspection System for the Diagnosis of Printed Circuits Based on Neural Networks”, Industry Applications Conference, Hong Kong, Vol. 1, pp. 680-684, 2005.
[4] 陳宗達，「CMOS玻璃蓋片自動光學檢測機台之設計及開發」，國立交通大學，工業工程及管理研究所，碩士論文，2005。
[5] Chang, C. Y., Chang, C. H., Li, C. H., and Jeng, M. D., “Learning Vector Quantization Neural Networks for Led Wafer Defect Inspection”, International Journal of Innovative Computing, Information and Control, Vol. 4, No. 10, pp. 2565-2579, 2008.
[6] Chen, L. F., Su, C. T., and Chen, M. H., “A Neural-Network Ap-proach for Defect Recognition in TFT-LCD Photolithography Process”, IEEE Transactions on Electronics Packaging Manufacturing, Vol. 32, No. 1, pp. 1-8, 2009.
[7] Francesco, A., Filippo, A., and Attilio, D. N., “Calibration of an Inspection System for Online Quality Control of Satin Glass”, IEEE Transactions on Instrumentation and Measurement, Vol. 59, No. 5, pp. 1035-1046, 2010.
[8] Stockham, T. G. Jr., “Image Processing in the Context of a Visual Model”, Proceedings of the IEEE, Vol. 60, No. 7, pp. 828-842, 1972.
[9] Wang, M., Pan, J., Chen, S., and Li, H., “A Method of Removing the Uneven Illumination Phenomenon for Optical Remote Sensing Image”, Geoscience and Remote Sensing Symposium, Vol. 5, Korea, 2005.
[10] 張紀鈴、馮曉毅、夏超，「AOI檢測系統光照不均的校正方法研究」，電子量測技術 (Eletronic Measurement Tech-nology)，第七期，第三十卷，pp.20-23，2007。
[11] Kim, T. Y., “Contrast Enhancement Using Brightness Preserving Bi-Histogram Equalization”, IEEE Transactions on Consumer Electronics, Vol. 3, No. 1, pp. 1-8, 1997.
[12] Kim, J. Y., Kim, L. S., and Hwang, S. H., “An Advanced Contrast Enhancement Using Partially Overlapped Sub-Block Histogram Equalization”, IEEE Transactions on Circuits and Systems for Video Technology, Vol. 11, No. 4, pp. 475-484, 2001.
[13] Wang, C., and Ye, Z., “Brightness Preserving Histogram Equalization with Maximum Entropy: A Variational Perspective”, IEEE Transactions on Consumer Electronics, Vol. 51, No. 4, pp. 1326-1334, 2005.
[14] 李界磐，「以影像亮度為基礎之適應性影像增強法的研究」，國立台灣科技大學，電機工程學系，碩士論文，2007。
[15] Turgay, C., and Tardi, T., “Automatic Image Equalization and Contrast Enhancement Using Gaussian Mixture Modeling”, IEEE Transactions on Image Processing, Vol. 21, No. 1, pp. 145-156, 2012.
[16] Hasikin, K., and Isa, N. A. M., “Enhancement of the Low Contrast Image Using Fuzzy Set Theory”, 2012 UKSim 14th International Conference on Computer Modelling and Simulation (UKSim), pp. 371-376, Cambridge, United Kingdom, 2012.
[17] Chen, F. L., and Liu, S. F.,“A Neural-Network Approach to Recognize Defect Spatial Pattern in Semiconductor Fabrication”, IEEE Transactions on Semiconductor Manufacturing, Vol. 13, No. 3, pp. 366-373, 2000.
[18] Feng, H., Ye, J., and Pease, R. F., “Self Inspection of Integrated Circuits Pattern Defects Using Support Vector Machines”, Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures, Vol. 23, No. 6, pp. 3085-3089, 2005.
[19] Acciani, G., Brunetti, G., and Fornarelli, G., “Application of Neural Networks in Optical Inspection and Classification of Solder Joints in Surface Mount Technology”, IEEE Transactions on Industrial Informatics, Vol. 2, No. 3, pp. 200-209, 2006.
[20] Chang, C. Y., Li, C. H., Lin, S. Y., and Jeng, M.,“Application of Two Hopfield Neural Networks for Automatic Four-Element LED Inspection”, IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews, Vol. 39, No. 3, pp. 352-365, 2009.
[21] Duan, G., Wang, H., Liu, Z., and Chen, Y. W., “A Machine Learning-Based Framework for Automatic Visual Inspection of Microdrill Bits in PCB Production”, IEEE Transactions on Sys-tems, Man, and Cybernetics, Part C: Applications and Reviews, Vol. 42, No. 6, pp. 1679-1689, 2012.
[22] Wu, H., Zhang, X., Xie, H., Kuang, Y., and Ouyang, G., “ Classification of Solder Joint Using Feature Selection Based on Bayes and Support Vector Machine”, IEEE Transactions on Components, Packaging and Manufacturing Technology, Vol. 3, No. 3, pp. 516-522, 2013.
[23] Kelly, M. C., Croarken, M., Flood, R., and Robson, E., “The History of Mathematical Tables”, Oxford University Press, New York, USA, 2003.
[24] 鍾國亮，「影像處理與電腦視覺(第四版)」，東華書局，臺灣台北，2008。
[25] Gonzalez, R. G., and Woods, R. E., “Digital Image Processing 3rd Edition”, Prentice Hall, Taiwan, Taipei, 2008.
[26] Tanimoto, S. L., “Template Matching in Pyramids”, Computer Graphics and Image Processing, Vol. 16, No. 4, pp. 356-369, 1981.
[27] Tsai, D. M., and Lin, C. T., “Fast Normalized Cross Correlation for Defect Detection”, Pattern Recognition Letters, Vol. 24, No. 15, pp. 2625-2631, 2003.
[28] Mattoccia, S., Tombari, F. Di., and Stefano, L., “Fast Full-Search Equivalent Template Matching by Enhanced Bounded Correlation”, IEEE Transactions on Image Processing, Vol. 17, No. 4, pp. 528-538, 2008.
[29] Otsu, N., “A Threshold Selection Method from Gray Level Histogram”, IEEE Transactions on Systems, Man, and Cybernetics, Vol. 9, No. 1, pp. 62-66, 1979.
[30] Kapur, J. N., Sahoo, P. K., and Wong, A. K. C., “A New Method for Gray-Level Picture Thresholding Using the Entropy of the Histogram”, Computer Vision, Graphics, and Image Processing, Vol. 29, pp. 273-285, 1985.
[31] Ridler, T. W., and Calvard, S., “Picture Thresholding Using an Iterative Selection Method”, IEEE Transactions on Systems, Man, and Cybernetics, Vol. 8, No. 8, pp. 630-632, 1978.
[32] 蘇春木、張孝德，「機器學習：類神經網路、模糊系統以及基因演算法則」，全華圖書股份有限公司，臺灣臺北，2007。
[33] Rumelart, D. E., Hinton, G. E., and Williams, R. J., “Learning Representations by Back-Propagating Errors”, Nature, Vol. 323, No. 9, pp. 533-536, 1986.
[34] Cortes, C., and Vapnik, V., “Support-Vector Network”, Machine Learning, Vol. 20, No. 3, pp. 273-297, 1995.
[35] Vapnik, V. N., “The Nature of Statistical Learning Theory”, Springer-Verlag New York, Inc., New York, USA, 1995.
[36] Vapnik, V. N., “Statistical Learning Theory”, John Wiley & Sons, Inc., New York, 1998.
[37] Bellman, R., “Dynamic Programming and Lagrange Multipliers”, Proceedings of the National Academy of Sciences of the United States of America, Vol. 42, No. 10, pp. 767-769, 1956.
[38] 陳建堂，「偏光膜外觀瑕疵之影像檢測系統開發」，國立台灣科技大學，高分子系，碩士論文，2008。
[39] 葉怡成，「類神經網路模式應用與實作」，儒林書局，臺灣台北，2003。
[40] 羅華強，類神經網路-MATLAB的應用，高立圖書股份有限公司，臺灣台北，2005。
[41] 連國珍，數位影像處理MATLAB，儒林圖書公司，臺灣台北，2007。