現今印刷電路板(Printed Circuit Board, PCB)產業蓬勃發展，生產逐漸精密化且線寬越來越小，為了提升印刷電路板製程的品質，必須要進行有效率的檢測。傳統是以人工目視辨別印刷電路板上的缺陷，不僅耗時且誤判率高，因此業界開始應用「自動光學檢測」(Automated Optical Inspection，AOI)設備來檢測產品缺陷，檢測出各類缺陷之後，若要將各類光阻圖案缺陷進行分類，需要提出另一種有效率的人工智慧模型，來針對數位微影所產生的光阻圖案缺陷進行分類與辨識，以達到產線最即時的回饋與修正。本文提出一種卷積神經網路(Convolution Neural Network, CNN)應用在影像的辨識與分類，針對六種最為常見的光阻圖案缺陷進行分類，並且在分類缺陷的任務中獲得將近90%的準確率，證明卷積神經網路是可以被應用在光阻圖案缺陷分類上。本文亦於台科大實驗室自行搭建的數位微影(Digital Lithography)曝光機台，採用數位微反射鏡裝置(Digital Micromirror Device , DMD)，去進行實際曝光的光阻圖案特徵探討，包含二維以及三維兩種主要型態的光阻特徵，並整合二維及三維的自動化光學檢測，再根據其光阻輪廓的不同，定義為不同的光阻缺陷，透過本文所開發的缺陷分類模型進行缺陷分類，再根據數位微影缺陷修正程序，進行落點位置的修正，重新進行曝光實驗驗證修正結果，完成整套智慧化數位微影曝光與檢測程序的開發。

The Printed Circuit Board (PCB) industry is developing well. Process is becoming more sophisticated and the line width is getting smaller and smaller. In order to improve the quality of the printed circuit board manufacturing process, efficient inspection must be carried out. Traditionally, artificial vision is used to identify defects on printed circuit boards. It is not only very time-consuming, but has a high rate of misjudgment. Therefore, the industry began to use Automated Optical Inspection (AOI) equipment to detect product defects. After detecting different defects, if we need to classify different photoresist pattern defects, it is necessary to propose another efficient artificial intelligence model to classify and identify different types of photoresist pattern defects, and then achieve the most immediate feedback and correction of the production line. This dissertation proposes a convolutional neural network (CNN) application in the identification and classification of images, and classify the six most common photoresist pattern defects. This model achieves nearly 90% accuracy in the task of classifying defects. It also proves that convolutional neural networks can be applied to the classification of photoresist pattern defects. This research also built a digital lithography machine in Laboratory of National Taiwan University of Science and Technology, and we adopt Digital Micromirror Device (DMD) to analyze the characteristics of the photoresist pattern of the actual exposure experiments. The results include two main types of photoresist features, two-dimensional and three-dimensional, and we integrate 2D and 3D automated optical inspection. They will be defined as different photoresist defects according to different profile. Next, the defect classification model developed in this research is used to classify different defects, and then according to the digital lithography defect correction procedure to correct exposure position of beam, and conduct exposure experiments to verify the correction results. This research completed the development of the entire intelligent digital lithography exposure and inspection procedure.

[1] E. J. Hansotte, E. C. Carignan, and W. D. Meisburger, "High speed maskless lithography of printed circuit boards using digital micromirrors," in Emerging Digital Micromirror Device Based Systems and Applications III, 2011, vol. 7932: International Society for Optics and Photonics, p. 793207.
[2] J. H. Bruning, "Optical lithography: 40 years and holding," in Optical Microlithography XX, 2007, vol. 6520: International Society for Optics and Photonics, p. 652004.
[3] Q. Li and S. Ren, "A real-time visual inspection system for discrete surface defects of rail heads," IEEE Transactions on Instrumentation and Measurement, vol. 61, no. 8, pp. 2189-2199, 2012.
[4] H.-C. Liao, Z.-Y. Lim, Y.-X. Hu, and H.-W. Tseng, "Guidelines of automated optical inspection (AOI) system development," in 2018 IEEE 3rd International Conference on Signal and Image Processing (ICSIP), 2018: IEEE, pp. 362-366.
[5] W. S. Noble, "What is a support vector machine?," Nature biotechnology, vol. 24, no. 12, pp. 1565-1567, 2006.
[6] H. Shang, D. Sun, P. Yu, Q. Sun, J. Gao, and T. J. Hall, "Analysis for system errors in measuring the sidewall angle of a silica waveguide with confocal laser scanning microscope (CLSM)," Measurement Science and Technology, vol. 30, no. 2, p. 025004, 2019.
[7] M.-J. Wang and C.-L. Huang, "Evaluating the eye fatigue problem in wafer inspection," IEEE Transactions on Semiconductor Manufacturing, vol. 17, no. 3, pp. 444-447, 2004.
[8] H. Xiao et al., "Capturing buried defects in metal interconnections with electron beam inspection system," in Metrology, Inspection, and Process Control for Microlithography XXVII, 2013, vol. 8681: International Society for Optics and Photonics, p. 86810F.
[9] M. Saito, T. Hayashi, K. Fujihara, K. Saito, J. Lin, and R. Midorikawa, "Study of ADI (after develop inspection) using electron beam," in Metrology, Inspection, and Process Control for Microlithography XX, 2006, vol. 6152: International Society for Optics and Photonics, p. 615248.
[10] K. He, G. Gkioxari, P. Dollár, and R. Girshick, "Mask r-cnn," in Proceedings of the IEEE international conference on computer vision, 2017, pp. 2961-2969.
[11] G. Ghiasi, T.-Y. Lin, and Q. V. Le, "Nas-fpn: Learning scalable feature pyramid architecture for object detection," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp. 7036-7045.
[12] Y. Ren, C. Zhu, and S. Xiao, "Object detection based on fast/faster RCNN employing fully convolutional architectures," Mathematical Problems in Engineering, vol. 2018, 2018.
[13] M. Lokanath, K. S. Kumar, and E. S. Keerthi, "Accurate object classification and detection by faster-RCNN," in IOP Conference Series: Materials Science and Engineering, 2017, vol. 263, no. 5: IOP Publishing, p. 052028.
[14] P. Malge and R. Nadaf, "PCB defect detection, classification and localization using mathematical morphology and image processing tools," International journal of computer applications, vol. 87, no. 9, 2014.
[15] V. H. Gaidhane, Y. V. Hote, and V. Singh, "An efficient similarity measure approach for PCB surface defect detection," Pattern Analysis and Applications, vol. 21, no. 1, pp. 277-289, 2018.
[16] B. Kaur, G. Kaur, and A. Kaur, "Detection and classification of printed circuit board defects using image subtraction method," in 2014 Recent Advances in Engineering and Computational Sciences (RAECS), 2014: IEEE, pp. 1-5.
[17] S. Ren, K. He, R. Girshick, and J. Sun, "Faster r-cnn: Towards real-time object detection with region proposal networks," Advances in neural information processing systems, vol. 28, pp. 91-99, 2015.
[18] W. Liu et al., "Ssd: Single shot multibox detector," in European conference on computer vision, 2016: Springer, pp. 21-37.
[19] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, "You only look once: Unified, real-time object detection," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 779-788.
[20] X. Wang, A. Shrivastava, and A. Gupta, "A-fast-rcnn: Hard positive generation via adversary for object detection," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 2606-2615.
[21] T.-Y. Lin et al., "Microsoft coco: Common objects in context," in European conference on computer vision, 2014: Springer, pp. 740-755.
[22] M. Everingham, S. A. Eslami, L. Van Gool, C. K. Williams, J. Winn, and A. Zisserman, "The pascal visual object classes challenge: A retrospective," International journal of computer vision, vol. 111, no. 1, pp. 98-136, 2015.
[23] R. Girshick, "Fast r-cnn," in Proceedings of the IEEE international conference on computer vision, 2015, pp. 1440-1448.
[24] B. Hu and J. Wang, "Detection of PCB surface defects with improved faster-RCNN and feature pyramid network," IEEE Access, vol. 8, pp. 108335-108345, 2020.
[25] R. Gavrilescu, C. Zet, C. Foșalău, M. Skoczylas, and D. Cotovanu, "Faster R-CNN: an approach to real-time object detection," in 2018 International Conference and Exposition on Electrical And Power Engineering (EPE), 2018: IEEE, pp. 0165-0168.
[26] D. Chen, Z. Yuan, G. Hua, N. Zheng, and J. Wang, "Similarity learning on an explicit polynomial kernel feature map for person re-identification," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015, pp. 1565-1573.
[27] T.-Y. Lin, P. Dollár, R. Girshick, K. He, B. Hariharan, and S. Belongie, "Feature pyramid networks for object detection," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 2117-2125.
[28] S. Faghih-Roohi, S. Hajizadeh, A. Núnez, R. Babuska, and B. De Schutter, "Deep convolutional neural networks for detection of rail surface defects," in 2016 International joint conference on neural networks (IJCNN), 2016: IEEE, pp. 2584-2589.
[29] D. Soukup and R. Huber-Mörk, "Convolutional neural networks for steel surface defect detection from photometric stereo images," in International Symposium on Visual Computing, 2014: Springer, pp. 668-677.
[30] Y. J. Cha, W. Choi, G. Suh, S. Mahmoudkhani, and O. Büyüköztürk, "Autonomous structural visual inspection using region‐based deep learning for detecting multiple damage types," Computer‐Aided Civil and Infrastructure Engineering, vol. 33, no. 9, pp. 731-747, 2018.
[31] R. Ding, L. Dai, G. Li, and H. Liu, "TDD-net: a tiny defect detection network for printed circuit boards," CAAI Transactions on Intelligence Technology, vol. 4, no. 2, pp. 110-116, 2019.
[32] W. Huang and P. Wei, "A PCB dataset for defects detection and classification," arXiv preprint arXiv:1901.08204, 2019.
[33] J. Zhong, Z. Liu, Z. Han, Y. Han, and W. Zhang, "A CNN-based defect inspection method for catenary split pins in high-speed railway," IEEE Transactions on Instrumentation and Measurement, vol. 68, no. 8, pp. 2849-2860, 2018.
[34] A. Kumar, "Computer-vision-based fabric defect detection: A survey," IEEE transactions on industrial electronics, vol. 55, no. 1, pp. 348-363, 2008.
[35] H.-F. Ng, "Automatic thresholding for defect detection," Pattern recognition letters, vol. 27, no. 14, pp. 1644-1649, 2006.
[36] X. Xie, "A review of recent advances in surface defect detection using texture analysis techniques," ELCVIA: electronic letters on computer vision and image analysis, pp. 1-22, 2008.
[37] D. Forsyth and J. Ponce, Computer vision: A modern approach. Prentice hall, 2011.
[38] E. Hjelmås and B. K. Low, "Face detection: A survey," Computer vision and image understanding, vol. 83, no. 3, pp. 236-274, 2001.
[39] H. Y. Ngan, G. K. Pang, and N. H. Yung, "Automated fabric defect detection—a review," Image and vision computing, vol. 29, no. 7, pp. 442-458, 2011.
[40] X. Wei, Z. Yang, Y. Liu, D. Wei, L. Jia, and Y. Li, "Railway track fastener defect detection based on image processing and deep learning techniques: A comparative study," Engineering Applications of Artificial Intelligence, vol. 80, pp. 66-81, 2019.
[41] J. Sauvola and M. Pietikäinen, "Adaptive document image binarization," Pattern recognition, vol. 33, no. 2, pp. 225-236, 2000.
[42] D. A. Van Dyk and X.-L. Meng, "The art of data augmentation," Journal of Computational and Graphical Statistics, vol. 10, no. 1, pp. 1-50, 2001.
[43] D. M. Hawkins, "The problem of overfitting," Journal of chemical information and computer sciences, vol. 44, no. 1, pp. 1-12, 2004.
[44] S. Sukhbaatar, A. Szlam, J. Weston, and R. Fergus, "End-to-end memory networks," arXiv preprint arXiv:1503.08895, 2015.
[45] J. Wang, Y. Yang, J. Mao, Z. Huang, C. Huang, and W. Xu, "Cnn-rnn: A unified framework for multi-label image classification," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 2285-2294.
[46] S. Ioffe and C. Szegedy, "Batch normalization: Accelerating deep network training by reducing internal covariate shift," in International conference on machine learning, 2015: PMLR, pp. 448-456.
[47] X. Glorot, A. Bordes, and Y. Bengio, "Deep sparse rectifier neural networks," in Proceedings of the fourteenth international conference on artificial intelligence and statistics, 2011: JMLR Workshop and Conference Proceedings, pp. 315-323.
[48] Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, "Gradient-based learning applied to document recognition," Proceedings of the IEEE, vol. 86, no. 11, pp. 2278-2324, 1998.
[49] B. C. Russell, A. Torralba, K. P. Murphy, and W. T. Freeman, "LabelMe: a database and web-based tool for image annotation," International journal of computer vision, vol. 77, no. 1-3, pp. 157-173, 2008.
[50] G. M. Gallatin, "Resist blur and line edge roughness," in Optical Microlithography XVIII, 2005, vol. 5754: International Society for Optics and Photonics, pp. 38-52.