印刷電路板(Printed circuit board, PCB)在電子工業佔著舉足輕重的地位，在先進檢測中，基於機器視覺的自動化光學檢測（Automated Optical Inspection, AOI）為重要的方法，高效與可靠的影像視覺檢測成為AOI重要組成部分，其中影像清晰度(Image Sharpness)為關鍵的檢測標準，經過影像融合能夠有效的進一步提升PCB的檢測影像清晰度，並有效減少檢測的重工率進而提高良率。影像拼接(Image Stitching)技術擁有多項優點，其中可有效的在擷取PCB上每一個區塊影像後並結合的方法。另外受限於光學鏡頭的場景深度(Depth of focus, DOF)限制，PCB每一區塊只有在景深限制內被拍攝時具清晰的成像，反之則是模糊的影像。因此現有CCD不容易拍攝出完整PCB所有區域皆清晰的影像，而本研究所提出的多重對焦影像融合技術可以解決這個問題，將多張具有相同場景、不同深度的影像融合成出一張所有物體相對清晰的影像。本論文提出之影像融合(Image Fusion)方法，利用圖像清晰度擷取以及PCB作分類切割圖像的多重對焦影像融合方法，並且與傳統影像融合演法結果圖比較清晰度。首先，分類出三種不同規格的PCB影像的特徵，並將多張具有相同場景、不同深度對焦PCB影像分別進行圖像切割，計算圖像對區域清晰度作分類，再來依分類的結果擷取出不同深度對焦影像的最佳清晰度PCB影像區塊，進行影像融合取得合成影像為實驗結果。所提出方法可改善傳統影像融合技術與少量對焦圖像融合的限制，並有效增強圖像的清晰度，進一步改善PCB每一個區塊影像品質，並利用影像拼接技術結合每一個PCB區塊圖成完整影像。

Printed circuit board (PCB) play an important role in the electronics industry. In advanced inspection, efficient and reliable image visual inspection is the most important part of machine vision-based automated optical inspection (AOI). Image sharpness is the key detection standard. Using image fusion to improve the image sharpness during the inspection process in order to reduce the rework rate and increase the yield. Image stitching has lots of advantage, capture every region of the PCB and reunion them is one of the advantages. In addition, due to the Depth of Focus (DOF) limitation of the optical lens, each region of the PCB will be clearly imaged only when it is captured within the depth of focus limit otherwise will a blur. Therefore, the existing CCD is not easy to capture clear images of all areas of the complete PCB, so the multi-focus image fusion technology proposed in this thesis can solve this problem by merging multiple images within the same scene and different depths into one image. The image fusion method proposed in this thesis uses a multi-focus image fusion method of image sharpness capture and PCB for images segmentation and compares the image sharpness with the traditional image fusion performance map. First of all, classified three different types of PCB and segment these images of PCB which shared with the same scene but different depths of the focus. And classified the image region depends on image sharpness. After that, select the best image sharpness in this region and merge them shown in the result of the experiment. The proposed method can improve the limitation of the traditional image fusion technology and a small number of image fusion, and effectively enhance the sharpness of the image, further improve the image quality of each block of the PCB, and merge the image of each PCB region into a complete image by using image stitching technology.

[1] TPCA"PCB市場2019年新機會！5G與AI將促其持續成長" Internet: https://medium.com/@trendforcemkt/拓樸觀點- PCB市場2019年新機會！5G與AI將促其持續成長-15d53beabcb1,Dec.24 ,2018[July.03, 2019]
[2]TPCA."PCB技術藍圖(2017-2021版)"Internet: https://www.tpca.org.tw/Message/MessageView?id=27&mid=125,May.02 ,2018[July.03, 2019]
[3] Fan, Jun, et al. "Evaluation of SMT decoupling design in a functioning high-speed PCB." 2001 IEEE EMC International Symposium. Symposium Record. International Symposium on Electromagnetic Compatibility (Cat. No. 01CH37161). Vol. 2. IEEE, 2001.pp. 1097-1101
[4] Li, Ziyin, and Qi Yang. "System design for PCB defects detection based on AOI technology." 2011 4th International Congress on Image and Signal Processing. Vol. 4. IEEE, 2011. pp. 1988-1991.
[5] Brown, Matthew, and David G. Lowe. "Automatic panoramic image stitching using invariant features." International journal of computer vision 74.1. 2007. pp. 59-73.
[6] Rahm, Erhard, and Philip A. Bernstein. "A survey of approaches to automatic schema matching." the VLDB Journal Vol.10. No. 4. 2001. pp. 334-350.
[7] Brown M, and Lowe. "Recognizing panoramas. "In Proceedings of the International Conference on Computer Vision. Vol. 3. 2003, pp. 1218–1225.
[8] Harris, Christopher G., and Mike Stephens. "A combined corner and edge detector." Alvey vision conference. Vol. 15. No. 50. 1988. pp. 10-5244.
[9] Liu, Yu, Shuping Liu, and Zengfu Wang. "Multi-focus image fusion with dense SIFT." Information Fusion 23 (2015): pp. 139-155.
[10] Canny, John. "A computational approach to edge detection." Readings in computer vision. Morgan Kaufmann, 1987. pp. 184-203.
[11] HUANG, You-qun, Yu Fu, and Guang-kun MA. "Cylindrical panoramic image stitching method based on RANSAC algorithm." Journal of Shenyang University of Technology vol. 30, no. 4,2008. pp.46 1-465.
[12] He, Kaiming, Huiwen Chang, and Jian Sun. "Rectangling panoramic images via warping." ACM Transactions on Graphics (TOG) 32.4 (2013): pp. 79.
[13] Kavan, Ladislav, et al. "Dual quaternions for rigid transformation blending." Trinity College Dublin (2006).
[14] Yen, Jonathan, William C. Kress, and Robert Poe. "Digital image tone adjustment." U.S. Patent No. 8,483,508. 9 Jul. 2013.
[15] Duan, Jiang, et al. "Tone-mapping high dynamic range images by novel histogram adjustment." Pattern Recognition vol. 43, no. 5,2010. pp. 1847-1862.
[16] Antunes, Mark, et al. "All-in-focus imaging using a series of images on different focal planes." International Conference Image Analysis and Recognition. Springer, Berlin, Heidelberg, 2005, pp174-181.
[17] Levin, Anat, et al. "Image and depth from a conventional camera with a coded aperture." ACM Transactions on Graphics (TOG)26.3 (2007): pp. 70.
[18] Liang, Lin, et al. "Real-time texture synthesis by patch-based sampling." ACM Transactions on Graphics (TOG) 20.3 (2001): pp. 127-150.
[19] Zhong, Zhang, and Rick S. Blum. "A categorization of multiscale-decomposition-based image fusion schemes with a performance study for a digital camera application." Proceedings of the IEEE87.8 (1999): pp. 1315-1326.
[20]Zhao, Hengjun, et al. "Multi-focus image fusion based on the neighbor distance." Pattern Recognition 46.3 (2013): pp. 1002-1011.
[21] Saha, Ashirbani, Gaurav Bhatnagar, and QM Jonathan Wu. "Mutual spectral residual approach for multifocus image fusion." Digital Signal Processing 23.4 (2013): pp. 1121-1135.
[22] Liu, Yu, Shuping Liu, and Zengfu Wang. "Multi-focus image fusion with dense SIFT." Information Fusion 23 (2015): pp. 139-155.
[23] D.K. Sahu, and M.P. Parsai. "Different image fusion techniques–a critical review." International Journal of Modern Engineering Research (IJMER) 2.5 (2012): pp. 4298-4301.
[24] Huang, Wei, and Zhongliang Jing. "Evaluation of focus measures in multi-focus image fusion." Pattern recognition letters 28.4 (2007): pp. 493-500.
[25] Pajares, Gonzalo, and Jesus Manuel De La Cruz. "A wavelet-based image fusion tutorial." Pattern recognition 37.9 (2004): pp. 1855-1872.
[26] Pei, Lili, Yaohong Zhao, and Haibo Luo. "Application of wavelet-based image fusion in image enhancement." 2010 3rd International Congress on Image and Signal Processing. Vol. 2. IEEE, 2010. pp. 649-653.
[27]Drori, Iddo, Daniel Cohen-Or, and Hezy Yeshurun. "Fragment-based image completion." ACM Transactions on Graphics (TOG). Vol. 22. No. 3. ACM, 2003. pp. 303-312.
[28] Chien, Chun-Liang, and Din-Chang Tseng. "Color image enhancement with exact HSI color model." International Journal of Innovative Computing, Information and Control 7.12 (2011): pp. 6691-6710.
[29] Uyttendaele, Matthew, Ashley Eden, and Richard Skeliski. "Eliminating ghosting and exposure artifacts in image mosaics." Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001. Vol. 2. IEEE, 2001. pp. 509–516.
[30] Zhang, Fan, and Feng Liu. "Parallax-tolerant image stitching." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2014. pp. 3262-3269.
[31] Singh, Kuldeep, and Rajiv Kapoor. "Image enhancement using exposure based sub image histogram equalization." Pattern Recognition Letters 36 (2014): pp. 10-14.
[32]Dennis P. Curtin. "USING YOUR DIGITAL CAMERA." "Internet:http://www.shortcourses.com/use/using3-7.html,2007[July.03, 2019].
[33] Park, Tae-Hyoung, and Hwa-Jung Kim. "Path planning of automatic optical inspection machines for PCB assembly systems." 2005 International Symposium on Computational Intelligence in Robotics and Automation. IEEE, 2005. pp. 249-254.
[34] Li, Hui, B. S. Manjunath, and Sanjit K. Mitra. "Multisensor image fusion using the wavelet transform." Graphical models and image processing 57.3 (1995): pp. 235-245.
[35] Zhao, Hengjun, et al. "Multi-focus image fusion based on the neighbor distance." Pattern Recognition 46.3 (2013): pp. 1002-1011.
[36] Huang, Wei, and Zhongliang Jing. "Evaluation of focus measures in multi-focus image fusion." Pattern recognition letters 28.4 (2007): pp. 493-500.
[37] Saha, Ashirbani, Gaurav Bhatnagar, and QM Jonathan Wu. "Mutual spectral residual approach for multifocus image fusion." Digital Signal Processing 23.4 (2013): pp. 1121-1135.
[38] Piella, Gemma. "A region-based multiresolution image fusion algorithm." Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002.(IEEE Cat. No. 02EX5997). Vol. 2. IEEE, 2002, pp.1557-1564.
[39] Chen, Cheng-I., Phen-Lan Lin, and Po-Whei Huang. "A new fusion scheme for multi-focus images based on dynamic salient weights on discriminative edge points." 2010 International Conference on Machine Learning and Cybernetics. Vol. 1. IEEE, 2010, pp. 351-356.
[40] Qiang, Ge, et al. "Detection method of PCB component based on automatic optical stitching algorithm." Circuit World 41.4 (2015): pp. 133-136.