本研究結合自動光學檢測、雷射加工技術與智能化技術，發展一套智動化的電路板修復方法。使用者放置有缺陷的電路板於平台上，透過影像拍攝與處理，獲得缺陷端點座標資訊後，本研究所開發的程式，即能自動快速規劃一個考量端點效應(熱累積效應)的最佳雷射掃描路徑，以達到高效率且精準穩定修復電路板缺陷的目標。
為拓展技術的應用範圍，本研究進一步建立考慮端點效應(熱累積效應)的雷射掃描路徑規劃技術，運用深度學習與影像辨識，以及改良壓縮影像與全檢測技術，提升路徑規劃的分析效率。目前研究成果發現深度學習的正確率為0.7，而改良影像壓縮全檢技術在誤差3^°內的正確率為0.86，後者的計算時間在所測試範例樣本中，平均至少可以縮短10倍以上的檢測時間。同時，本研究所開發的雷射掃描路徑規劃技術，亦可應用於雷射加法製造的製程優化，改進現行一般雷射積層製造路徑優化未考慮熱累積效應的缺點。
此外，本研究也開發可處理電路板短路缺陷的雷射掃描路徑規劃技術，依使用者需求分為兩種功能：
第一類是對短路區域進行切斷，目的是快速破壞短路；第二類是對短路區域進行精準的完全清除，目標應用場域是高頻、高靈敏度電子元件的電路，解決原本製程殘留的不規則金屬或殘膠所造成的電磁干擾雜訊。

The current study combines automatic optical inspection and detection with image analysis and optimization technologies to develop an intelligent laser path planning method for ultra-precision printed circuit board (PCB) defects repairment. Once the user placed the defective PCB on the working stage and captured the image of the board, a series of image processing, positioning will be conducted to attain information of defects. The repairing path based on this information of defects will then be determined automatically and accurately by using the rule-based method developed in this study.
To further improve the efficiency and effectiveness for the path planning analysis, Deep Learning and a new method based on image compression was applied, separately, with considering heat accumulation during the laser process to optimize laser scanning strategy. The accuracy to the optimized scanning direction achieved by Deep Learning and by the new image compression method is 0.7 and 0.86, respectively. The later method can reduce the analyzing time at least 10 times than the exhaustive method (full-scan).
The method developed in this study can not only be applied for repairing open circuits but also short circuits of PCB. For the short circuits, in addition to line cutting, a full clean of the defect is possible thanks to the image process and analysis of the circuits technology that was developed in conjunction with the current study. Thus, the method presented here is especially valuable for repairing high precision circuit boards and electronics.

摘要 I
Abstract II
誌謝 III
目錄 IV
圖目錄 VI
表目錄 IX
第一章 緒論 1
1.1 研究動機與目的 1
第二章 文獻回顧 3
2.1 雷射掃描技術 3
2.1.1 印刷電路板 3
2.1.2 雷射修復電路基板技術 4
2.1.3 雷射掃描路徑策略規劃 5
2.2 影像處理 10
2.2.1 卷積運算法(Convolution) 10
2.2.2 形態學(Morphology) 12
2.2.3 像素連通性(Pixel Connectivity) 13
2.2.4 濾波函數 14
第三章 研究方法 16
3.1 架構設計 16
3.2 程式設計 19
第四章 研究結果 35
4.1 影像壓縮結果 36
4.2 最佳掃描路徑結果圖 39
4.3 斷路與短路之路徑規劃結果 40
第五章 結論與未來展望 44
5.1 結論 44
5.2 未來展望 45
Reference 46

[1] M. Moganti, F. Ercal, C. H. Dagli, and S. Tsunekawa, "Automatic PCB inspection algorithms: a survey," Computer vision and image understanding, vol. 63, no. 2, pp. 287-313, 1996.
[2] S. Luo, P. T. Hoang, and T. Liu, "Direct laser writing for creating porous graphitic structures and their use for flexible and highly sensitive sensor and sensor arrays," Carbon, vol. 96, pp. 522-531, 2016.
[3] B. Lange, "PCB machining and repair via laser," On Board Technol, vol. 2, pp. 12-14, 2005.
[4] R. Oron, "Performance optimization of electronic circuits laser repair," in Laser Applications in Microelectronic and Optoelectronic Manufacturing (LAMOM) XIX, vol. 8967： International Society for Optics and Photonics, p. 89671H, 2014.
[5] C.-H. Tsai and C.-J. Chen, "Application of iterative path revision technique for laser cutting with controlled fracture," Optics and lasers in Engineering, vol. 41, no. 1, pp. 189-204, 2004.
[6] G. C. Sih, "Strain-energy-density factor applied to mixed mode crack problems," International Journal of fracture, vol. 10, no. 3, pp. 305-321, 1974.
[7] T. Chwan-Huei and M. Chien-Ching, "Thermal weight function of cracked bodies subjected to thermal loading," Engineering Fracture Mechanics, vol. 41, no. 1, pp. 27-40, 1992.
[8] Arimoto S, Kawamura S, Miyazaki F. Bettering operation of robots by learning. J Robotic System, 1984.
[9] Y.-M. Huang and H.-Y. Lan, "Path planning effect for the accuracy of rapid prototyping system," The International Journal of Advanced Manufacturing Technology, vol. 30, no. 3-4, pp. 233-246, 2006.
[10] Y.-K. Liu and M.-T. Lee, "Laser direct synthesis and patterning of silver nano/microstructures on a polymer substrate," ACS applied materials & interfaces, vol. 6, no. 16, pp. 14576-14582, 2014.
[11] K. Ren, Y. Chew, J. Fuh, Y. Zhang, and G. Bi, "Thermo-mechanical analyses for optimized path planning in laser aided additive manufacturing processes," Materials & Design, vol. 162, pp. 80-93, 2019.
[12] T. Manoj and A. S. Rubia, "A Survey and Evaluation of Edge Detection Operators： Application to Text Recognition," 2012.
[13] S. W. Smith, "The Scientist and Engineer's Guide to Digital Signal Processing," 1999.
[14] I. Sobel, "Neighborhood coding of binary images for fast contour following and general binary array processing," Computer graphics and image processing, vol. 8, no. 1, pp. 127-135, 1978.
[15] S. Villumsen, "Process time optimization of robotic remote laser cutting by utilizing customized beam patterns and redundancy space task sequencing," 2016.
[16] C. Negara, "Fast polarization state detection by division-of-amplitude in a simple configuration setup," in Proceedings of the 2015 Joint Workshop of Fraunhofer IOSB and Institute for Anthropomatics, Vision and Fusion Laboratory, vol. 24： J. Beyerer and A. Pak (KIT Scientific Publishing, Karlsruhe, 2016), pp. 75-89, 2016.
[17] P. Han et al., "An Improved Adaptive Correction Algorithm For Non-uniform Illumination Panoramic Image," in 2019 IEEE 2nd International Conference on Electronic Information and Communication Technology (ICEICT)： IEEE, pp. 258-262, 2019.
[18] X. Yang, X. Shen, J. Long, and H. Chen, "An improved median-based Otsu image thresholding algorithm," Aasri Procedia, vol. 3, pp. 468-473, 2012.
[19] K. He, X. Zhang, S. Ren, and J. Sun, "Deep residual learning for image recognition," in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 770-778, 2016.
[20] C. Szegedy et al., "Going deeper with convolutions," in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 1-9, 2015.
[21] R. Keys, "Cubic convolution interpolation for digital image processing," IEEE transactions on acoustics, speech, and signal processing, vol. 29, no. 6, pp. 1153-1160, 1981.
[22] V. A. Adibhatla, H.-C. Chih, C.-C. Hsu, J. Cheng, M. F. Abbod, and J.-S. Shieh, "Defect detection in printed circuit boards using you-only-look-once convolutional neural networks," Electronics, vol. 9, no. 9, p. 1547, 2020.