在TFT-LCD 的微影製程中，由於製程本身的精密要求，良率的控制便扮演著舉足輕重的角色。在微影製程結束欲進入下個製程以前，往往利用自動光學檢查 (AOI) 將面板上的缺陷位置記錄下來而形成缺陷影像。這些缺陷影像再由有經驗的工程師或作業員以人工辨識的方式歸類其發生的原因。顯然的，這種以人工辨識的方式可能會因為個人的差異，導致淺在的誤判或時間上的浪費。本篇論文提出一應用類神經網路於TFT-LCD微影製程缺陷辨識之方法，並利用四種神經網路：倒傳遞神經網路，放射基準機能網路，學習向量量化網路1 (LVQ1)，學習向量量化網路2 (LVQ2)，來探討最後的執行結果與進行比較。研究的結果發現本論文所提出的方法可有效的進行微影製程缺陷之辨識。

Since the advent of high qualification and tiny technology, yield control in the photolithography process has played an important role in the manufacture of Thin Film Transistor-Liquid Crystal Displays (TFT-LCDs). Through an Auto Optic Inspection (AOI), defect points from the panels are collected and the defect images are generated after the photolithography process. The defect images are usually identified by experienced engineers or operators. Evidently, human identification may produce potential misjudgments and cause time loss. This study therefore proposes a neural-network approach for defect recognition in the TFT-LCD photolithography process. There were four neural-network methods adopted for this purpose, namely, backpropagation, radial basis function, learning vector quantization 1, and learning vector quantization 2. A comparison of the performance of these four types of neural-networks was illustrated. The results show that the proposed approach can effectively recognize the defect images in the photolithography process.

References
1. C.Weber, B. Moslehi, and M. Dutta, “An integrated framework For yield management and defect/fault reduction,” IEEE Transactions Semiconductor Manufacturing, vol. 8, pp. 110-120, May 1995.
2. S. P. Cunningham and S. MacKinnon, “Statistical methods for visual defect metrology,” IEEE Transactions Semiconductor Manufacturing, vol. 11, pp. 48-53, Feb. 1998.
3. L. Breaux and B.Singh, “Automatic defect classification system for patterned semiconductor wafers,” Semiconductor Manufacturing, 1995, pp.68-73.
4. J.-M. Zhang, R.-M. Lin, and M.-J. J. Wang, “The development of an automatic post-sawing inspection system using computer vision techniques,” Computers in Industry, vol. 40, pp. 51–60, 1999.
5. F.-L. Chen and S.-F. Liu, “A neural-network approach to recognize defect spatial pattern in semiconductor fabrication,” IEEE Transactions Semiconductor Manufacturing, vol. 13, pp. 366–373, Aug. 2000.
6. K. K. Sreenivasan, M. Srinath, and A. Khotanzad, “Automated vision system for inspection of IC pads and bonds,” IEEE Transactions on Components, Hybrids, Manufacturing Technology, vol. 16, pp. 333–338, May 1993.
7. D. Barschdorff and L. Monostori, “Neural networks: Their applications and perspectives in intelligent machining,” Computers in Industry, vol. 17, pp. 101–119, 1991.
8. Su, C.-T., Yang T., and Ke C.-M., “A Neural-Network Approach for Semiconductor Wafer Post-Sawing Inspection,” IEEE Transactions Semiconductor Manufacturing, vol. 15,no. 2, pp. 260-266, May 2002.
9. Nicolaos B. Karayiannis, James C. Bezdek, Nikhil R. Pal, Richard J. Hathaway, and Pin-I Pai, “Repairs to GLVQ: A New Family of Competitive Learning Schemes,” IEEE Transactions Neural Networks., vol. 7,no. 5, pp. 1062-1071, Sep. 1996.
10. F. Duvivier, “Automatic detection of spatial signature on wafermaps in a high volume production,” Defect Fault Tolerance VLSI Syst., 1999, pp. 68-73.
11. R. P. Lippmann, “An introduction to computing with neural nets,” IEEE Acoustics, Speech and Signal Pcocessing Mag., vol. 4, pp. 4-22, 1987.

12. Fausett, L., Fundamentals of Neural Networks, Prentice-Hall International, 1994.
13. J.-S. R. Jang, C.-T. Sun, and E. Mizutani, Neuro-Fuzzy and Soft Computing. Upper Saddle River, NJ: Prentice Hall, 1997.
14. V. K. Gupta, J. G. Chen, and M. B. Murtaza, “A learning vector quantization neural network model for the classification of industrial construction projects,” Omega, vol. 25, no. 6, pp. 715-727, 1997.