Defects Detection of TFT Lines of Flat Panel Displays Using an Evolutionary Optimized Recurrent Neural Network

Hapu Arachchilage Abeysundara, Hiroshi Hamori, Takeshi Matsui, Masatoshi Sakawa

Department of Electrical Systems and Mathematical Engineering, Faculty of Engineering, Hiroshima University, Hiroshima, Japan.
Research & Development Department, OHT Incorporation, Fukuyama, Japan.

DOI: 10.4236/ajor.2014.43011                

Abstract

This paper proposes an evolutionary optimized recurrent neural network for inspection of open/short defects on thin film transistor (TFT) lines of flat panel displays (FPD). The inspection is performed on digitized waveform data of voltage signals that are captured by a capacitor based non-contact sensor through scanning over TFT lines on the surface of mother glass of FPD. Irregular patterns on the waveform, sudden deep falls (open circuits) or sharp rises (short circuits), are classified and detected by employing the optimized recurrent neural network. The topology parameters of the recurrent neural network are optimized by a multiobjective evolutionary optimization process using a selected training data set. This method is an extension to our previous work, which utilized a feed-forward neural network, to address the drawbacks in it. Experimental results show that this method can detect defects on more realistic and noisy data than both of the previous method and the conventional threshold based method.

Keywords

Non-Contact Defects Inspection, Recurrent Neural Networks, Evolutionary Optimization, Open Short Detection

Conflicts of Interest

The authors declare no conflicts of interest.

References

[1]	Liu, Y.H., Lin, S.H., Hsueh, Y.L. and Lee, M.J. (2009) Automatic Target Defects Identification for TFT-LCD Array Process Using FCM Based Fuzzy SVDD Ensemble. International Journal of Expert Systems with Applications, 36, 1978-1998. http://dx.doi.org/10.1016/j.eswa.2007.12.015
[2]	Liu, Y.H. and Chen, Y.J. (2011) Automatic Defect Detection for TFT-LED Array Process Using Quasi-Conformal Kernel Support Vector Data Description. International Journal Neural of Molecular Science, 12, 5762-5781.
[3]	Lu, C.-J. and TSai, D.-M. (2005) Automatic Defects Inspections for LCD Using Singular Value Decomposition. International Journal of Advanced Manufacturing Technology, 25, 53-61. http://dx.doi.org/10.1007/s00170-003-1832-6
[4]	Lu, C.-J. and TSai, D.-M. (2004) Defects Inspections of Patterned TFT-LCD Panels Using a Fast Sub-Image Base SVD. Proceedings of 5th Asia Pacific Industrial Engineering and Management Systems Conference, Gold Coast, 1215 December 2004, 3.3.1-3.3.16.
[5]	Hamori, H., Sakawa, M., Katagiri, M. and Matsui, T. (2010) A Fast Non-Contact Inspection System Based on a Dual Channel Measurement System. Journal of Japan Institute of Electronic Packaging, 13, 562-568. (in Japanese)
[6]	Hamori, H., Sakawa, M., Katagiri, M. and Matsui, T. (2010) Fast Non-Contact Flat Panel Inspection through a Dual Channel Measurement System. Proceedings of International Conference on Computers and Industrial Engineers, Awaji, 25-28 July 2010, 1-6. http://dx.doi.org/10.1109/ICCIE.2010.5668229
[7]	Hamori, H., Sakawa, M., Katagiri, M. and Matsui, T. (2011) A Defect Position Identification System Based on a Dual Channel Measurement System. Journal of Japan Institute of Electronics, Information and Communication Engineers, J94-C, 323-333. (in Japanese)
[8]	Hamori, H., Sakawa, M., Katagiri, M. and Matsui, T. (2011) A Dual Channel Defect Position Identification Method for Touch Panel Manufacturing Process. Proceedings of International Conference on Electronics Packaging, Shanghai, 8-11 August 2011, 732-736.
[9]	Abeysundara, H.A., Hamori, H., Matsui, T. and Sakawa, M. (2013) Defects Detection on TFT lines of Flat Panels Using a Feed Forward Neural Network. International Journal of Artificial Intelligence Research, 2, 1-12.
[10]	Abeysundara, H.A., Hamori, H., Matsui, T. and Sakawa, M. (2013) A Neural Network Approach for Non-Contact Defects Inspection of Flat Panel Displays. 17th International Conference in Knowledge Based and Intelligent Information and Engineering Systems, Kita Kyushu, 9-11 September 2013, 28-38.
[11]	Delgado, M., Cuellar, M.P. and Pegalajar, M.C. (2008) Multiobjective Hybrid Optimization and Training of Recurrent Neuural Networks. IEEE Transactions on Systems, Man and Cybernetics: Part B Cybernetics, 38, 381-403. http://dx.doi.org/10.1109/TSMCB.2007.912937
[12]	Katagiri, H., Nishizaki, I., Hayashida, T. and Kadoma, T. (2011) Multiobjective Evolutionary Optimization of Training and Topology of Recurrent Neural Networks for Time Series Prediction. The Computer Journal, 55, 325-336. http://dx.doi.org/10.1093/comjnl/bxr042
[13]	Rumelhart, D.E. and Mcclelland, D.E. (1986) Parallel Distributed Processing: Explorations in the Microstructure of Cognition, Volume 1, Foundations. The MIT Press, Cambridge.
[14]	Rojas, P. (1996) Neural Networks—A Systematic Introduction. Springer-Verlag, Berlin, Heidelberg.
[15]	Deb, K., Pratap, A., Agarwal, S. and Meyarivan, T. (2002) A Fast and Elitist Multiobjective Genetic Algorithm: NSGA-II. IEEE Transactions on Evolutionary Computation, 6, 182-197. http://dx.doi.org/10.1109/4235.996017
[16]	Huang, B.Q., Rashid, T. and Kechadi, M.T. (2007) Multi-Context Recurrent Neural Networks for Time Series Applications. World Academy of Science, Engineering and Technology, 10, 4448457.
[17]	Ang, J.H., Goh, C.K., Teoh, E.J. and Mamum, A.A. (2007) Multi-Objective Evolutionary Recurrent Neural Network for System Identification. Proceedings of IEEE Congress on Evolutionary Computations, Singapore, 25-28 September 2007, 1586-1592.
[18]	Husken, M. and Stagge, P. (2003) Recurrent Neural Networks for Time series Classification. Neurocomputing, 50, 223-235.
[19]	Dolinsky, J. and Takagi, H. (2008) RNN with Recurrent Output Layer for Learning of Naturalness. Neural Information Processing: Letters and Reviews, 12, 31-42.
[20]	Zang, G.P. (2000) Neural Networks for Classification: A Survey. IEEE Transactions on System, Man and Cybernetics, Part C: Applications and Reviews, 30, 451-462. http://dx.doi.org/10.1109/5326.897072