不同的種類的深度卷積神經網路 (DCNN) 用於晶圓圖識別、分類問題，已經在過去的研究中被提出。然而輸入的圖像解析度用於提出的模型的分類表現影響及訓練集的類別分佈不平均的問題，從過去到現在並未被考慮在過去的研究中。本研究提出一個基於 DCNN 架構下的一種模型並結合殘差學習單元，稱為最佳化殘差學習深度卷積神經網路 (Opt-ResDCNN) 模型，應用於晶圓圖缺陷辨識、分類並考慮不同的輸入圖像解析度和在訓練時的訓練集不平衡問題。本研究提出的模型藉由平衡公式，平衡訓練集提升模型效果，並且與過去發表的模型及不同解析度大小做缺陷辨識和分類的準確度 (Accuracy)、精確度 (Precision)、召回率 (Recall)、F1 score比較。利用公開的晶圓資料集 (WM-118K dataset)，所提出的模型可以得到平均的分類準確度為 99.896%、98.351%、90.277%及98.879%，在分別輸入圖像得解析度為26*26、64*64、96*96及256*256下，儘管在十次的嚴謹實驗證明下測試，本研究所提出的模型在不同的輸入圖像解析度下的表現指標均優於過去所發表的結果。

Different deep convolution neural network (DCNN) models have been proposed for wafer map pattern identification and classification tasks in previous studies. However, factors such as the effect of input image resolution on the classification performance of the proposed models and the class imbalance issue in the training set have not been considered in the previous studies. This thesis proposes a DCNN-based model with residual blocks called Opt-ResDCNN model for wafer map defect pattern identification and classification by considering different input image resolutions and class imbalance issues during the model training. In this thesis the proposed model used balanced training set by balance function to improve the performance and compared with the previously published defect pattern recognition and classification models in terms of accuracy, precision, recall, and F1 score for different input image sizes. Using a publicly available wafer map dataset (WM-811K), the proposed method can obtain an average classification accuracy result of 99.896%, 98.351%, 90.277%, and 98.879%, for 26*26, 64*64, 96*96, and 256*256, input image resolutions respectively. The proposed model outperforms previously published results in all performance metrics for different input image resolution even used ten-iteration rigorous experiments verification.

References
[1] TSMC (2020, Dec). TSMC Recognized with 2021 IEEE Corporate Innovation Award. Accessed: Apr. 06, 2021. [Online]. Available: https://pr.tsmc.com/english/news/2765.
[2] R. Wang and N. Chen, “Defect pattern recognition on wafers using convolutional neural networks,” Qual. Reliab. Eng. Int., vol. 36, no. 4, pp. 1245-1247, Jun. 2020. doi: 10.1002/qre.2627.
[3] T. Yuan, W. Kuo, and S. J. Bae, “Detection of spatial defect patterns generated in semiconductor fabrication processes,” IEEE Trans. Semicond. Manuf., vol. 24, no. 3, pp. 392-403, Aug. 2011. doi: 10.1109/TSM.2011.2154870.
[4] F. L. Chen and S. F. Liu, “A neural-network approach to recognize defect spatial pattern in semiconductor fabrication,” IEEE Trans. Semicond. Manuf., vol. 13, no. 3, pp. 366-373, Aug. 2000. doi: 10.1109/66.857947.
[5] S. S. Gleason, K. W. Tobin, T. P. Karnowski, and F. Lakhani, “Rapid yield learning through optical defect and electrical test analysis,” Proc. SPIE-Int. Soc. Opt. Eng., vol. 3332, pp. 232-242, 1998, doi: 10.1117/12.308731.
[6] T. Ishida, I. Nitta, D. Fukuda, and Y. Kanazawa, “Deep learning-based wafer-map failure pattern recognition framework,” in Proc. of IEEE Int. Symp. Quality Electron. Des., Santa Clara, USA, Mar. 2019, pp. 291-297. doi: 10.1109/ISQED.2019.8697407.
[7] M. Wu, J. R. Jang, and J. Chen, “Wafer map failure pattern recognition and similarity ranking for large-scale data sets,” IEEE Trans. Semicond. Manuf., vol. 28, no. 1, pp. 1-12, Feb. 2015. doi: 10.1109/TSM.2014.2364237.
[8] J.Gu et al., “Recent advances in convolutional neural networks,” Pattern Recognit., vol. 77, pp. 354-377, May 2018. doi: 10.1016/j.patcog.2017.10.013.
[9] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in IEEE Conf. Comput. Vis Pattern Recognit.(CVPR)., Las Vegas, USA, Dec. 2016, pp. 770-77. doi: 10.1109/CVPR.2016.90.
[10] G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, “Densely connected convolutional networks,” in IEEE Conf. Comput. Vis Pattern Recognit.(CVPR)., Honolulu, USA, Nov. 2017, pp. 2261-2269. doi: 10.1109/CVPR.2017.243.
[11] Y. Lecun, L. Bottou, Y. Bengio, and P. Haffner, “Gradient-based learning applied to document recognition,” Proc. IEEE, vol. 86, no. 11, pp. 2278-324, Nov. 1998. doi: 10.1109/5.726791.
[12] J. Yu and X. Lu, “Wafer map defect detection and recognition using joint local and nonlocal linear discriminant analysis,” IEEE Trans. Semicond. Manuf., vol. 29, no. 1, pp. 33-43, Feb. 2016. doi: 10.1109/TSM.2015.2497264.
[13] Mengying Fan, Qin Wang, and B.van der Waal, “Wafer defect patterns recognition based on OPTICS and multi-label classification,” in Proc. IEEE Adv. Inf. Manag. Commun. Electron. Autom. Control Conf., Xi’an, China, Oct. 2016, pp. 912-915. doi: 10.1109/IMCEC.2016.7867343.
[14] M. Piao, C. H. Jin, J. Y. Lee, and J. Y. Byun, “Decision tree ensemble-based wafer map failure pattern recognition based on radon transform-based features,” IEEE Trans. Semicond. Manuf., vol. 31, no. 2, pp. 259-257, May 2018. doi: 10.1109/TSM.2018.2806931.
[15] M. Saqlain, B. Jargalsaikhan, and J. Y. Lee, “A voting ensemble classifier for wafer map defect patterns identification in semiconductor manufacturing,” IEEE Trans. Semicond. Manuf., vol. 32, no. 2, pp. 171-182, May 2019. doi: 10.1109/TSM.2019.2904306.
[16] A. Shawon, M. O. Faruk, M. BinHabib, and A. M. Khan, “Silicon wafer map defect classification using deep convolutional neural network with data augmentation,” in Proc. IEEE Int. Conf. Comput. Commun., Chengdu, China, Dec. 2019, pp. 1995-1999. doi: 10.1109/ICCC47050.2019.9064029.
[17] J. Yu, X. Zheng, and J. Liu, “Stacked convolutional sparse denoising auto-encoder for identification of defect patterns in semiconductor wafer map,” Comput. Ind., vol. 109, pp.121-133, Aug. 2019. doi: 10.1016/j.compind.2019.04.015.
[18] J. Wang, Z. Yang, J. Zhang, Q. Zhang and W. K. Chien, “AdaBalGAN: an improved generative adversarial network with imbalanced learning for wafer defective pattern recognition,” IEEE Trans. Semicond. Manuf., vol. 32, no. 3, pp. 310-319, Agu. 2019. doi: 10.1109/TSM.2019.2925361.
[19] J. Yu and J. Liu, “Two-dimensional principal component analysis-based convolutional autoencoder for wafer map defect detection,” IEEE Trans. Ind. Electron., Aug 2020 (Early Access). doi: 10.1109/TIE.2020.3013492.
[20] Y. Ji and J. H. Lee, “Using GAN to improve CNN performance of wafer map defect type classification : yield enhancement,” in 31st Annu. SEMI Adv. Semicond. Manuf. Conf. ( ASMC), Aug. 2020, pp. 1-6. doi: 10.1109/ASMC49169.2020.9185193.
[21] C. H. Jin, H. J. Kim, Y. Piao, M. Li, and M. Piao, “Wafer map defect pattern classification based on convolutional neural network features and error-correcting output codes,” J. Intell. Manuf., vol. 31, no. 8, pp. 1861-1875, Dec. 2020. doi: 10.1007/s10845-020-01540-x.
[22] T. H. Tsai and Y. C. Lee, “A light-weight neural network for wafer map classification based on data augmentation,” IEEE Trans. Semicond. Manuf., vol. 33, no. 4, pp. 663-672, Nov.2020. doi: 10.1109/TSM.2020.3013004.
[23] K.He, X. Zhang, S. Ren, and J. Sun, “Identity mappings in deep residual networks,” In: Leibe B., Matas J., Sebe N., Welling M. (eds) Computer Vision – ECCV 2016, Lecture Notes in Computer Science, vol 9908, pp. 630-645. Springer, Cham. Sep. 2016. https://doi.org/10.1007/978-3-319-46493-0_38.
[24] K. Maksim et al., “Classification of wafer maps defect based on deep learning methods with small amount of data,” in Proc. Int. Conf. Eng. Telecom., Dolgoprudny, Russia, 2019, pp. 1-5, Nov. 2019. doi: 10.1109/EnT47717.2019.9030550.
[25] K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” Apr. 2015. [Online]. Available: https://arXiv:1409.1556, 2014 - arxiv.org.
[26] A. G. Howard et al., “MobileNets: efficient convolutional neural networks for mobile vision applications,” Apr. 2017. [Online]. Available: https://arxiv.org/abs/1704.04861v1.
[27] Z. Shen and J. Yu, “Wafer map defect recognition based on deep transfer learning,” in Proc. IEEE Int. Conf. Ind. Eng. Eng. Manag. (IEEM), Macao, China, Dec. 2019, pp. 1568-1572. doi: 10.1109/IEEM44572.2019.8978568.
[28] J. M. Johnson and T. M. Khoshgoftaar, “Survey on deep learning with class imbalance,” J. Big Data, vol. 6, no. 1, Mar. 2019. doi: 10.1186/s40537-019-0192-5.
[29] H. Lee, M. Park, and J. Kim, “Plankton classification on imbalanced large scale database via convolutional neural networks with transfer learning,” in Proc. IEEE Int. Conf. Image Process., Phoenix, USA, Sep. 2016, pp. 3713-3717. doi: 10.1109/ICIP.2016.7533053.
[30] S. Pouyanfar et al., “Dynamic sampling in convolutional neural networks for imbalanced data classification,” in Proc. IEEE Conf. Multimed. Inf. Process. Retr., Miami, USA, Apr. 2018, pp. 112-117. doi: 10.1109/MIPR.2018.00027. [31] M. Buda, A. Maki, and M. A. Mazurowski, “A systematic study of the class imbalance problem in convolutional neural networks,” Neural Networks, vol. 106, pp. 249-259, Oct. 2018. doi: 10.1016/j.neunet.2018.07.011.
[32] P. Hensman and D. Masko, “The impact of imbalanced training data for convolutional neural networks,” Degree project in Computer Science, Stockholm, Sweden, May 2015.
[33] S. Wang, W. Liu, J. Wu, L. Cao, Q. Meng, and P. J. Kennedy, “Training deep neural networks on imbalanced data sets,” in Proc. Int. Joint Conf. Neural Netw., Vancouver, Canada, Jul. 2016, pp. 4368-4374. doi: 10.1109/IJCNN.2016.7727770.
[34] S. H. Khan, M. Hayat, M. Bennamoun, F. A. Sohei, and R. Togneri, “Cost-sensitive learning of deep feature representations from imbalanced data,” IEEE Trans. Neural Networks Learn. Syst., vol. 29, no. 8, pp. 3573-3587, Aug.2018. doi: 10.1109/TNNLS.2017.2732482.
[35] H. Wang, Z. Cui, Y. Chen, M. Avidan, A. BenAbdallah, and A. Kronzer, “Predicting hospital readmission via cost-sensitive deep learning,” IEEE/ACM Trans. Comput. Biol. Bioinforma., vol. 15, no. 6, pp. 1968-1978, Nov.2018. doi: 10.1109/TCBB.2018.2827029.
[36] C. Zhang, K. C. Tan, and R. Ren, “Training cost-sensitive deep belief networks on imbalance data problems,” in Proc. Int. Joint Conf. Neural Netw., Vancouver, Canada, Jul. 2016, pp. 4362-4367. doi: 10.1109/IJCNN.2016.7727769.
[37] C. Huang, Y. Li, C. C. Loy, and X. Tang, “Learning deep representation for imbalanced classification,” in Proc. IEEE Conf. Comput. Vis. Pattern Recognit., Las Vegas, USA, Jun. 2016, pp. 5375-5384. doi: 10.1109/CVPR.2016.580.
[38] T. Y. Lin, P. Goyal, R. Girshick, K. He, and P. Dollar, “Focal loss for dense object detection,” in Proc. IEEE Int. Conf Comput. Vis., Venice, Italy, Oct. 2017, pp. 2999-3007. doi: 10.1109/ICCV.2017.324.
[39] Y. Zhang, L. Shuai, Y. Ren, and H. Chen, “Image classification with category centers in class imbalance situation,” in Proc. Youth Acad. Ann. Conf. Chinese Assoc. Autom., Nanjing, China, May 2018, pp. 359-363. doi: 10.1109/YAC.2018.8406400.
[40] S. Ando and C. Y. Huang, “Deep over-sampling framework for classifying imbalanced data,” in Ceci M., Hollmén J., Todorovski L., Vens C., Džeroski S. (eds) Machine Learning and Knowledge Discovery in Databases. ECML PKDD, Lect. Notes Comput. Sci., vol 10534. Springer, Cham, 2017, doi:10.1007/978-3-319-71249-9_46.
[41] Q. Dong, S. Gong, and X. Zhu, “Imbalanced deep learning by minority class incremental rectification,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 41, no. 6, pp. 1367-1381, Jun. 2019. doi: 10.1109/TPAMI.2018.2832629.
[42] H. Lee, J. Kim, B. Kim, and S .Kim, “Convolutional autoencoder based feature extraction in radar data analysis,” Conf. Soft Comput. Intell. Syst. 19th Int. Symp. Adv. Intell. Syst., Toyama, Japan, May 2019, pp. 81-84, doi: 10.1109/SCIS-ISIS.2018.00023.
[43] B.Hou and R.Yan, “Convolutional auto-encoder based deep feature learning for finger-vein verification,” IEEE Int. Symp. Med. Meas. Appl., Rome, Italy, Aug. 2018, pp. 1-5. doi: 10.1109/MeMeA.2018.8438719.