自動化檢測是數位轉型的重要過程，雖然深度學習模型已經廣泛應用於自動化檢測當中，但深度學習模型不能客觀地解釋結果，稱為低可解釋性。當深度學習模型是低可解釋性的狀態，發生判斷錯誤時人類難以找出錯誤的根本原因，更無法加以修正來提高模型的準確率，只能透過漫無目的訓練模型來祈禱下一次深度學習模型不會犯錯。本研究提出一種整合方法，將物件偵測的實例分割深度學習模型、分群演算法、相似度演算法與計數演算法整合，來實現可解釋的自動化檢測過程。本研究透過一個球鞋鞋底的實際研究案例表明，可解釋的檢測方法可以快速調整錯誤並提高準確率，顏色檢測方面最終測試的績效為F1 score 96.89%，噴墨點檢測方面為MAPE 0.9081%。本研究解決了數位轉型中自動化檢測的挑戰，第一，提高深度學習模型的可解釋性；第二，加強深度學習模型的可修正性。實際應用方面，可以結合遮罩面積計算物體尺寸、利用圖像相似度找出對稱配雙的產品，以及使用遮罩形狀判斷數字、字母。

Automated detection is an important process in digital transformation. Although deep learning models have been widely used in automated detection, deep learning models cannot objectively explain the results, which is called low interpretability. When the deep learning model is in a state of low interpretability, it is difficult for humans to find the root cause of the error when a judgment error occurs, and it is impossible to correct it to improve the accuracy of the model. Only by training the model aimlessly to pray for the next deep learning Models don't make mistakes. This study proposes an integration method that integrates instance segmentation deep learning models, clustering algorithms, similarity algorithms, and counting algorithms for object detection to achieve an interpretable automated detection process. This study shows through a real-world case study of sneaker soles, that interpretable detection methods can quickly adjust for errors and improve accuracy, with final test performances of 96.89% for color detection and 0.9081% for MAPE for inkjet dot detection. This study addresses the challenges of automated detection in the digital transformation, first, improving the interpretability of deep learning models; second, enhancing the correctability of deep learning models. In practical applications, it is possible to calculate the size of objects along with the area of the mask, use the image similarity to find products with symmetrical pairs and use the shape of the mask to judge numbers and letters. For example: matching pairs of sneakers of the same size, quality grading with fruit spots, etc.

摘要…………………………………………………………………………...……….2
目錄…………………………………………………………….……….……………..4
第一章 介紹…………………………………………………......................................6
第二章 文獻回顧……………………………………………......................................8
2-1. 數位轉型………………………………........................................8
2-2. 模型可解釋性………………………………................................8
2-3. 小結………………………………..............................................11
第三章 研究方法………………………………........................................................11
3-1. 資料前處理………………………………..................................13
3-2. 模型訓練………………………………………………………..14
3-2-1. 卷積運算網路………………………………..14
3-2-2. 區域提案網路和RoIAlign…...……………...14
3-2-3. 輸出層………………………………………..15
3-3. 分群演算法……………………………………………………..16
3-4. 分析及整合……………………………………………………..18
3-4-1. 相似度演算法………………………………..18
3-4-2. 計數演算法…………………………………..19
3-5. 衡量指標………………………………………………………..20
第四章 案例研究……………………………………………………………………21
4-1. 資料前處理……………………………………………………..22
4-2. 模型訓練………………………………………………………..23
4-3. 分群演算法……………………………………………………..24
4-4. 分析及整合……………………………………………………..25
4-4-1. CIEDE2000……………………………………25
4-4-2. 計數演算法…………………………………..26
4-5. 績效評估......................................................................................27
4-6. 討論……………………………………………………………..29
第五章 結論…………………………………………………………………………31
參考文獻……………………………………………………………………………..33
附錄…………………………………………………………………………………..41

[1] Bacchi, S., Oakden-Rayner, L., Zerner, T., Kleinig, T., Patel, S., & Jannes, J. (2019).
Deep learning natural language processing successfully predicts the cerebrovascular
cause of transient ischemic attack-like presentations. Stroke, 50(3), 758-760.
[2] Buhrmester, V., Münch, D., & Arens, M. (2021). Analysis of explainers of black
box deep neural networks for computer vision: A survey. Machine Learning and
Knowledge Extraction, 3(4), 966-989.
[3] Carvalho, D. V., Pereira, E. M., & Cardoso, J. S. (2019). Machine learning
interpretability: A survey on methods and metrics. Electronics, 8(8), 832.
[4] Chakraborty, S., Tomsett, R., Raghavendra, R., Harborne, D., Alzantot, M., Cerutti,
F., ... & Gurram, P. (2017, August). Interpretability of deep learning models: A survey
of results. In 2017 IEEE smartworld, ubiquitous intelligence & computing, advanced
& trusted computed, scalable computing & communications, cloud & big data
computing, Internet of people and smart city innovation. IEEE.
[5] Chang, F., Liu, M., Dong, M., & Duan, Y. (2019). A mobile vision inspection
system for tiny defect detection on smooth car-body surfaces based on deep ensemble
learning. Measurement Science and Technology, 30(12), 125905.
[6] Chen, J., Liu, Z., Wang, H., Núñez, A., & Han, Z. (2017). Automatic defect
detection of fasteners on the catenary support device using deep convolutional neural
network. IEEE Transactions on Instrumentation and Measurement, 67(2), 257-269.
[7] Chiu, M. C., Lee, Y. H., & Chen, T. M. (2022). Integrating content-based image
retrieval and deep learning to improve wafer bin map defect patterns classification.
Journal of Industrial and Production Engineering, 1-15.
33
[8] Chiu, M. C., Wen, C. Y., Hsu, H. W., & Wang, W. C. (2022). Key wastes selection
and prediction improvement for biogas production through hybrid machine learning
methods. Sustainable Energy Technologies and Assessments, 52, 102223.
[9] Chiu, M. C., Tsai, H. Y., & Chiu, J. E. (2022). A novel directional object detection
method for piled objects using a hybrid region-based convolutional neural network.
Advanced Engineering Informatics, 51, 101448.
[10] Chiu, M. C., & Chen, T. M. (2021). Applying Data Augmentation and Mask RCNN-Based Instance Segmentation Method for Mixed-Type Wafer Maps Defect
Patterns Classification. IEEE Transactions on Semiconductor Manufacturing, 34(4),
455-463.
[11] Chiu, M. C., Tsai, C. D., & Li, T. L. (2020). An integrative machine learning
method to improve fault detection and productivity performance in a cyber-physical
system. Journal of Computing and Information Science in Engineering, 20(2), 021009.
[12] Colyer, S. L., Evans, M., Cosker, D. P., & Salo, A. I. (2018). A review of the
evolution of vision-based motion analysis and the integration of advanced computer
vision methods towards developing a markerless system. Sports medicine-open, 4(1),
24.
[13] Du, M., Liu, N., & Hu, X. (2019). Techniques for interpretable machine learning.
Communications of the ACM, 63(1), 68-77.
[14] Gilpin, L. H., Bau, D., Yuan, B. Z., Bajwa, A., Specter, M., & Kagal, L. (2018,
October). Explaining explanations: An overview of interpretability of machine learning.
In 2018 IEEE 5th International Conference on data science and advanced analytics
(DSAA) (pp. 80-89). IEEE.
34
[15] Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT press.
Honegger, M. (2018). Shedding light on black box machine learning algorithms:
Development of an axiomatic framework to assess the quality of methods that explain
individual predictions. arXiv preprint arXiv:1808.05054.
[16] Jiao, L., Zhang, F., Liu, F., Yang, S., Li, L., Feng, Z., & Qu, R. (2019). A Survey
of Deep Learning-Based Object Detection. IEEE Access, 7, 128837-128868.
[17] Kang, K., Li, H., Yan, J., Zeng, X., Yang, B., Xiao, T., Zhang, C., Wang, Z., Wang,
R., Wang X., et al., “T-cnn: Tubelets with convolutional neural networks for object
detection from videos,” IEEE Transactions on Circuits and Systems for Video
Technology, vol. 28, no. 10, pp. 2896–2907, 2018.
[18] Kang, H. S., Lee, J. Y., Choi, S., Kim, H., Park, J. H., Son, J. Y., ... & Do Noh, S.
(2016). Smart manufacturing: Past research, present findings, and future directions.
International journal of precision engineering and manufacturing-green technology,
3(1), 111-128.
[19] Kleinberg, J., & Mullainathan, S. (2019, June). Simplicity creates inequity:
implications for fairness, stereotypes, and interpretability. In Proceedings of the 2019
ACM Conference on Economics and Computation (pp. 807-808).
[20] Ku, C. C., Chien, C. F., & Ma, K. T. (2020). Digital transformation to empower
smart production for Industry 3.5 and an empirical study for textile dyeing. Computers
& Industrial Engineering, 142, 106297.
[21] Kumar, S. S., Abraham, D. M., Jahanshahi, M. R., Iseley, T., & Starr, J. (2018).
Automated defect classification in sewer closed circuit television inspections using
deep convolutional neural networks. Automation in Construction, 91, 273-283.
35
[22] Kusiak, A. (2018). Smart manufacturing. International Journal of Production
Research, 56(1-2), 508-517.
[23] Krizhevsky, A., & Sutskever, I. (2012). H. Geoffrey E.,“Alex Net,”. Adv. Neural
Inf. Process. Syst, 25, 1-9.
[24] Laato, S., Tiainen, M., Islam, A. N., & Mäntymäki, M. (2022). How to explain AI
systems to end users: a systematic literature review and research agenda. Internet
Research, 32(7), 1-31.
[25] Lin, T. Y., Dollár, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017a).
Feature pyramid networks for object detection. In Proceedings of the IEEE conference
on computer vision and pattern recognition (pp. 2117-2125).
[26] Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C. Y., & Berg, A. C.
(2016, October). Ssd: Single shot multibox detector. In European conference on
computer vision (pp. 21-37). Springer, Cham.
[27] Lin, T. Y., Goyal, P., Girshick, R., He, K., & Dollár, P. (2017b). Focal loss for
dense object detection. In Proceedings of the IEEE international conference on
computer vision (pp. 2980-2988).
[28] Lv, C., Zhang, Z., Shen, F., Zhang, F., & Su, H. (2020). A fast surface defect
detection method based on background reconstruction. International Journal of
Precision Engineering and Manufacturing, 21(3), 363-375.
[29] Lv, Q., & Song, Y. (2019). Few-shot learning combine attention mechanism-based
defect detection in bar surface. ISIJ International, 59(6), 1089-1097.
36
[30] Ma, J., Shao, W., Ye, H., Wang, L., Wang, H., Zheng, Y., & Xue, X. (2018).
Arbitrary-oriented scene text detection via rotation proposals. IEEE Transactions on
Multimedia, 20(11), 3111-3122.
[31] Murdoch, W. J., Singh, C., Kumbier, K., Abbasi-Asl, R., & Yu, B. (2019).
Definitions, methods, and applications in interpretable machine learning. Proceedings
of the National Academy of Sciences, 116(44), 22071-22080.
[32] Nakagawa, K., Ito, T., Abe, M., & Izumi, K. (2019). Deep recurrent factor model:
interpretable non-linear and time-varying multi-factor model. arXiv preprint
arXiv:1901.11493.
[33] Nambisan, S., Wright, M., & Feldman, M. (2019). The digital transformation of
innovation and entrepreneurship: Progress, challenges and key themes. Research Policy,
48(8), 103773.
[34] Napoletano, P., Piccoli, F., & Schettini, R. (2018). Anomaly detection in
nanofibrous materials by CNN-based self-similarity. Sensors, 18(1), 209.
[35] Park, J.; Kwon, B.; Park, J.; Kang, D. Machine learning-based imaging system for
surface defect inspection. Int. J. Proc. Eng. Manuf. Green Technol. 2016, 3, 303–310
[36] Pintelas, E., Livieris, I. E., & Pintelas, P. (2020). A grey-box ensemble model
exploiting black-box accuracy and white-box intrinsic interpretability. Algorithms,
13(1), 17.
[37] Rashmi, M., & Guddeti, R. M. R. (2020, January). Skeleton based Human Action
Recognition for Smart City Application using Deep Learning. In 2020 International
Conference on COMmunication Systems & NETworkS (COMSNETS) (pp. 756-761).
IEEE.
37
[38] Ren, S., He, K., Girshick, R., & Sun, J. (2015). Faster r-cnn: Towards real-time
object detection with region proposal networks. In Advances in neural information
processing systems (pp. 91-99).
[39] Rudin, C. (2019). Stop explaining black box machine learning models for high
stakes decisions and use interpretable models instead. Nature Machine Intelligence,
1(5), 206-215.
[40] Schmidhuber, J. (2015). “Deep Learning in Neural Networks: An Overview”.
Neural Networks., 61: 85-117. arXiv:1404.7828.
[41] Schallmo, D., Williams, C. A., & Boardman, L. (2017). Digital transformation of
business models—best practice, enablers, and roadmap. International Journal of
Innovation Management, 21(08), 1740014.
[42] Shoeibi, A., Khodatars, M., Alizadehsani, R., Ghassemi, N., Jafari, M., Moridian,
P., ... & Shi, P. (2020). Automated detection and forecasting of covid-19 using deep
learning techniques: A review. arXiv preprint arXiv:2007.10785.
[43] Stiglic, G., Kocbek, P., Fijacko, N., Zitnik, M., Verbert, K., & Cilar, L. (2020).
Interpretability of machine learning‐based prediction models in healthcare. Wiley
Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 10(5), e1379.
[44]Szegedy, C., Toshev, A., & Erhan, D. (2013). Deep neural networks for object
detection. In Advances in neural information processing systems (pp. 2553-2561).
[45]Talo, M., Baloglu, U. B., Yıldırım, Ö., & Acharya, U. R. (2019). Application of
deep transfer learning for automated brain abnormality classification using MR images.
Cognitive Systems Research, 54, 176-188.
38
[46] Wang, J., Ma, Y., Zhang, L., Gao, R. X., & Wu, D. (2018). Deep learning for smart
manufacturing: Methods and applications. Journal of Manufacturing Systems, 48, 144-
156.
[47] Wang, J., Fu, P., & Gao, R. X. (2019). Machine vision intelligence for product
defect inspection based on deep learning and Hough transform. Journal of
Manufacturing Systems, 51, 52-60.
[48] Wang, P., Fan, E., & Wang, P. (2021). Comparative analysis of image
classification algorithms based on traditional machine learning and deep learning.
Pattern Recognition Letters, 141, 61-67.
[49] Wang, Y., Liu, M., Zheng, P., Yang, H., & Zou, J. (2020). A smart surface
inspection system using faster R-CNN in cloud-edge computing environment.
Advanced Engineering Informatics, 43, 101037.
[50] Yuan, Z. C., Zhang, Z. T., Su, H., Zhang, L., Shen, F., & Zhang, F. (2018). Visionbased defect detection for mobile phone cover glass using deep neural networks.
International Journal of Precision Engineering and Manufacturing, 19(6), 801-810.
[51] Yun, J. P., Shin, W. C., Koo, G., Kim, M. S., Lee, C., & Lee, S. J. (2020).
Automated defect inspection system for metal surfaces based on deep learning and data
augmentation. Journal of Manufacturing Systems, 55, 317-324.
[52] Yi, L., Li, G., & Jiang, M. (2017). An end‐to‐end steel strip surface defects
recognition system based on convolutional neural networks. steel research international,
88(2), 1600068.
39
[53] Yin, X., Niu, Z., He, Z., Li, Z. S., & Lee, D. H. (2020). Ensemble deep learning
based semi-supervised soft sensor modeling method and its application on quality
prediction for coal preparation process. Advanced Engineering Informatics, 46, 101136.
[54] Zhang, L., Tan, J., Han, D., & Zhu, H. (2017). From machine learning to deep
learning: progress in machine intelligence for rational drug discovery. Drug discovery
today, 22(11), 1680-1685.
[55] Zhang, Q., & Zhu, S. C. (2018). Visual interpretability for deep learning: a survey.
arXiv preprint arXiv:1802.00614.