自動光學檢查(Automated Optical Inspection, AOI)，是一種搭配機器視覺的光學影像檢測系統，用於改良傳統上人工手動檢測的缺點，是在工業製程中相當常見的檢測技術。
現有AOI檢測技術大致上可分為比對為主的檢測技術(Matching-based)以及分類為主的檢測技術(Classification-based)，兩者在檢測上都有各自的缺點。Matching-basesd的檢測技術在照射光線及物品擺放位置固定時能得到良好的檢測效能，但只要光線或位置發生偏移或旋轉，就容易導致誤判。Classification-based的檢測技術雖然能夠直接使用現有的強大模型進行使用，但缺點是必須要對瑕疵部分另外進行資料的收集以及標記。
現今工廠的自動化生產線上的產品種類相當繁多、且外觀各異，單一產品表面不同的區域就含有不同的材質與形狀，而這些元素就導致這些區域檢測的標準不統一，若使用單一標準進行檢測就會導致檢測效能的低落。
本論文的目的為開發出一套能夠解決上述困難的演算法則。本論文使用非監督式學習演算法－Autoencoder來針對現有AOI檢測技術的缺點進行改善。並且針對工廠生產線上種類繁多、外觀各異的產品表面利用影像分割技術進行分別檢測。

[1]Chang, K. H. (2012). Development of optical inspection system for surface mount device light emitting diodes. Master thesis, National Sun Yat-sen University, Taiwan.

[2] 陳宗達 (2005). Design and Development of a CMOS Glass Lid AOI Machine, 國立交通大學工業工程與管理學系碩士論文.

[3] Wang, T., Chen, Y., Qiao, M., & Snoussi, H. (2018). A fast and robust convolutional neural network-based defect detection model in product quality control. The International Journal of Advanced Manufacturing Technology, 94(9-12), 3465-3471.

[4] Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. In Advances in neural information processing systems (pp. 1097-1105).

[5]Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556.

[6]Wang, Y., Yao, H., & Zhao, S. (2016). Auto-encoder based dimensionality reduction. Neurocomputing, 184, 232-242.

[7]Baldi, P. (2012, June). Autoencoders, unsupervised learning, and deep architectures. In Proceedings of ICML workshop on unsupervised and transfer learning (pp. 37-49).

[8] Long, J., Shelhamer, E., & Darrell, T. (2015). Fully convolutional networks for semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 3431-3440).

[9] Hinton, G. E., & Salakhutdinov, R. R. (2006). Reducing the dimensionality of data with neural networks. science, 313(5786), 504-507.

[10] Tabernik, D., Šela, S., Skvarč, J., & Skočaj, D. (2020). Segmentation-based deep-learning approach for surface-defect detection. Journal of Intelligent Manufacturing, 31(3), 759-776.

[11] Yao, W., Zeng, Z., Lian, C., & Tang, H. (2018). Pixel-wise regression using U-Net and its application on pansharpening. Neurocomputing, 312, 364-371.

[12]Ronneberger, O., Fischer, P., & Brox, T. (2015, October). U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical image computing and computer-assisted intervention (pp. 234-241). Springer, Cham.

[13] Brunelli, R. (2009). Template matching techniques in computer vision: theory and practice. John Wiley & Sons.

[14] Wang, B., & Lin, C. Y. (2015). A Robust Method for AOI System Design, Tatung University, Taiwan.

[15] Wei, P., Liu, C., Liu, M., Gao, Y., & Liu, H. (2018). CNN-based reference comparison method for classifying bare PCB defects. The Journal of Engineering, 2018(16), 1528-1533.

[16] Faghih-Roohi, S., Hajizadeh, S., Núñez, A., Babuska, R., & De Schutter, B. (2016, July). Deep convolutional neural networks for detection of rail surface defects. In 2016 International joint conference on neural networks (IJCNN) (pp. 2584-2589). IEEE.

[17]Tao, X., Zhang, D., Ma, W., Liu, X., & Xu, D. (2018). Automatic metallic surface defect detection and recognition with convolutional neural networks. Applied Sciences, 8(9), 1575.

[18] S. Boda (2009), Feature-based image registration, M. Tech Thesis, National Institute of Technology, Rourkela.

[19] McLaughlin, R. A., Hipwell, J., Penney, G. P., Rhode, K., Chung, A., Noble, J. A., & Hawkes, D. J. (2001). Intensity-based registration versus feature-based registration for neurointerventions. Proc. Med. Image. Understanding Analysis.