COVID-19 疫情下，佩戴口罩被視為減少病毒傳播風險的關鍵之一。然而由於民眾在公共場所必須佩戴口罩，因此在人臉辨識（Face Recognition）上的困難度就增加了。本研究設計出了一套結合口罩佩戴辨識與戴口罩人臉辨識的系統，能判斷口罩佩戴與否、口罩是否正確佩戴，同時進行戴口罩人臉辨識。

　　我們的系統僅使用MTCNN與FaceNet兩個模型。利用FaceNet作為預訓練模型，先訓練出戴口罩人臉辨識的部分，再利用已訓練完的戴口罩人臉辨識模型，作為口罩佩戴辨識的預訓練模型，以同一個網路架構即達成三項任務，因此大幅降低了訓練成本。即使訓練成本很低，我們仍在任務上達到了一定的水準。

　　Under the COVID-19 outbreak, wearing masks is seen as one of the keys to reducing the risk of virus transmission. However, because people must wear masks in public places, the difficulty of Face Recognition is increased. In this study, we design a system combining face mask wearing recognition and masked face recognition, which can check whether the face mask is worn or not, whether the face mask is correctly worn, and at the same time, masked face recognition can be carried out.

　　We only use two models, MTCNN and FaceNet, in our system. FaceNet is used as the pre-trained model to train the masked face recognition model first, and then the trained masked face recognition model is used as the pre-trained model of the face mask checking models. Three tasks can be achieved in the same network architecture, so the training cost is greatly reduced. Even though the training costs are low, we still achieve a certain level of performance on the tasks.

誌謝　i
摘要　ii
Abstract　iii
目錄　iv
圖目錄　vi
表目錄　vii
第一章　緒論　1
1.1　研究背景　1
1.2　研究動機　2
1.3　論文架構　2
第二章　相關研究　3
2.1　人臉偵測與對齊　3
2.2　人臉特徵擷取與比對　5
2.3　Transfer Learning　14
第三章　實驗　15
3.1　實驗流程　15
3.2　實驗環境　16
3.3　資料集　17
3.4　戴口罩的人臉辨識　18
3.5　口罩佩戴辨識　21
第四章　實驗結果與討論　25
4.1　實驗結果　25
4.2　問題討論　28
第五章　結論與未來展望　29
5.1　結論　29
5.2　未來展望　29
參考文獻　31

[1] P. Viola and M. Jones, "Rapid object detection using a boosted cascade of simple features," Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001, Kauai, HI, USA, 2001, pp. I-I, doi: 10.1109/CVPR.2001.990517.
[2] P. Viola and M. Jones, "Robust real-time face detection," Int. Journal of Computer Vision, vol.57, no.2, pp.137-154, 2004.
[3] T. Ojala, M. Pietikainen and D. Harwood, "Performance evaluation of texture measures with classification based on Kullback discrimination of distributions," Proceedings of 12th International Conference on Pattern Recognition, Jerusalem, Israel, 1994, pp. 582-585 vol.1, doi: 10.1109/ICPR.1994.576366.
[4] N. Dalal and B. Triggs, "Histograms of oriented gradients for human detection," 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05), San Diego, CA, USA, 2005, pp. 886-893 vol. 1, doi: 10.1109/CVPR.2005.177.
[5] P. F. Felzenszwalb, R. B. Girshick, D. McAllester and D. Ramanan, "Object Detection with Discriminatively Trained Part-Based Models," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 32, no. 9, pp. 1627-1645, Sept. 2010, doi: 10.1109/TPAMI.2009.167.
[6] D. E. King, (2009). "Dlib-ml: A Machine Learning Toolkit,". Journal of Machine Learning Research. 10 (Jul): 1755–1758. CiteSeerX 10.1.1.156.3584
[7] Xin Jin and Xiaoyang Tan. “Face alignment in-the-wild,” Computer Vision and Image Understanding, 1–22, Sept. 2017. DOI:https://doi.org/10.1016/j.cviu.2017.08.008
[8] T. F. Cootes, G. J. Edwards, and C. J. Taylor, “Active appearance models,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 23, no. 6, pp. 681-685, 2001.
[9] X. Yu, J. Huang, S. Zhang, W. Yan, and D. Metaxas, “Pose-free facial landmark fitting via optimized part mixtures and cascaded deformable shape model,” in IEEE International Conference on Computer Vision, 2013, pp. 1944-1951.
[10] X. Zhu, and D. Ramanan, “Face detection, pose estimation, and landmark localization in the wild,” in IEEE Conference on Computer Vision and Pattern Recognition, 2012, pp. 2879-2886.
[11] X. P. Burgos-Artizzu, P. Perona, and P. Dollar, “Robust face landmark estimation under occlusion,” in IEEE International Conference on Computer Vision, 2013, pp. 1513-1520.
[12] X. Cao, Y. Wei, F. Wen, and J. Sun, “Face alignment by explicit shape regression,” International Journal of Computer Vision, vol 107, no. 2, pp. 177-190, 2012.
[13] J. Zhang, S. Shan, M. Kan, and X. Chen, “Coarse-to-fine auto-encoder networks (CFAN) for real-time face alignment,” in European Conference on Computer Vision, 2014, pp. 1-16.
[14] K. Zhang, Z. Zhang, Z. Li, and Y. Qiao, “Joint face detection and alignment using multitask cascaded convolutional networks,” IEEE Signal Processing Letters, vol. 23, no. 10, pp. 1499–1503, 2016.
[15] F. Schroff, D. Kalenichenko, and J. Philbin, "Facenet: A unified embedding for face recognition and clustering," in Proc. IEEE Conf. on Computer Vision and Pattern Recognition, Boston, MA, Jun.7-12, 2015, pp.815-823.
[16] M. D. Zeiler and R. Fergus. Visualizing and understanding convolutional networks. CoRR, abs/1311.2901, 2013. 2, 3, 4, 6
[17] C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich, "Going deeper with convolutions," in Proc. IEEE Int. Conf. on Computer Vision and Pattern Recognition (CVPR), Boston, MA, Jun.7-12, 2015, pp.1-9, doi: 10.1109/CVPR.2015.7298594.
[18] Davidsandberg, “facenet: Face recognition using Tensorflow,” https://github.com/davidsandberg/facenet, accessed: 2021-01-15.
[19] C. Szegedy, S. Ioffe, V. Vanhoucke, and A. A. Alemi, “Inception-v4, Inception- ResNet and the impact of residual connections on learning,” In Thirty-First AAAI Conference on Artificial Intelligence, 2017.
[20] A. G. Howard, M. Zhu, B. Chen et al., "Mobilenets: efficient convolutional neural networks for mobile vision applications," arXiv preprint arXiv:1704.04861, 2017.
[21] K. He, X. Zhang, S. Ren and J. Sun, "Deep Residual Learning for Image Recognition," 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, 2016, pp. 770-778, doi: 10.1109/CVPR.2016.90.
[22] M. Lin, Q. Chen, and S. Yan, “Network in network,” arXiv:1312.4400, Mar. 2014.
[23] C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens and Z. Wojna, "Rethinking the Inception Architecture for Computer Vision," in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 2016 pp. 2818-2826. doi: 10.1109/CVPR.2016.308
[24] L. Goldmann, U. J. Monich and T. Sikora, "Components and Their Topology for Robust Face Detection in the Presence of Partial Occlusions," in IEEE Transactions on Information Forensics and Security, vol. 2, no. 3, pp. 559-569, Sept. 2007, doi: 10.1109/TIFS.2007.902019.
[25] M. De Marsico, M. Nappi and D. Riccio, "FARO: FAce Recognition Against Occlusions and Expression Variations," in IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans, vol. 40, no. 1, pp. 121-132, Jan. 2010, doi: 10.1109/TSMCA.2009.2033031.
[26] S. J. Pan and Q. Yang, "A Survey on Transfer Learning," in IEEE Transactions on Knowledge and Data Engineering, vol. 22, no. 10, pp. 1345-1359, Oct. 2010, doi: 10.1109/TKDE.2009.191.
[27] J. Yosinski, J. Clune, Y. Bengio, and H. Lipson, “ How transferable are features in deep neural networks?” in Advances in Neural Information Processing Systems, 2014.
[28] X-zhangyang, “Real-World Masked Face Dataset,” https://github.com/X-zhangyang/Real-World-Masked-Face-Dataset, accessed: 2021-01-15.
[29] Aqeelanwar, “MaskTheFace,” https://github.com/aqeelanwar/MaskTheFace, accessed: 2021-01-15.
[30] Gary B. Huang, Manu Ramesh, Tamara Berg, and Erik Learned-Miller. “Labeled Faces in the Wild: A Database for Studying Face Recognition in Unconstrained Environments” University of Massachusetts, Amherst, Technical Report 07-49, October, 2007.
[31] DataFlair, “Face Mask Data,” https://data-flair.training/blogs/download-face-mask-data/, accessed: 2021-01-15.
[32] Adnane Cabani, Karim Hammoudi, Halim Benhabiles, and Mahmoud Melkemi, "MaskedFace-Net - A dataset of correctly/incorrectly masked face images in the context of COVID-19", Smart Health, ISSN 2352-6483, Elsevier, 2020. https://doi.org/10.1016/j.smhl.2020.100144 [Preprint version available at arXiv:2008.08016]
[33] Dong Yi, Zhen Lei, Shengcai Liao and Stan Z. Li, “Learning Face Representation from Scratch,” 2014. arXiv preprint arXiv:1411.7923.
[34] Sebastian Ruder, “An overview of gradient descent optimisation algorithms,” 2016. arXiv preprint arXiv:1609.04747