花樣口罩辨識｜國立中央大學博碩士論文系統

簡易檢索 / 詳目顯示

回結果列表

研究生：	姜亭光 Ting-Guang Jiang
論文名稱：	花樣口罩辨識 The patterned face mask detection
指導教授：	梁德容
口試委員:
學位類別：	碩士 Master
系所名稱：	資訊電機學院 - 資訊工程學系在職專班 Executive Master of Computer Science & Information Engineering
論文出版年：	2022
畢業學年度：	110
語文別：	中文
論文頁數：	60
中文關鍵詞：	影像辨識、人臉偵測、深度學習、口罩辨識
外文關鍵詞：	Image recognition, Face detection, Deep learning, Mask detection
相關次數：	點閱：10 下載：0
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

由於2019年冠狀病毒COVID-19的迅速傳播，已影響了全世界，各國政府與人民正面臨著健康危機，COVID-19是一種呼吸系統疾病，可導致受影響個體出現嚴重的肺炎病例。這種疾病是通過與感染者的直接接觸，以及當感染者咳嗽、打噴嚏或將病毒呼出到空氣中時釋放的唾液珠、呼吸道飛沫而獲得的，因此世界衛生組織WHO提出建議，人們戴口罩在預防COVID-19傳播方面非常有效。

隨著近年人工智慧與電腦影像技術的發展，把相關技術應用於偵測人們是否有佩戴口罩，可以大幅降低對人力的依賴，實現有效地佩戴口罩偵測系統，並能搭配自動化警示機制，使得COVID-19傳播風險大幅下降。

目前應用於口罩辨識的方法遇上佩戴花樣與罩面是深色口罩時，偵測效果較差，因此本研究提出方法針對此問題作出改善，並提高整體辨識準確率。本論文目標是提出一套口罩辨識方法，使用深度學習的架構結合影像處理的演算法達成目標，首先透過人臉識別分類器將人臉影像資料中人臉的部分擷取出來，再利用CNN預訓練模型的基礎改進模型網路結構，經過逐漸調整並優化方法後，實驗結果顯示準確率達到95%，由此證明本方法具備一定程度之可用性。

In 2019, governments and people around the world were facing health
issues due to the rapid spread of the coronavirus, a respiratory
disease that causes severe pneumonia. COVID-19 transmits when people
breathe in air contaminated by droplets and small airborne particles
containing the virus, so the World Health Organization (WHO) recommends
people wear masks to prevent the spread of COVID-19.

In recent years, with development of artificial intelligence and
computer imaging technology, the face mask-wearing detection algorithm
automatically detects masks which can decrease the waste of human
resources and creates alert which can reduce the spread of the disease.

The current method used in mask detection has low accuracy when
people wearing masks with patterns or dark-colored. Therefore, in this
study, improvement will be proposed in order to increase accuracy.

The goal of this research is to propose several mask detection methods by using the deep learning algorithm combined with the image processing algorithm. First, the face part in the image is extracted by the face recognition classifier, and then use the pre-trained CNN model as a basis to gradually improve its model network structure.

According to the experiment, the accuracy rate reaches 95%, which
proves that this method has a certain degree of accuracy after optimization.

摘要 ............................................................. i 
Abstract ................................................. ...... ii 
致謝 ............................................................. iii 
目錄 ............................................................. iv 
圖目錄 ........................................................... vi 
表目錄 ........................................................... viii 
一、 緒論 ......................................................... 1
  1-1 研究背景 .................................................... 1
  1-2 研究動機與目的 ............................................... 2
  1-3 研究目標 .................................................... 3
  1-4 論文架構 .................................................... 3
二、 文獻探討 ...................................................... 4
  2-1 影像分類深度學習模型 .......................................... 4
  2-2 物件偵測技術 ................................................. 7
  2-3 口罩辨識技術 ................................................ 15
三、 解決方案 ..................................................... 19
  3-1 資料前處理 .................................................. 19
  3-2 CNN 模型訓練 ................................................ 21
  3-3 損失函數 .................................................... 24
四、 實驗設計與結果 ................................................. 25
  4-1 實驗環境 .................................................... 25
  4-2 人臉影像資料分析 ............................................. 25
  4-3 K-fold Cross Validation .................................... 27
  4-4 衡量指標 .................................................... 28
  4-5 實驗一:模型訓練 .............................................. 30
    4-5-1 實驗流程 ................................................ 30
    4-5-2 實驗結果 ................................................ 31
  4-6 實驗二:應用人臉識別分類器 ...................................... 32
    4-6-1 實驗流程 ................................................ 32
    4-6-2 實驗結果 ................................................ 33
  4-7 實驗結果比較 ................................................. 34
    4-7-1 數據比較 ................................................ 34
    4-7-2 模型優化比較 ............................................. 35
    4-7-3 辨識失敗特徵比較 .......................................... 36
  4-8 與現有方法比較 ............................................... 40 
五、 結論與未來展望 ................................................. 42 
  5-1 結論 ........................................................ 42 
  5-2 未來展望 ..................................................... 43 
參考文獻 ........................................................... 44
                                

[1] Y. Lecun, L. Bottou, Y. Bengio and P. Haffner, "Gradient-based learning applied to document recognition", in Proceedings of the IEEE, vol. 86, no. 11, pp. 2278-2324, 1998, doi: 10.1109/5.726791.
[2] C. Szegedy et al., "Going deeper with convolutions", in 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2015, pp. 1-9, doi: 10.1109/CVPR.2015.7298594.
[3] K. He, X. Zhang, S. Ren and J. Sun, "Deep Residual Learning for Image Recognition", in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 2016, pp. 770-778, doi: 10.1109/CVPR.2016.90.
[4] K. Simonyan, A. Zisserman. "Very Deep Convolutional Networks for Large-Scale Image Recognition", in 3rd International Conference on Learning Representations (ICLR), 2015, arXiv preprint arXiv:1409.1556, 2014.
[5] F. A. Mohammed Ali, and M. S. Al-Tamimi, "Face mask detection methods and techniques: A review", International Journal of Nonlinear Analysis and Applications, vol. 13, no. 1, pp. 3811-3823, 2022, doi: 10.22075/ijnaa.2022.6166.
[6] N. Jindal, H. Singh, and P. S. Rana, "Face mask detection in COVID-19: a strategic review", Multimed Tools Appl, pp. 1-30, 2022, doi: 10.1007/s11042-022-12999-6.
[7] M. Sandler, A. Howard, M. Zhu, A. Zhmoginov and L. -C. Chen, "MobileNetV2: Inverted Residuals and Linear Bottlenecks", in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018, pp. 4510-4520, doi: 10.1109/CVPR.2018.00474.
[8] J. Redmon and A. Farhadi, "Yolov3: An incremental improvement", arXiv preprint arXiv:1804.02767, 2018.
[9] S. Ren, K. He, R. Girshick and J. Sun, "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks", in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 06, pp. 1137-1149, 2017, doi: 10.1109/TPAMI.2016.2577031.
[10] J. Redmon, S. Divvala, R. Girshick and A. Farhadi, "You Only Look Once: Unified, Real-Time Object Detection", in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2016, pp. 779-788, doi: 10.1109/CVPR.2016.91.
[11] W. Liu et al., "SSD: Single shot multibox detector", in Proceedings of the European Conference on Computer Vision (ECCV), 2016, vol. 9905, pp. 21-37, doi: 10.1007/978-3-319-46448-0_2.
[12] J. Redmon and A. Farhadi, "YOLO9000: Better, Faster, Stronger", in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 6517-6525, doi: 10.1109/CVPR.2017.690.
[13] Bochkovskiy, A., Wang, C. Y., and Liao, H. Y. M., "Yolov4: Optimal speed and accuracy of object detection", arXiv preprint arXiv:2004.10934, 2020.
[14] Wang, C. Y., Bochkovskiy, A., and Liao, H. Y. M., "YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors", arXiv preprint arXiv:2207.02696, 2022.
[15] Y. Lee, J. -w. Hwang, S. Lee, Y. Bae and J. Park, "An Energy and GPU-Computation Efficient Backbone Network for Real-Time Object Detection,", in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2019, pp. 752-760, doi: 10.1109/CVPRW.2019.00103.
[16] L. Xie et al., "Elan: Towards Generic and Efficient Elastic Training for Deep Learning", in 2020 IEEE 40th International Conference on Distributed Computing Systems (ICDCS), 2020, pp. 78-88, doi: 10.1109/ICDCS47774.2020.00018.
[17] A. Navada, A.N. Ansari, S. Patil and B.A. Sonkamble, "Overview of use of decision tree algorithms in machine learning", in 2011 IEEE Control and System Graduate Research Colloquium, 2011, pp. 37-42, doi: 10.1109/ICSGRC.2011.5991826.
[18] M. A. Hearst, S. T. Dumais, E. Osuna, J. Platt and B. Scholkopf, "Support vector machines", in IEEE Intelligent Systems and their Applications, Aug. 1998, vol. 13, no. 4, pp. 18-28, doi: 10.1109/5254.708428.
[19] M. Loey, G. Manogaran, M. H. N. Taha, and N. E. M. Khalifa, "A hybrid deep transfer learning model with machine learning methods for face mask detection in the era of the COVID-19 pandemic", Meas. J. Int. Meas. Confed., vol. 167, pp. 108-288, 2021, doi:10.1016/j.measurement.2020.108288.
[20] S. Yadav, "Deep Learning based Safe Social Distancing and Face Mask Detection in Public Areas for COVID-19 Safety Guidelines Adherence", Int. J. Res. Appl. Sci. Eng. Technol., vol. 8, no. 7, pp. 1368-1375, 2020, doi: 10.22214/ijraset.2020.30560.
[21] M. Jiang, X. Fan, "RetinaMask: a face mask detector", arXiv preprint arXiv:2005.03950, 2020.
[22] M. Ucar, "Face Mask Detection Using YOLOv4", in 4th International Conference on Data Science and Applications (ICONDATA'21), 2021, pp. 411-416.
[23] P. Nagrath et al., "SSDMNV2: A real time DNN-based face mask detection system using single shot multibox detector and MobileNetV2", Sustainable Cities and Society. 2021 Aug;71:102964, doi: 10.1016/j.scs.2021.102964.
[24] P. Viola and M. Jones, "Rapid object detection using a boosted cascade of simple features", in Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), 2001, pp. I-I, doi: 10.1109/CVPR.2001.990517.
[25] Tommy Huang, "機器/深度學習: 基礎介紹-損失函數(loss function)", [Online]. Available: https://medium.com/chih-sheng-huang821/2dcac5ebb6cb [Accessed 14 July 2022].
[26] G. Shiming et al., "mafa-dataset", [Online]. Available: https://www.kaggle.com/datasets/revanthrex/mafadataset [Accessed 25 March 2022].

簡易檢索 / 詳目顯示

相關論文