Automatic Control of Using Medical Mask in Public Space by Deep Transfer Learning Approach

Abstract

The main transmission routes of COVID-19, an international public health emergency, are respiratory droplets and physical contact. As one of the comprehensive strategies to prevent and fight against disease the outbreak, the use of medical masks in the public sphere has been made mandatory in many societies. In this context, automatic control of the use of medical masks in the public sphere is crucial in the fight against the outbreak. This study aimed to detect the use of medical masks automatically from images of the public sphere by the transfer learning approach. By transfer learning approach to deep architecture, it is aimed to obtain effective solutions in medical mask detection with fine-tuning of pre-trained parameters. The image data set offered by Human in the Loop (HITL) as open access was used for the automatic detection of medical masks. The SqueezeNet based on transfer learning approach proposed in this study achieved a classification accuracy of 99.20%. In addition, the AUC (area under the ROC curve) value was found as 0.998. To emphasize the superiority of the transfer learning approach, the SqueezeNet architecture, which does not contain trained parameters, was also applied to the same data set and the obtained performance metrics were compared. The model trained from scratch on a limited number of image dataset offered classification accuracy and AUC performances as 94.75% and 0.976, respectively. As a result, it has been observed that the deep architecture, which has been trained in a very short time with the transfer learning approach, has an impressive performance in detecting the use of medical masks.

Keywords

• Transfer learning
• Deep learning
• SqueezeNet
• Classification
• Image processing
• COVID-19

Derin Transfer Öğrenimi Yaklaşımı ile Kamusal Alanda Medikal Maske Kullanımının Otomatik Kontrolü

Abstract

Uluslararası kamu sağlığı acil durumu olan COVID-19 hastalığının başlıca bulaşma yolları, solunum damlacıkları ve fiziksel temastır. Hastalığın yayılımını önlemek ve salgınla mücadele etmenin kapsamlı stratejilerinden biri olarak kamusal alanda medikal maske kullanımı birçok toplumda zorunlu kılınmıştır. Bu kapsamda, kamusal alanda medikal maske kullanımının otomatik olarak kontrolü, salgınla mücadelede önem arz etmektedir. Bu çalışmada, transfer öğrenimi yaklaşımı ile kamusal alandan alınan görüntülerden medikal maske kullanımının otomatik olarak tespit edilmesi amaçlanmıştır. Derin mimariye transfer öğrenimi yaklaşımı uygulanarak, öğrenilmiş parametrelerinin ince ayarı ile medikal maske tespitinde etkili çözümlerin elde edilmesi amaçlanmıştır. Medikal maske kullanımının otomatik olarak tespitinde, Human in the Loop (HITL) tarafından erişime açık olarak sunulan görüntüler kullanılmıştır. SqueezeNet tabanlı transfer öğrenimi yaklaşımı ile %99,20 oranında sınıflandırma doğruluğu elde edilmiştir. ROC eğrisi altında kalan alanın (AUC) büyüklüğü ise 0,998 olarak elde edilmiştir. Ayrıca, transfer öğrenimi yaklaşımının üstünlüğünü vurgulamak için eğitilmiş parametre içermeyen SqueezeNet mimarisi de aynı veri seti üzerinde uygulanmış ve elde edilen performans değerleri karşılaştırılmıştır. Sınırlı sayıda görüntü veri kümesi üzerinde eğitilen mimari ile sınıflandırma doğruluğu ve AUC performansları sırasıyla %94,75 ve 0,976 olarak elde edilmiştir. Transfer öğrenimi yaklaşımı ile çok kısa sürede eğitilen derin mimarinin medikal maske kullanımı tespitinde etkileyici bir performans sergilediği gözlemlenmiştir.

Keywords

• Transfer öğrenimi
• Derin Öğrenme
• Sınıflandırma
• SqueezeNet
• Görüntü İşleme
• COVID-19

References

1. [1] Toğaçar M, Ergen B, Cömert Z. COVID-19 detection using deep learning models to exploit Social Mimic Optimization and structured chest X-ray images using fuzzy color and stacking approaches. Comput. Biol. Med. 2020;121:103805.
2. [2] Pereira RM, Bertolini D, Teixeira LO, Silla Jr CN, Costa YMG. Covid-19 identification in chest x-ray images on flat and hierarchical classification scenarios. Comput Methods Programs Biomed. 2020;94;105532.
3. [3] ICTV 2020 [internet]. [cited 14.02.2020]. Available from: https://talk.ictvonline.org/.
4. [4] WHO 2021 [internet]. [cited 30.05.2021]. Available from: https://www.who.int/emergencies/diseases/novel-coronavirus-2019/.
5. [5] Loey M, Monogaran G, Taha MHN, Khalifa NEM. A hybrid deep transfer learning model with machine learning methods for face mask detection in the era of the COVID-19 pandemic, Measurement. 2021;167: 108288.
6. [6] Butt C, Gill J, Chun D, Babu BA. Deep learning system to screen coronavirus disease 2019 pneumonia. Appl Intell. 2020;1-7.
7. [7] Ozturk T, Talo M, Yildirim EA, Baloglu UB, Yildirim O, Acharya UR. Automated detection of covid-19 cases using deep neural networks with x-ray images. Comput. Biol. Med. 2020;121:103792.
8. [8] Ucar F, Korkmaz D. COVIDiagnosis-Net: Deep Bayes-SqueezeNet based diagnosis of the coronavirus disease 2019 (COVID-19) from X-ray images. Med. Hypotheses. 2020;140: 109761.

9. [9] Batagelj B, Peer P, Štruc V, Dobrišek S. How to Correctly Detect Face-Masks for COVID-19 from Visual Information?. Appl. Sci. 2021;11:2070.
10. [10] Khandelwal P, Khandelwal A, Agarwal S. Using Computer Vision to enhance Safety of Workforce in Manufacturing in a Post COVID World. arXiv 2020, arXiv:2005.05287.
11. [11] Qin B, Li D. Identifying Facemask-wearing Condition Using Image Super-Resolution with Classification Network to Prevent COVID-19. Sensors. 2020;20:5236.
12. [12] HITL 2021 [internet]. [cited 02.04.2021]. Available from: https://humansintheloop.org/resources/datasets/medical-mask-dataset/
13. [13] Zhang Q, Yang LT, Chen Z, Li P. A survey on deep learning for big data. Inform. Fus. 2019;42:146-157.
14. [14] Fayek HM, Lech M, Cavedon L. Evaluating deep learning architectures for speech emotion recognition. Neural Network. 2017;92:60-68.
15. [15] Alom MZ, Taha TM, Yakopcic C, Westberg S, Sidike P, Nasrin MS, et al. A state-of-the-art survey on deep learning theory and architectures. Electronics. 2019;8:292.
16. [16] Acharya UR, Oh SL, Hagiwara Y, Tan JH, Adeli H. Deep convolutional neural network for the automated detection and diagnosis of seizure using EEG signals, Comput. Biol. Med. 2018;100:270-278.
17. [17] Iandola FN, Han S, Moskewicz MW, Ashraf K, Dally WJ, Keutzer K. SqueezeNet: AlexNet-Level Accuracy with 50x Fewer Parameters And< 0.5 MB Model Size, arXiv preprint. 2016;arXiv:1602.07360.
18. [18] Minaee S, Kafieh R, Sonka M, Yazdani S, Jamalipour Soufi G. Deep-COVID: predicting COVID-19 from chest X-ray images using deep transfer learning, Med. Image Anal. 2020;65:101794.
19. [19] Xiao T, Liu L, Li K, Qin WJ, Yu SD, Li ZC. Comparison of transferred deep neural networks in ultrasonic breast masses discrimination, Biomed Res. Int. 2018;2018:4605191.
20. [20] Rao TS, Devi SA, Dileep P, Ram MS. A Novel Approach To Detect Face Mask To Control Covid Using Deep Learning, European Journal of Molecular & Clinical Medicine. 2020;7(6):658-668.