Robotic based mask detection to prevent epidemic diseases transmitted through droplets using pre-trained deep learning models

Öz

The Coronavirus disease, which emerged in Wuhan, China in December 2019 and spread rapidly all over the world, infected healthy people by being transmitted by small droplets. Medical experts have stated that the most effective fight against the Coronavirus disease is the need for people in contact to wear masks. Despite this, some people violated the obligation to wear masks. In this study, mask detection performances of pre-trained Convolutional Neural Network (CNN) models such as NasNetMobile, MobileNetV3Small, ResNet50, DenseNet121 and EfficientNetV2B0, which were previously trained, were evaluated in order to automatically detect people who violate the mask wearing obligation. At the end of this evaluation, DenseNet121 architechture has become the most successful model. This model has been tested with the image obtained from the camera on a robotic system with six Degrees of Freedom (6-DOF). The human face images taken from the camera were processed using the Jetson Xavier NX development board. As a result, this study will help the officers who carry out mask inspections in public areas and will significantly reduce the spread of new outbreaks similar to the Coronavirus.

Anahtar Kelimeler

• Mask detection
• Epidemic diseases
• Convolutional neural network
• Transfer learning
• Robotic

Önceden eğitilmiş derin öğrenme modelleri kullanılarak damlacık yoluyla bulaşan salgın hastalıkları önlemek için robotik tabanlı maske tespiti

Öz

Aralık 2019’da Çin’in Wuhan şehrinde ortaya çıkan ve tüm dünyada hızla yayılan Koronavirüs hastalığı solunum yolu sonucu oluşan küçük damlacıklar ile bulaşarak sağlıklı insanları enfekte etmiştir. Tıp uzmanları Koronavirüs hastalığına karşı en etkili mücadelenin temas halindeki kişilerin maske takması gerekliliğini belirtmişlerdir. Buna rağmen bazı kişiler maske takma zorunluluğunu ihlal etmişlerdir. Bu çalışmada maske takma zorunluluğunu ihlal eden kişilerin otomatik olarak tespit edilebilmesi için önceden eğitilmiş olan NaNetMobile, MobileNetV3Small, ResNet50, DenseNet121 ve EfficientNetV2B0 gibi derin sinir ağı modellerinin maske tanıma performansları değerlendirilmiştir. Bu değerlendirme sonucunda en başarılı DenseNet121 modeli ele edilmiştir. Bu model 6- Serbestlik Derecesine (6-DOF) sahip robotik bir sisteminin üzerinde yer alan kameradan elde edilen görüntü ile doğrulanmıştır. Kameradan alınan insane ait yüz görüntüleri yüksek kare hızlarında NVIDIA tarafından geliştirilen Jetson Xavier geliştirme kartı kullanılarak işlenmiştir. Sonuç olarak, bu çalışma toplu alanlarda maske denetimi gerçekleştiren görevlilere yardımcı olacak ve Koronavirüs benzeri çıkabilecek yeni salgınların yayılımı önemli ölçüde azaltacaktır.

Anahtar Kelimeler

• Maske Tanıma
• Epidemik hastalıklar
• Evrişimli sinir ağı
• Transfer öğrenme
• Robotik

Kaynakça

1. World Health Organization, Novel coronavirus (‎2019-nCoV): situation report, 1, https://apps.who.int/ iris/bitstream/handle/10665/330760/nCoVsitrep21Jan2020-eng.pdf?sequence=3&isAllowed=y, Accessed 20 March 2023.
2. World Health Organization, Naming the coronavirus disease (COVID-19) and the virus that causes it, https://www.who.int/emergencies/diseases/novel-coronavirus-2019/technical-guidance/naming-the-coronavirus-disease-(covid-2019)-and-the-virus-that-causes-it, Accessed 20 March 2023.
3. P. Bahl, C. Doolan, C. de Silva, A. A. Chughtai, L. Bourouiba and C. R. MacIntyre, Airborne or Droplet precautions for health workers treating COVID-19?. The Journal of Infectious Diseases, 225, 9, 1561-1568, 2022, https://doi.org/10.1093/infdis/jiaa189.
4. G. Jignesh Chowdary, N. S. Punn, S. K. Sonbhadro and S. Agarwal, Face mask detection using transfer learning of InceptionV3. In: L. Bellatreche, V. Goyal, H. Fujita, A. Mondal, P. K. Reddy, Eds. Big Data Analytics. BDA 2020. Lecture Notes in Computer Science, 12581, Springer, Cham, pp. 81-90, 2020.
5. A. Cabani, K. Hammoudi, H. Benhabiles and M. Melkemi, MaskedFace-Net -- A dataset of correctly/incorectly masked face images in the context of COVID-19. Smart Health, 19, 1-5, 2021. https://doi.org/10.1016/j.smhl.2020.100144.
6. X. Kong, K. Wang, S. Wang, X Wang, X. Jiang, Y. Guo, G. Shen, X. Chen and Q. Ni, Real-time mask identification for COVID-19: An edge computing-based deep learning framework. IEEE Internet of Things Journal, 8 (21), 1-10, 2021. https:// 10.1109/JIOT.2021.3051844.
7. J. Brownlee, A Gentle introduction to transfer learning for deep learning, https://machinelearni ngmastery.com/transfer-learning-for-deep-learning/ Accessed 22 February 2023.
8. JavaTPoint, Introduction to transfer learning in ML, https://www.javatpoint.com/transfer-learning-in-mach ine-learning. Accessed 30 February 2023.

9. W. Wang, and J. Gang, Application of convolutional neural network in natural language processing. In 2018 International Conference on Information Systems and Computer Aided Education (ICISCAE), IEEE, pp. 64-70, Changchun, China, 2018.
10. D. N. N. Tran, L. H. Pham, H. H. Nguyen and J. W. Jeon, A Vision-Based method for real-time traffic flow estimation on edge devices. IEEE Transactions on Intelligent Transportation Systems, 1-15, 2023. https:// 10.1109/TITS.2023.3264796.
11. H. Avula, R. Ranjith, and A. S. Pillai, CNN based recognition of emotion and speech from gestures and facial expressions. In 2022 6th International Conference on Electronics, Communication and Aerospace Technology, IEEE, pp. 1360-1365, Coimbatore, India, 2022.
12. T. Kalaiselvi, P. Sriramakrishnan, and K. Somasundaram, Survey of using GPU CUDA programming model in image analysis. Informatics in Medicine Unlocked, 9, 133-144, 2017. https://doi.org/10.1016/j.imu.2017.08.001.
13. NVIDIA, Jetson Xavier NX Development Kit, Santa Clara, California, ABD, User Guide DA_09814-002, 19 May 2020.
14. M. Tan, B. Chen, R. Pang, V. Vasudevan, M. Sandler, A. Howard, and Q. V. Le, Mnasnet: Platform-aware neural architecture search for mobile. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, (CVPR), pp. 2820-2828, Long Beach, CA, USA, 2019.
15. A. Howard, M. Sandler, G. Chu, L. C. Chen, B. Chen, M. Tan, and H. Adam, Searching for mobilenetv3. In Proceedings of the IEEE/CVF international conference on computer vision pp. 1314-1324, Seoul, Korea (South), 2019.
16. P. Dwivedi, Understanding and coding a ResNet in Keras, https://towardsdatascience.com/understanding-and-coding-a-resnet-in-keras-446d7ff84d33, Accessed 10 January 2023.
17. N. Radwan, Leveraging sparse and dense features for reliable state estimation in urban environments. Ph.D. Thesis, Technische Fakultat¨ Albert-Ludwigs-Universitat Freiburg, Germany, 2019.
18. N. H. Shabrina, R. A. Lika and S. Indarti, Deep learning models for automatic identification of plant-parasitic nematode. Artificial Intelligence in Agriculture, 7, 1-12, 2023. https://doi.org/10.1016/j.aiia.2022.12.002.
19. F. Eryılmaz and H. A. Karacan, Akciğer X-Ray görüntülerinden COVID-19 tespitinde hafif ve geleneksel evrişimsel sinir ağ mimarilerinin karşılaştırılması. Düzce Üniversitesi Bilim ve Teknoloji Dergisi, 9 (6), 26-39, 2021. https://doi.org/10.29130/dubited.1011829.
20. J. Jodriguez, The evolution of Google’s MobileNet architechtures to improve computer vision models, https://medium.com/dataseries/the-evolution-of-googl es-mobilenet-architectures-to-improve-computer-visio n-models-ffb483ffcc0a, Accessed 12 February 2023.
21. S. Bangar, ResNet architecture expained, https://medium.com/@siddheshb008/resnet-architectu re-explained-47309ea9283d, Accessed 01 January 2023.
22. M. Shafiq and Z. Gu, Deep residual learning for image recognition: a survey. Applied Sciences, 12 (18), 8972, 1-43, 2022. https://doi.org/10.3390/app12188972.
23. Z. Feng, An Overview of ResNet architecture and ıts variations, https://builtin.com/artificial-intelligence/ resnet-architecture, Accessed 18 February 2023.
24. S-H. Tsang, Review: DenseNet—Dense convolutional network (Image classification), https://towardsdatasci ence.com/review-densenet-image-classification-b6631 a8ef803, Accessed 22 January 2023.
25. B. Lodhi and J. Kang, Multipath-DenseNet: A Supervised ensemble architecture of densely connected convolutional networks. Information Sciences, 482, 63-72. 2019. https://doi.org/10.1016/j.ins.2019.01.012.
26. G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, Densely connected convolutional networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 4700-4708, Honolulu, HI, USA, 2017.
27. M. Tan and Q. Le, Efficientnet: Rethinking model scaling for convolutional neural networks. In International conference on machine learning, pp. 6105-6114, PMLR, 2019. https://doi.org/10.48550/ar Xiv.1905.11946.
28. H. Alhichri, A. S. Alswayed, Y. Bazi, N. Ammour and N. A. Alajlan, Classification of remote sensing images using EfficientNet-B3 CNN model with attention. IEEE Access, 9, 14078-14094, 2021. 10.110 9/ACCESS.2021.3051085.
29. D. Misra, Mish: A self regularized non-monotonic activation function. arXiv preprint arXiv:1908.08681, 2020, https://doi.org/10.48550/arXiv.1908.08681.
30. S. Balaji, Face-Mask-Detection, 2021, https://github.com/balajisrinivas/Face-Mask-Detection /tree/master /dataset/ with_mask, Accessed 14 January 2023.