Identification of people without a mask using Deep Learning
DOI:
https://doi.org/10.52428/20758944.v18i52.236Keywords:
Computer vision, Deep learning, Convolutional neural networks, Object detection, YOLOAbstract
This work makes use of convolutional neural networks to detect people with and without mask, due to the current situation of COVID-19 and in accordance with the biosecurity measures instructed by government and health institutions, it has been proven in a way that the use of surgical masks or chinstraps help to reduce the risk of contagion of the disease, the need to be able to detect or identify people who are not wearing a mask becomes evident, not complying with this biosecurity measure and putting a group of the population at risk.
Initially, a training repository was established consisting of images of people with and without masks, these images were obtained from different sources.
Three types of convolutional neural networks have been trained and compared, Faster R-CNN, SSD (Single Shot MultiBox Detector) and YOLO (You Only Look Once), each one performs the detection of people with and without masks, standing out one from the other due to its speed, precision, or performance.
To obtain the object detection models, Darknet and TensorFlow Object Detection API frameworks have been used, Google Colab was used too, which, being a free provider, it also provided powerful computational features.
Downloads
References
Bisong, E. (2019). Google Colaboratory. En Building Machine Learning and Deep Learning Models on Google Cloud Platform. Apress. https://doi.org/10.1007/978-1-4842-4470-8_7 DOI: https://doi.org/10.1007/978-1-4842-4470-8
Bochkovskiy, A. (2021). GitHub - AlexeyAB/darknet: YOLOv4 / Scaled-YOLOv4 / YOLO - Neural Networks for Object Detection (Windows and Linux version of Darknet). Recuperado el 27 de Marzo de 2021, de GitHub: https://github.com/AlexeyAB/darknet
Bochkovskiy, A. (2021). Running a YOLOv4 Object Detector with Darknet in the Cloud! (GPU ENABLED). Recuperado el 27 de Marzo de 2021, de Google Colab: https://colab.research.google.com/drive/1_GdoqCJWXsChrOiY8sZMr_zbr_fH-0Fg
Bochkovskiy, A., Wang, C.-Y., & Liao, H.-Y. M. (2020). YOLOv4: Optimal Speed and Accuracy of Object Detection. arXiv:2004.10934.
Girshick, R. (2015). Fast R-CNN. Proceedings of the IEEE International Conference on Computer Vision, (págs. 1440-1448). https://doi.org/10.1109/ICCV.2015.169 DOI: https://doi.org/10.1109/ICCV.2015.169
Joshi, N. (2020). How to fine-tune your artificial intelligence algorithms. Recuperado el 20 de Abril de 2022, de Allerin: https://www.allerin.com/blog/how-to-fine-tune-your-artificial-intelligence-algorithms
Khandelwal, R. (2019). COCO and Pascal VOC data format for Object detection. Recuperado el 28 de Marzo de 2021, de Towards Data Science: https://towardsdatascience.com/coco- data-format-for-object-detection-a4c5eaf518c5
Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.-Y., & Berg, A. (2016). SSD: Single Shot MultiBox Detector. European Conference on Computer Vision, (pp. 21-37). https://doi.org/10.1007/978-3-319-46448-0_2 DOI: https://doi.org/10.1007/978-3-319-46448-0_2
Padilla, R., Netto, S. L., & da Silva, E. A. (2020). A Survey on Performance Metrics for Object-Detection Algorithms. International Conference on Systems Signals and Image Processing. https://doi.org/10.1007/978-3-319-46448-0_2 DOI: https://doi.org/10.1109/IWSSIP48289.2020.9145130
Pan, S. J., Yang, Q. (2010). A Survey on Transfer Learning. IEEE Transactions on Knowledge and Data Engineering, vol. 22, (pp. 1345-1359), doi: https://doi.org/10.1007/978-3-319-46448-0_2 DOI: https://doi.org/10.1109/TKDE.2009.191
Ponnusamy, A. (24 de Marzo de 2021). Preparing Custom Dataset for Training YOLO Object Detector. Obtenido de Vision Geek: https://www.visiongeek.io/2019/10/preparing- custom-dataset-for-training-yolo-object-detector.html
Redmon, J. (2013-2016). Darknet: Open Source Neural Networks in C. Obtenido de http://pjreddie.com/darknet/
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You Only Look Once: Unified, Real-Time Object Detection. IEEE. https://doi.org/10.1109/CVPR.2016.91 DOI: https://doi.org/10.1109/CVPR.2016.91
TensorFlow 2 Detection Model Zoo. (1 de Mayo de 2021). Obtenido de GitHub:
https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf2_d etection_zoo.md
TensorFlow. (1 de Mayo de 2021). Obtenido de www.tensorflow.org https://doi.org/10.1007/978- 1-4842-6373-0_1
TensorFlow Object Detection API. (1 de Mayo de 2021). Obtenido de https://github.com/tensorflow/models/tree/master/research/object_detection
Additional Files
Published
How to Cite
Issue
Section
License
Copyright (c) 2022 Giovanny German Rocha Vallejo
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License 4.0 that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
