Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Sep 9;19(9):e0308460.
doi: 10.1371/journal.pone.0308460. eCollection 2024.

Innovation in public health surveillance for social distancing during the COVID-19 pandemic: A deep learning and object detection based novel approach

Mohammad Arifuzzaman  1   2 , Rizwan Shaikh  3   2 , Iftekhar Bhuiyan  2 , S M Raiyan Chowdhury  3   2   4 , Tauhidul Islam Atoul  2
Affiliations

Innovation in public health surveillance for social distancing during the COVID-19 pandemic: A deep learning and object detection based novel approach

Mohammad Arifuzzaman et al. PLoS One. .

Abstract

The Corona Virus Disease (COVID-19) has a huge impact on all of humanity, and people's disregard for COVID-19 regulations has sped up the disease's spread. Our study uses a state-of-the-art object detection model like YOLOv4 (You Only Look Once, version 4), a very effective tool, on real-time 25fps, 1920 X 1080 video data streamed live by a camera-mounted Unmanned Aerial Vehicle (UAV) quad-copter to observe proper maintenance of social distance in an area of 35m range in this study. The model has demonstrated remarkable efficacy in identifying and quantifying instances of social distancing, with an accuracy of 82% and little latency. It has been able to work efficiently with real-time streaming at 25-30 ms. Our model is based on CSPDarkNet-53, which was trained on the MS COCO dataset for image classification. It includes additional layers to capture feature maps from different phases. Additionally, the model's neck is made up of PANet, which is used to aggregate the parameters from various CSPDarkNet-53 layers. The CSPDarkNet-53's 53 convolutional layers are followed by 53 more layers in the model head, for a total of 106 completely convolutional layers in the design. This architecture is further integrated with YOLOv3, resulting in the YOLOv4 model, which will be used by our detection model. Furthermore, to differentiate humans The aforementioned method was used to evaluate drone footage and count social distance violations in real time. Our findings show that our model was reliable and successful at detecting social distance violations in real-time with an average accuracy of 82%.

PubMed Disclaimer

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1
Fig 1. Block diagram of the proposed model.
Fig 2
Fig 2. The ground and body coordinate systems.
Fig 3
Fig 3. Euler angles.
Fig 4
Fig 4. Directions of rotation of the motors.
Fig 5
Fig 5. Schematic diagram of the quadcopter.
Fig 6
Fig 6. Images of the quadcopter from different angles.
Fig 7
Fig 7. Drone in flight for data collection.
Fig 8
Fig 8. Flowchart for our proposed model.
Here, SDV = Number of Social Distance Violations, SDM = Number of Social Distance Maintained.
Fig 9
Fig 9. Model architecture of YOLOv3.
Fig 10
Fig 10. Model architecture of YOLOv4.
Fig 11
Fig 11. Bounding box with location priors and location prediction.
Fig 12
Fig 12. Process for determining the focal length (f) of the camera.
Fig 13
Fig 13. Process of estimating distance (d) of ROIs from the camera.
Fig 14
Fig 14. (a–c) Benchmarking the distance calibration (Courtesy to the physics laboratory of East West University).
Fig 15
Fig 15. Raw frames before object detection (a—d).
Fig 16
Fig 16. Processed frames with detected ROIs (a—d).
Fig 17
Fig 17. Images after depth calculation (a—d).
Fig 18
Fig 18. Processed images displaying the count of social distance violations (a—d).
Fig 19
Fig 19. Histogram of accuracy for detecting ROIs (From Video-1).
Fig 20
Fig 20. Histogram of accuracy for detecting social distance violations (From Video-1).
Fig 21
Fig 21. Histogram of accuracy for detecting ROIs (From Video-2).
Fig 22
Fig 22. Histogram of accuracy for detecting social distance violations (From Video-2).
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Lin TY, Maire M, Belongie S, Hays J, Perona P, Ramanan D, et al. Microsoft coco: Common objects in context. InEuropean conference on computer vision 2014 Sep 6 (pp. 740–755). Springer, Cham.
    1. He K, Zhang X, Ren S, Sun J. Spatial pyramid pooling in deep convolutional networks for visual recognition. IEEE transactions on pattern analysis and machine intelligence. 2015. Jan 9;37(9):1904–16. doi: 10.1109/TPAMI.2015.2389824 - DOI - PubMed
    1. Redmon J, Farhadi A. YOLO9000: better, faster, stronger. InProceedings of the IEEE conference on computer vision and pattern recognition 2017 (pp. 7263–7271).
    1. Wang CY, Liao HY, Wu YH, Chen PY, Hsieh JW, Yeh IH. CSPNet: A new backbone that can enhance learning capability of CNN. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops 2020 (pp. 390–391).
    1. Meivel S, Sindhwani N, Anand R, Pandey D, Alnuaim AA, Altheneyan AS, et al. Mask Detection and Social Distance Identification Using Internet of Things and Faster R-CNN Algorithm. Computational Intelligence and Neuroscience. 2022. Feb 1;2022. doi: 10.1155/2022/2103975 - DOI - PMC - PubMed