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
. 2021 Nov 8:2021:4529107.
doi: 10.1155/2021/4529107. eCollection 2021.

SAI-YOLO: A Lightweight Network for Real-Time Detection of Driver Mask-Wearing Specification on Resource-Constrained Devices

Zuopeng Zhao  1   2 , Kai Hao  1   2 , Xiaoping Ma  1   2 , Xiaofeng Liu  1   2 , Tianci Zheng  1   2 , Junjie Xu  1   2 , Shuya Cui  1   2
Affiliations

SAI-YOLO: A Lightweight Network for Real-Time Detection of Driver Mask-Wearing Specification on Resource-Constrained Devices

Zuopeng Zhao et al. Comput Intell Neurosci. .

Abstract

Frequent occurrence and long-term existence of respiratory diseases such as COVID-19 and influenza require bus drivers to wear masks correctly during driving. To quickly detect whether the mask is worn correctly on resource-constrained devices, a lightweight target detection network SAI-YOLO is proposed. Based on YOLOv4-Tiny, the network incorporates the Inception V3 structure, replaces two CSPBlock modules with the RES-SEBlock modules to reduce the number of parameters and computational difficulty, and adds a convolutional block attention module and a squeeze-and-excitation module to extract key feature information. Moreover, a modified ReLU (M-ReLU) activation function is introduced to replace the original Leaky_ReLU function. The experimental results show that SAI-YOLO reduces the number of network parameters and calculation difficulty and improves the detection speed of the network while maintaining certain recognition accuracy. The mean average precision (mAP) for face-mask-wearing detection reaches 86% and the average precision (AP) for mask-wearing normative detection reaches 88%. In the resource-constrained device Raspberry Pi 4B, the average detection time after acceleration is 197 ms, which meets the actual application requirements.

PubMed Disclaimer

Conflict of interest statement

The authors declare that there are no conflicts of interest regarding the publication of this study.

Figures

Figure 1
Figure 1
CSPBlock structure.
Figure 2
Figure 2
Structure of base convolutional layer.
Figure 3
Figure 3
Neck structure.
Figure 4
Figure 4
Attention module. (a) Channel attention. (b) Spatial attention. (c) Attention module.
Figure 5
Figure 5
Structure of RES-SEBlock. (a) RES-SEBlock. (b) SE-Block.
Figure 6
Figure 6
The proposed SAI-YOLO network structure.
Figure 7
Figure 7
Masked_Imgs dataset.
Figure 8
Figure 8
Effect comparisons on MAFA dataset: (a) SAI-YOLO and (b) YOLOv4-Tiny.
Figure 9
Figure 9
Average precision of the SAI-YOLO in Masked_Imgs dataset.
Figure 10
Figure 10
Detection results on Masked_Imgs.
Figure 11
Figure 11
Illumination test. (a) Strong light exposure and (b) poor light exposure.
Figure 12
Figure 12
Different illumination test results.
Figure 13
Figure 13
Occlusion test.
Figure 14
Figure 14
Detection test on Raspberry Pi 4B.
See this image and copyright information in PMC

References

    1. Rahmani A., Mirmahaleh S. Y. H. Coronavirus disease (COVID-19) prevention and treatment methods and effective parameters: a systematic literature review. Sustainable Cities and Society . 2020;64 doi: 10.1016/j.scs.2020.102568.102568 - DOI - PMC - PubMed
    1. Shorfuzzaman M., Hossain M. S., Alhamid M. F. Towards the sustainable development of smart cities through mass video surveillance: a response to the COVID-19 pandemic. Sustainable Cities and Society . 64(2021) doi: 10.1016/j.scs.2020.102582.102582 - DOI - PMC - PubMed
    1. Matuschek C., Moll F., Fangerau H., et al. Face masks: benefits and risks during the COVID-19 crisis. European Journal of Medical Research . 2020;25(32) doi: 10.1186/s40001-020-00430-5. - DOI - PMC - PubMed
    1. Zhao Z., Zhang Z., Xu X., Xu Y., Yan H., Zhang L. A lightweight object detection network for real-time detection of driver handheld call on embedded devices. Computational Intelligence and Neuroscience . 2020;2020:12. doi: 10.1155/2020/6616584.6616584 - DOI - PMC - PubMed
    1. Hubel D. H., Wiesel T. N. Early exploration of the visual cortex. Neuron . 1998;20(3):401–412. doi: 10.1016/s0896-6273(00)80984-8. - DOI - PubMed