Real-Time Face Mask Detection Method Based on YOLOv3
The rapid outbreak of COVID-19 has caused serious harm and infected tens of millions of people worldwide. Since there is no specific treatment, wearing masks has become an effective method to prevent the transmission of COVID-19 and is required in most public areas, which has also led to a growing d...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-04-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/10/7/837 |
| _version_ | 1797539170329034752 |
|---|---|
| author | Xinbei Jiang Tianhan Gao Zichen Zhu Yukang Zhao |
| author_facet | Xinbei Jiang Tianhan Gao Zichen Zhu Yukang Zhao |
| author_sort | Xinbei Jiang |
| collection | DOAJ |
| description | The rapid outbreak of COVID-19 has caused serious harm and infected tens of millions of people worldwide. Since there is no specific treatment, wearing masks has become an effective method to prevent the transmission of COVID-19 and is required in most public areas, which has also led to a growing demand for automatic real-time mask detection services to replace manual reminding. However, few studies on face mask detection are being conducted. It is urgent to improve the performance of mask detectors. In this paper, we proposed the Properly Wearing Masked Face Detection Dataset (PWMFD), which included 9205 images of mask wearing samples with three categories. Moreover, we proposed Squeeze and Excitation (SE)-YOLOv3, a mask detector with relatively balanced effectiveness and efficiency. We integrated the attention mechanism by introducing the SE block into Darknet53 to obtain the relationships among channels so that the network can focus more on the important feature. We adopted GIoUloss, which can better describe the spatial difference between predicted and ground truth boxes to improve the stability of bounding box regression. Focal loss was utilized for solving the extreme foreground-background class imbalance. Besides, we performed corresponding image augmentation techniques to further improve the robustness of the model on the specific task. Experimental results showed that SE-YOLOv3 outperformed YOLOv3 and other state-of-the-art detectors on PWMFD and achieved a higher 8.6% mAP compared to YOLOv3 while having a comparable detection speed. |
| first_indexed | 2024-03-10T12:42:25Z |
| format | Article |
| id | doaj.art-4171678ee9964969ae5db19fc77e4e49 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-10T12:42:25Z |
| publishDate | 2021-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-4171678ee9964969ae5db19fc77e4e492023-11-21T13:46:34ZengMDPI AGElectronics2079-92922021-04-0110783710.3390/electronics10070837Real-Time Face Mask Detection Method Based on YOLOv3Xinbei Jiang0Tianhan Gao1Zichen Zhu2Yukang Zhao3Software College, Northeastern University, Shenyang 110004, ChinaSoftware College, Northeastern University, Shenyang 110004, ChinaSoftware College, Northeastern University, Shenyang 110004, ChinaSchool of Computer Science and Engineering, Northeastern University, Shenyang 110004, ChinaThe rapid outbreak of COVID-19 has caused serious harm and infected tens of millions of people worldwide. Since there is no specific treatment, wearing masks has become an effective method to prevent the transmission of COVID-19 and is required in most public areas, which has also led to a growing demand for automatic real-time mask detection services to replace manual reminding. However, few studies on face mask detection are being conducted. It is urgent to improve the performance of mask detectors. In this paper, we proposed the Properly Wearing Masked Face Detection Dataset (PWMFD), which included 9205 images of mask wearing samples with three categories. Moreover, we proposed Squeeze and Excitation (SE)-YOLOv3, a mask detector with relatively balanced effectiveness and efficiency. We integrated the attention mechanism by introducing the SE block into Darknet53 to obtain the relationships among channels so that the network can focus more on the important feature. We adopted GIoUloss, which can better describe the spatial difference between predicted and ground truth boxes to improve the stability of bounding box regression. Focal loss was utilized for solving the extreme foreground-background class imbalance. Besides, we performed corresponding image augmentation techniques to further improve the robustness of the model on the specific task. Experimental results showed that SE-YOLOv3 outperformed YOLOv3 and other state-of-the-art detectors on PWMFD and achieved a higher 8.6% mAP compared to YOLOv3 while having a comparable detection speed.https://www.mdpi.com/2079-9292/10/7/837computer visionCOVID-19deep learningface mask detectionYOLOv3 |
| spellingShingle | Xinbei Jiang Tianhan Gao Zichen Zhu Yukang Zhao Real-Time Face Mask Detection Method Based on YOLOv3 Electronics computer vision COVID-19 deep learning face mask detection YOLOv3 |
| title | Real-Time Face Mask Detection Method Based on YOLOv3 |
| title_full | Real-Time Face Mask Detection Method Based on YOLOv3 |
| title_fullStr | Real-Time Face Mask Detection Method Based on YOLOv3 |
| title_full_unstemmed | Real-Time Face Mask Detection Method Based on YOLOv3 |
| title_short | Real-Time Face Mask Detection Method Based on YOLOv3 |
| title_sort | real time face mask detection method based on yolov3 |
| topic | computer vision COVID-19 deep learning face mask detection YOLOv3 |
| url | https://www.mdpi.com/2079-9292/10/7/837 |
| work_keys_str_mv | AT xinbeijiang realtimefacemaskdetectionmethodbasedonyolov3 AT tianhangao realtimefacemaskdetectionmethodbasedonyolov3 AT zichenzhu realtimefacemaskdetectionmethodbasedonyolov3 AT yukangzhao realtimefacemaskdetectionmethodbasedonyolov3 |