Zhang, Yuxi, Al-Ataby, Ali and Al-Naima, Fawzi
(2022)
A Deep Learning-Based Tool for Face Mask Detection and Body Temperature Measurement.
In: 2022 5th International Conference on Signal Processing and Information Security (ICSPIS), 2022-12-7 - 2022-12-8.
|
PDF
A Deep Learning-Based Tool for Detection of Face Mask and Measuring Body Temperature.pdf - Author Accepted Manuscript Download (831kB) | Preview |
Abstract
Due to the COVID-19 pandemic outbreak, wearing a mask and ensuring normal body temperature in overcrowded areas such as workplaces have become obligatory. In this paper, a deep learning-based tool for automatic mask detection and temperature measurement at the entrance of workplaces was developed to save costs of manual supervision and reduce human contact for safety concerns. Using Python, image/video processing techniques related to face and object detection are used to process image input from a webcam. A deep learning algorithm called MobileNetV2 was used to build the face mask detector model. Moreover, a non-contact thermal sensor, the MLX90614, along with Arduino, was employed to measure body temperature. The mask detection and temperature measurements are displayed correctly on a Graphical User Interface (GUI). Besides, an additional function related to the Internet of Things (IoT) was implemented, which sends high-temperature alerts to smartphones. It has been verified that the model can achieve an accuracy of about 98%. The developed system experiences a limitation when other objects are used to cover the mouth and nose in that they may still be classified as masks. However, compared to the mask detection systems available commercially, it can provide correct detection results when using the hand to pretend to be wearing a mask.
| Item Type: | Conference or Workshop Item (Unspecified) |
|---|---|
| Divisions: | Faculty of Science and Engineering > School of Electrical Engineering, Electronics and Computer Science |
| Depositing User: | Symplectic Admin |
| Date Deposited: | 22 Feb 2023 12:41 |
| Last Modified: | 14 Mar 2024 18:38 |
| DOI: | 10.1109/icspis57063.2022.10002688 |
| Related URLs: | |
| URI: | https://livrepository.liverpool.ac.uk/id/eprint/3168539 |
Dimensions
Dimensions
