IoT-Based and Teachable Machine Platform for Covid-19 Prevention and Control

Siti Murni Hanum Md Ariff; Norlela Ishak; Mazidah Tajjudin

doi:10.37934/araset.56.1.163182

Authors

Siti Murni Hanum Md Ariff School of Electrical Engineering, College of Engineering, Universiti Teknologi Mara, 40450 Shah Alam, Selangor, Malaysia
Norlela Ishak School of Electrical Engineering, College of Engineering, Universiti Teknologi Mara, 40450 Shah Alam, Selangor, Malaysia
Mazidah Tajjudin School of Electrical Engineering, College of Engineering, Universiti Teknologi Mara, 40450 Shah Alam, Selangor, Malaysia

DOI:

https://doi.org/10.37934/araset.56.1.163182

Keywords:

Face mask detection, Teachable machine, Infrared sensor, MLX90614, Blynk, ESP32-CAM, Ultrasonic sensor, HC-SR04

Abstract

This paper aims to develop a solution for premises requiring compliance with Standard Operating Procedures (SOP) for COVID-19 safety. To address potential outbreaks and human errors in manual entrance supervision, a face mask detection model is developed using Teachable Machine in which a dataset of 4092 images is used to train the model. This system utilizes an ultrasonic sensor (HC-SR04) to gauge the distance and a non-contact infrared temperature sensor (MLX90614) for body temperature screening. Data is displayed on an I2C LCD and published on an IoT platform (Blynk's dashboard) for remote monitoring. A healthy body temperature (36.1℃ to 38℃) triggers the face mask detection model. If the person wears a mask, a green LED lights up, and the barrier opens for entry; otherwise, a red LED indicates no mask, and entry is denied. Notification is also prompted upon the detection of an unhealthy body temperature. The system achieves the following; 98.43% accuracy for body temperature scanning, 98% for face mask detection, and 97.5% success rate during the implementation of the face mask detection system.