Citation

Yunos, Nur Syahidah (2025) Containerised edge framework for real-time health monitoring. Masters thesis, Multimedia University.

Full text not available from this repository.

Abstract

The COVID-19 pandemic has emphasized the need for efficient health screening solutions in public spaces, especially schools. This research proposes an IoT-based health screening framework using edge computing and deep learning to automate mask detection, temperature monitoring, and attendance logging in a school environment. The system leverages a containerised architecture running on a Raspberry Pi, integrating a camera, MLX90614 thermal sensor, and RC522 RFID reader for real-time processing. Sensor data was processed locally at the edge, reducing latency and preserving privacy, with the final data uploaded to a cloud server for remote monitoring via Grafana dashboards. There were 10 CNN models analysed and out of them MobileNetV2, ResNet50 and DenseNet121 were benchmarked using transfer learning. The metrics such as training time, accuracy, F1-score, and real-time inference performance were evaluated. Based on the observation, ResNet50 achieved an accuracy of 99%, however, MobileNetV2 was selected for deployment due to its optimal balance between accuracy of 98%, fast inference, and lightweight structure which was suitable for embedded environments. The system was successfully deployed and tested, demonstrating high reliability and practical applicability for school-based health monitoring.

Item Type:	Thesis (Masters)
Additional Information:	Call No.: QA76.583 .N87 2025
Uncontrolled Keywords:	Edge computing
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75-76.95 Calculating machines
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	05 Feb 2026 08:27
Last Modified:	05 Feb 2026 08:27
URII:	http://shdl.mmu.edu.my/id/eprint/15200

Downloads

Edit (login required)