Manuscript received September 28, 2022; revised November 23, 2022; accepted May 9, 2023.
Abstract—The rise in crime rates has significantly proven the need for security in houses. Although various types of surveillance systems are already being used, most make use of single modes of detection which are insufficient to identify intrusions. This paper presents an affordable IoT based home security system that combines different ways to detect intruders. It has been implemented using a Raspberry Pi 4 device, an IP camera, a microphone, and sound and motion sensors. When people enter a monitored area, they are detected either through their movement, voice, or while knocking on the door. The number of individuals present is counted, followed by face recognition. The system can also recognise masked faces. The system uses a servo motor as a physical door lock, which is unlocked for authorized users. A speaker identification module is responsible for authenticating the voice of a person via the camera whenever the latter is heard but not seen. When a potential intrusion is detected, the footage is uploaded to the cloud and event details are transmitted to a cloud database. Owners are also immediately notified through messages and can access all the intrusion details on a web application.
Index Terms—Security, object detection, face recognition, masked face recognition, speaker identification, smart lock, immediate notifications
The authors are with the Information and Communication Technology Department, Faculty of Information, Communication and Digital Technologies, University of Mauritius, Mauritius.
*Correspondence: s.pudaruth@uom.ac.mu (S.P.)
[PDF]
Cite:Bhavesh Bhuckory and Sameerchand Pudaruth, "A Multimodal IoT-Based Home Intrusion Detection System," International Journal of Computer Theory and Engineering vol. 15, no. 3, pp. 117-124, 2023.
Copyright © 2023 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).
