Using IoT and YOLOv5 for real-time security monitoring of homes

Abstract Everyone wants their home to be as secure as possible against disturbances and threats. Many people have installed Closed Circuit Television (CCTV) as a home security system. Apart from the fact that it is relatively expensive to install, CCTV has shortcomings that make it difficult to integrate with stable, secure user devices in real time. The aim of this work is to develop a security system capable of detecting malicious people in a house using a Raspberry PI 3, infrared (PIR) and ultrasound sensors, a webcam for taking photos and YOLOv5 for detecting objects. Once the malicious person is detected, the homeowner is immediately informed of the status of their home on their mobile phone via the Telegram mobile application. This proposed system can be applied and used as a safe and very inexpensive means of prevention.