An Integrated IoT-Based Smart Home System for Safety and Automation

Abstract

This paper presents the design and implementation of a unified smart home system that combines facial recognition for access control, hazardous gas detection, and IoT-based appliance automation. The system employs an ESP32-S3 CAM microcontroller with a pre-trained DeepStack model for real-time facial identification, MQ-series gas sensors for gas detection, and a mobile application developed using MIT App Inventor for remote control of appliances. Google Sheets functions as a lightweight cloud backend for data logging and system monitoring. Experimental results confirmed the system’s high reliability under various lighting conditions, sub-second switching latency for appliances, and reliable gas detection at multiple distances. The results demonstrate that the proposed system provides a low-cost, modular, and scalable framework suitable for enhancing residential safety and convenience.