IoT-Based Smart Lock with Real-Time Person Detection Using YOLOv5 and Mobile App Integration

This article presents the design and implementation of a smart electronic lock system that integrates Internet of Things (IoT) technologies, computer vision, and multifactor authentication to enhance residential security. The solution is built upon a LilyGo ESP32-S3 microcontroller with an embedded camera and leverages a YOLOv5-based person detection model for real-time monitoring. A mobile application, developed in Flutter and integrated with Firebase, enables secure user interaction, remote access control, and real-time alerts. The system combines three authentication factors: a keypad code, visual detection of authorized individuals, and mobile app verification. Experimental validation with ten participants demonstrated an average detection accuracy of 96%, outperforming comparable systems. The proposed approach stands out for its low cost, modularity, and high adaptability to smart home environments. This research contributes to the development of accessible and robust security solutions using edge AI and distributed architectures.