Authors: Assistant Professor G.Chandram, Manasa Sai
Abstract: The unauthorized use of mobile phones within restricted environments such as examination halls, government offices, hospitals, research laboratories, and other secure facilities presents serious concerns related to security, privacy, operational efficiency, and regulatory compliance. Conventional mobile phone detection techniques primarily rely on monitoring radio frequency or network signals, making them ineffective when devices operate in airplane mode, remain disconnected from communication networks, or employ signal suppression mechanisms. To overcome these limitations, this paper presents Phone Patrol, an intelligent image-based surveillance framework developed for detecting unauthorized mobile phone usage in mobile-free environments. The proposed system integrates You Only Look Once version 5 (YOLOv5) for real-time mobile phone detection and You Only Look Once version 9 (YOLOv9) for face recognition, enabling accurate identification of individuals violating mobile usage policies. The framework is deployed on a Raspberry Pi 5, providing a compact and cost-effective Internet of Things (IoT) solution capable of processing live video streams and automatically issuing SMS notifications through the Twilio API whenever a violation is detected. Experimental evaluation demonstrates excellent detection performance, with the YOLOv5 model achieving a mean Average Precision (mAP50) of 0.992, confirming its capability to accurately recognize mobile phone usage under real-time operating conditions. Although the current implementation is intended as a prototype and requires further optimization for large-scale deployment, the proposed framework demonstrates the effectiveness of combining deep learning, edge computing, and automated notification services to enforce mobile-free policies. The developed system provides a reliable, network-independent surveillance solution that strengthens security, improves policy compliance, and supports intelligent monitoring in restricted environments.
