Volume 15, Issue 6

Safe Hostel: A Smart Security System for Girls Hostel Using AI and IoT

Author

Aparajita Sinha, Shraddha Sunil Kadam, Shraddha Vasant Bhor, Sonali Ashok Dhone

Abstract

Traditional hostel security systems are predominantly manual, reactive, and prone to errors, making it difficult to ensure reliable safety and prevent unauthorized access. To overcome these limitations, this paper presents "Safe Hostel," an intelligent loT-based security system designed for real-time monitoring and automated decision-making in hostel environments. The proposed system employs a dual-layer architecture consisting of' an loT-based sensing layer for motion detection and data acquisition, and an AI-based processing layer for face recognition and activity classification. The system integrates Raspberry Pi 4B, ESP32, ESP32-CAM module, Raspberry Pi Camera, USB Microphone, and a 32GB SD card for efficient surveillance, communication, and secure data storage. The ESP32 module manages device coordination and wireless communication, while the Raspberry Pi 4B performs real-time face recognition and activity classification using MediaPipe and dlib libraries. The system generates 128-dimensional facial embeddings for accurate identity verification and classifies activities into five categories: Authorized Person, Unauthorized Person, Suspicious Activity, Normal Activity, and Emergency Situation. Experimental results demonstrate an overall accuracy of' 94%, with high precision and recall for authorized user detection. The proposed system is cost-effective, scalable, and suitable for real-time edge-based deployment in smart hostel security applications.

Keywords: Hostel Security, Internet of' Things (loT), Artificial Intelligence (Al), Face Recognition, Raspberry Pi, ESP32, Activity Classification, Smart Surveillance

DOI: https://doi.org/10.62226/ijarst20262716

 

DOI

https://doi.org/10.62226/ijarst20262716

PAGES : 2158-2162 | 8 VIEWS | 4 DOWNLOADS

Download Full Article