EagleAI: AI Powered Real Time Restricted Area Surveillance

Abstract

Traditional security measures often fail to provide accurate and efficient monitoring, leading to vulnerabilities in restricted areas. This project implements an advanced real-time facial recognition-based surveillance system to enhance security and address these shortcomings. The system uses facial recognition technology for live monitoring, identifying authorized personnel and detecting unauthorized individuals. Attendance is recorded automatically in an Excel file, with timestamps for entry and exit. For unauthorized access, the system triggers multiple actions, including capturing and saving the individual\'s image, uploading it to Azure Blob Storage, sending email alerts with timestamps and image URLs, and activating a buzzer alarm alongside an on-screen warning. By integrating automated facial recognition with real-time alerts and cloud storage, the system significantly improves surveillance accuracy and efficiency, providing a robust solution for preventing unauthorized access in restricted areas.

Introduction

The project aims to enhance security and access control in sensitive areas by replacing traditional manual monitoring with an intelligent, automated surveillance system called EagleAI. This system uses AI-driven facial recognition to accurately identify authorized personnel, detect unauthorized access in real-time, and send immediate alerts to administrators. Beyond security, it also automates attendance tracking, improving operational efficiency in settings like schools, offices, and industries.

Objectives:

• Implement real-time, continuous automated monitoring without manual intervention.

• Precisely classify authorized individuals to maintain secure access.

• Trigger instant alerts on detecting unauthorized persons.

• Notify administrators promptly to facilitate rapid response.

Related Work:
Prior studies explore face detection and recognition using OpenCV, deep learning algorithms, and cloud-based machine learning, highlighting trade-offs between accuracy, latency, cost, and scalability.

Proposed Method:
The system integrates AI-powered face recognition with automated alerting, combining real-time surveillance, face encoding comparison, breach detection, and administrator notification via email and audio-visual warnings.

Methodology:

• Phase 1: Record and encode facial features of authorized personnel.

• Phase 2: Train models and monitor live video feeds for face recognition.

• On detecting unknown individuals, issue warnings, capture images, upload evidence to cloud storage, and send alerts.

• Automate attendance logging for authorized personnel.

System Design:

• Use of libraries such as MTCNN for face detection, Albumentations for data augmentation, and openpyxl for attendance logging.

• Cloud integration via Azure Blob Storage for secure image storage and remote access.

• Real-time operation until manually stopped.

Use Case & Sequence Diagrams:
Describe interactions between admin and system for face input, recognition, alert generation, and attendance marking.

Results:
Demonstrated effective security enforcement, accurate attendance management, and real-time unauthorized access detection with automated alerts, enhancing protection in restricted areas.

Conclusion

The project\'s motivation is to strengthen security measures and ensure effective access management, thereby reducing vulnerabilities in sensitive areas. Traditional approaches, which rely on manual checks and static monitoring systems, often lead to inefficiencies, delays, and errors. These shortcomings can compromise the safety of personnel, data, and infrastructure. This project aims to address these issues by introducing an intelligent system that automates monitoring, reduces manual dependency, and provides real-time responses to potential threats.

References

[1] R. T. H. Hasan and A. B. Sallow, “Face Detection and Recognition Using OpenCV,” Journal of Soft Computing and Data Mining, vol. 2, no. 2, pp. 86–97, Oct. 2021, doi: 10.30880/jscdm.2021.02.02.008. [2] D. Sattibabu, S. Yacoob, P. Tanuja, B. M. Koushika, K. V. Babu, and Ch. J. Kumar, “Intra Student Surveillance System: Detection And Identifying Unauthorized Wandering,” 2024, pp. 1201–1208. doi: 10.2991/978-94-6463-471-6_115. [3] A. S. Bein and A. Williams, “Development of Deep Learning Algorithms for Improved Facial Recognition in Security Applications,” IAIC Transactions on Sustainable Digital Innovation (ITSDI), vol. 5, no. 1, pp. 19–23, 2023. [4] K. Virendra, M. Kumar Bagwani, V. Kumar Tiwari, D. Kumar Chouhan, and A. Jain, “Optimizing Face Detection Performance with Cloud Machine Learning Services,” 2024.

Copyright

Copyright © 2025 Jahith Sai Krishna Varma, K Pallavi, Darshan R, K Prashanth Kumar, Dr. Phanindra Reddy K. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.