Smart AntiTheft System with Face Recognition

Abstract

The increasing rise in unauthorized access, theft, and physical tampering has created a strong demand for smarter and more secure access-control systems. Conventional locking methods such as mechanical locks and RFID-based systems are no longer fully reliable, as keys and cards can be lost, stolen, or duplicated. To address this issue, this project presents a Smart Anti-Theft System that combines Raspberry Pi Pico, TensorFlow Lite, OpenCV, and Firebase Cloud to build a low-cost, intelligent, and low-latency security solution. The proposed system performs real-time facial recognition locally on the device using edge AI, which improves privacy and reduces response time. A quantized CNN model in TensorFlow Lite processes captured image frames and makes unlock decisions in less than 500 milliseconds. When an authorized face is detected, the system activates a solenoid lock to grant access. If an unknown or unauthorized person is detected, the system immediately generates alerts and synchronizes logs to Firebase Cloud for remote monitoring. An Android companion application further enhances the system by providing features such as face enrollment, real-time activity tracking, remote lock control, and NLP-based interaction for smarter user communication. The integration of local AI processing with cloud-based monitoring makes the system both secure and practical. Overall, the proposed system offers a privacy-preserving, modular, and efficient anti-theft solution suitable for homes, offices, laboratories, and restricted-access areas.

Introduction

The project describes the development of a Smart Anti-Theft System that uses face recognition and edge AI to replace traditional security methods like keys, passwords, and RFID cards, which are vulnerable to theft, duplication, and misuse. It is built using a Raspberry Pi Pico, camera module, TensorFlow Lite, OpenCV, Firebase Cloud, and an Android application to enable intelligent, real-time access control.

The system uses facial recognition as the main authentication method, where a lightweight CNN model processes face data locally (edge AI) to reduce delay, improve privacy, and ensure fast decision-making. A solenoid lock is used for physical access control, granting or denying entry based on recognition results. Firebase Cloud is integrated for storing alerts, logs, and event data, while the Android app supports face enrollment, monitoring, and remote control.

The project is currently about 50–60% complete, with core modules designed and partially implemented. The methodology includes image capture, face detection, preprocessing, face recognition, decision-making, lock control, and cloud synchronization.

Testing shows that the system correctly identifies authorized users, blocks unauthorized access, triggers alerts, and logs events in the cloud. It performs well in normal conditions and maintains local functionality even without internet connectivity. However, performance can be affected by poor lighting, face angle, and image quality.

Conclusion

The proposed Smart Anti-Theft System shows that a modern access-control solution can be built by combining facial recognition, edge AI, embedded hardware control, and cloud connectivity. Instead of depending on traditional keys, passwords, or RFID cards, the system uses face-based authentication to provide a more secure and user-friendly method of entry. This directly reduces the risk of duplication, theft, and unauthorized use associated with conventional security methods. The project also demonstrates the practical importance of local processing using edge AI. By performing recognition near the device, the system can make faster decisions, reduce delay, and improve privacy because sensitive facial data does not need to be continuously sent to remote servers. At the same time, the integration of Firebase Cloud and Android monitoring adds remote visibility, alerting, and event logging, making the overall solution more intelligent and useful for real-world deployment. These design choices align well with current research trends in facial-recognition access control, IoT security, and privacy-aware biometric systems. Although the project is still under development, the current framework proves that the proposed model is technically feasible, scalable, and suitable for applications such as homes, offices, laboratories, and restricted areas. With further improvement in dataset quality, liveness detection, model optimization, and full hardware-software integration, the system can become a reliable smart security product. In conclusion, this work provides a strong foundation for a low-cost, AI-based anti-theft system that is secure, modular, and future-ready.

References

[1] R. Reaño, J. Carrión, and J. Mansilla, “Access Control Using Facial Recognition with Neural Networks for Restricted Zones,” in Proc. 19th Int. Conf. Web Information Systems and Technologies (WEBIST), 2023. [2] B. V. Satish Babu, J. Manoranjini, R. Changala, et al., “Biometric-Based Access Control Systems with Robust Facial Recognition in IoT Environments,” in 2024 3rd Int. Conf. Intelligent Techniques in Control, Optimization and Signal Processing (INCOS), 2024, doi: 10.1109/INCOS59338.2024.10527499. [3] M. Marimuthu, G. Mohanraj, J. Akilandeswari, and V. Sathiyapriya, “Facial Recognition Enabled Smart Security Lock System Using Machine Learning Approach,” EAI Endorsed Transactions on Internet of Things, vol. 11, 2025, doi: 10.4108/eetiot.5657. [4] S. S. F. A. Elnozahy, A. A. Elhady, K. M. Hosny, and M. M. Darwish, “Raspberry Pi-Based Face Recognition Door Lock System,” IoT, vol. 6, no. 2, 2025. E. Abdullahu, S. Wache, and M. Piangerelli, “Secure and Decentralized Hybrid Multi-Face Recognition for IoT Applications,” Sensors, vol. 25, no. 18, 2025, Art. no. 5880. [5] J. Guo, Y. Mu, Z. Liu, X. Ren, and H. Han, “Federated Learning for Biometric Recognition: A Survey,” Artificial Intelligence Review, vol. 57, 2024, Art. no. 208. [6] A. S. Mohammad, “IoT-MFaceNet: Internet-of-Things-Based Face Recognition Using MobileNetV2 and FaceNet Deep-Learning Implementations on a Raspberry Pi-400,” Journal of Low Power Electronics and Applications, vol. 14, no. 3, 2024. [7] Y. Li, Y. Li, and Z. Wang, “FaceCloseup: Enhancing Mobile Facial Authentication with Perspective Distortion-Based Liveness Detection,” Computers, vol. 14, 2025, Art. no. 254. [8] M. Kira, Z. Alajamy, A. Soliman, and M. Mazzara, “Trustworthy Face Recognition as a Service: A Multi-Layered Approach for Mitigating Spoofing and Ensuring System Integrity,” Future Internet, vol. 17, 2025, Art. no. 450. [9] E. L. Birgisdóttir, M. I. Kunkel, et al., “Exploring the Security of Mobile Face Recognition: Attacks, Defenses, and Future Directions,” Applied Sciences, vol. 15, 2025, Art. no. 13232. [10] H. Xing, et al., “Face Anti-Spoofing Based on Deep Learning,” Applied Sciences, vol. 15, 2025, Art. no. 6891. [11] A. Nurpeisova, et al., “Deep Residual Learning for Face Anti-Spoofing,” Technologies, vol. 13, no. 9, 2025, Art. no. 413. [12] A. Baran, et al., “Face the Challenge—Generalization of Presentation Attack Detection in Face Biometrics,” Sensors, vol. 25, no. 18, 2025, Art. no. 5792. [13] S. Brindha, D. M. Bharath, S. R. Gokulprasath, and M. Ravisrinivasan, “Industrial Attendance and Access Control System Using Face and Biometric Recognition,” in Proc. SCITEPRESS, 2025. [14] J. Silva, “Enhancing Facial Recognition While Protecting User Data,” in Proc. SCITEPRESS, 2025. [15] R. Pandurangan, “Enhanced Face Recognition Algorithm for Real-Time Applications,” in Proc. SCITEPRESS, 2025. [16] G. P. Priyadarshini, “Facial Recognition and Feature Mapping with Machine Learning,” in Proc. SCITEPRESS, 2025. [17] R. Karman, I. Indrabayu, and I. P. W. Kusuma, “Operational Limits of Near-Infrared Face Recognition: The Critical Impact of Distance on Identification Accuracy,” in Proc. SCITEPRESS, 2025. [18] J. Al-Nabulsi, et al., “IoT Solutions and AI-Based Frameworks for Masked-Face Recognition: A Review,” Sensors, vol. 23, no. 16, 2023, Art. no. 7193. [19] J. Saleem, et al., “Machine Learning-Enhanced Attribute-Based Authentication for Smart IoT Environments,” Sensors, vol. 25, no. 9, 2025, Art. no. 2779.

Copyright

Copyright © 2026 Prof. Satish C. Cholke, Dhonnar Dipali, Mahale Hemangi , Khalkar Vaishnavi , Salve Harshada . 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.