AI Driven ATM Premises using Raspberry pi Technology

Abstract

The project focuses on developing a smart ATM that leverages biometric security features like retina scanning, fingerprint recognition, and facial recognition to authenticate users instead of relying on traditional PINs. These advanced technologies aim to enhance transaction security, minimize fraud, and ensure that only the rightful account holder can access their funds. By integrating these features, the system intends to make banking safer, more efficient, and highly convenient for users. It emphasizes improving the reliability and ease of cash withdrawals while reducing the risk of unauthorized access. The innovative use of biometrics represents a shift towards modern, secure banking solutions. Overall, the project aspires to redefine the way individuals interact with ATMs, prioritizing safety anduserfriendliness.

Introduction

The ATM Examining System is an AI-driven, real-time monitoring solution designed to improve the security, maintenance, and operational efficiency of ATMs. It employs IoT sensors, surveillance cameras, and AI models—such as facial recognition and object detection (YOLOv8)—to continuously track ATM functionality and detect suspicious activities, especially fraud attempts involving identity concealment using masks or helmets.

Key features include helmet and mask detection to enforce security protocols, behavior analysis to identify unusual actions, and instant alerts sent to bank officials via SMS, email, or apps for rapid response. The system uses edge computing devices like Raspberry Pi 4B, connected to IP cameras powered via PoE, enabling efficient local image processing with minimal latency. A centralized dashboard provides administrators with real-time ATM status and security monitoring.

The system ensures only one user accesses the ATM at a time and prompts users to remove helmets or masks for proper identification. It logs security events for later review and generates reports on machine performance and incidents. This integration of AI, IoT, and edge computing enhances fraud prevention, maintenance workflows, and overall ATM safety, making it a cost-effective and robust tool for modern banking environments.

Conclusion

AI-driven ATM premises powered by Raspberry Pi offer numerous practical benefits beyond traditional surveillance systems. With the integration of computer vision and machine learning, the system can be trained to detect specific events such as loitering, vandalism, or even the absence of required safety gear like helmets or masks—ensuring compliance with safety protocols in certain environments. Additionally, features like facial recognition and motion tracking can be implemented to enhance identity verification and track customer interactions securely. The Raspberry Pi, despite its compact form factor, proves to be a highly capable platform for running lightweight AI models and interfacing with various sensors and cameras, making it a robust and versatile solution. This innovation not only supports cost-effective deployment but also encourages the adoption of smart technologies in sectors where high-end computing infrastructure may not be feasible. As digital banking continues to evolve, AI-based ATM premises using Raspberry Pi stand as a scalable and intelligent solution for modernizing ATM infrastructure while significantly improving safety, surveillance, and customer experience.

References

[1] Degadwala, Sheshang, et al. “Advancements in ATM Security for Movement and Tampering Detection: A Review.” International Journal of Scientific Research in Computer Science, Engineering and Information Technology, 2024. Available at: https://doi.org/10.32628/CSEIT2410587. [2] Sanserwal, Vishal, et al. “Comparative Analysis of Various Feature Descriptors for Efficient ATM Surveillance Framework.” Proceeding - IEEE International Conference on Computing, Communication and Automation, ICCCA 2017, 2017, pp. 539 – 44. Available at: https://doi.org/10.1109/CCAA.2017.8229860. [3] Liu, Fan, et al. “Abnormal Behavior Recognition System for ATM Monitoring by RGB-D Camera.” MM 2012 - Proceedings of the 20th ACM International Conference on Multimedia, 2012, pp. 1295 – 96. Available at: https://doi.org/10.1145/2393347.2396450. [4] Wu J-D, Ye S-H (2009) Driver identification using finger-vein patterns with Radon transform and neural network. Expert Syst Appl 36(3, Part 2):5793–5799. [5] Researcher Wu and Liu have described neural network and component analysis methodology used in finger vein authentication method. [6] Dr. Deepak B. Kadam, Dr. Madhukar S. Chavan, Mr. Shital A. Patil, \"Security System using Raspberry Pi,\" International Journal for Research in Applied Science & Engineering Technology (IJRA), Volume 7 Issue VI, June 2019, available at www.ijraset.com. [7] Huu-Quoc Nguyen, Ton Thi Kim Loan, Bui Dinh Mao, and Eui-Nam Huh, \"Low cost real-time system monitoring using Raspberry Pi,\" 2015 Seventh International Conference on Ubiquitous and Future Networks, Sapporo, 2015, pp. 857-859. [8] V. Menezes, V. Patchava, and M. S. D. Gupta, \"Surveillance and monitoring system using Raspberry Pi and SimpleCV,\" 2015 International Conference on Green Computing and Internet of Things (ICGCIoT), Noida, 2015, pp. 1276-1278. [9] A. Arora, P. Dutta, S. Bapat, V. Kulathumani, H. Zhang, V. Naik, V. Mittal, H. Cao, M. Demirbas, M. Gouda, Y. Choi, T. Herman, S. Kulkarni, U. Arumugam, M. Nesterenko, A. Vora, M. Miyashita, “A line in the sand: a wireless sensor network for target detection, classification, and tracking,” Computer Networks: The International Journal of Computer and Telecommunications Networking, v.46 n.5, p.605-634.

Copyright

Copyright © 2025 Sandeep AS, B Gowrish, B Ramakoti Reddy, Basavaraja P M, Mrs. Gouri D Malgi. 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.