Development of an Edge AI-Based Employee Attendance System to Improve the Reliability of IoT Attendance Systems

Ilham Wibawa Kusumah Dinata, Yanti Yanti, Indra Setiawan, Fajar Firmansyah

Abstract

Employee attendance management is essential for improving organizational discipline and operational efficiency. Conventional attendance systems based on identity cards, fingerprint scanners, and manual recording may be affected by attendance fraud, physical contact, maintenance requirements, and communication or server interruptions. This study presents a reliability-oriented IoT employee attendance prototype integrating an ESP32-CAM, a Raspberry Pi 3 Model B+ as an edge gateway, HTTP and MQTT communication, offline buffering, Face Recognition, local data storage, and a web-based dashboard. The Raspberry Pi manages image forwarding, local buffering, MQTT communication, and synchronization, while a Flask-based server performs face detection, identity verification, attendance recording, and dashboard visualization. Face Recognition uses a Dlib-based Histogram of Oriented Gradients (HOG) detector and 128-dimensional facial embeddings with a fixed tolerance of 0.5. The initial closed-set evaluation used 212 facial samples from 11 registered employees, with all evaluated samples correctly classified under controlled conditions. The system also achieved successful HTTP transmission, database storage, dashboard synchronization, and offline-buffer recovery in the reported functional tests, with an average server response time of 1.069 s and an average MQTT delay of 2.47 ms. These results demonstrate prototype feasibility and service continuity under the tested conditions. They should not be interpreted as evidence of general biometric accuracy or large-scale deployment readiness. Additional robustness, impostor-rejection, security, privacy, and multi-user evaluations are required for broader validation.

Keywords

Edge Computing; Face Recognition; Internet of Things (IoT), MQTT Communication, Offline Buffering

Full Text:

PDF

References

[1] M. A. Azkiya Aziz, S. Ismail, and N. Allias, “Deep Learning in Face Recognition for Attendance System: An Exploratory Study,” J. Comput. Res. Innov., vol. 7, no. 2, pp. 74–81, 2022, doi: 10.24191/jcrinn.v7i2.288.

[2] J. Patel, S. Gandhi, V. Katheriya, P. Pataliya, and A. Majumdar, “Enhancing Classroom Attendance Systems with Face Recognition through CCTV using Deep Learning,” Procedia Comput. Sci., vol. 258, pp. 3031–3041, 2025, doi: 10.1016/j.procs.2025.04.561.

[3] G. M. G. Makigod, J. J. B. Sebastian, E. R. Arboleda, and L. A. T. Ongquit, “Design and Development of IoT Based Automated Attendance System Using Facial Recognition,” J. Eng. Technol. Ind. Appl., vol. 10, no. 48, pp. 136–144, Jul. 2024, doi: 10.5935/jetia.v10i48.976.

[4] Ubong David Essien and Godwin Okon Ansa, “A deep learning-based face recognition attendance system,” Glob. J. Eng. Technol. Adv., vol. 17, no. 1, pp. 009–022, Oct. 2023, doi: 10.30574/gjeta.2023.17.1.0165.

[5] R. Singh and S. S. Gill, “Edge AI: A survey,” Internet Things Cyber-Physical Syst., vol. 3, no. March, pp. 71–92, Jan. 2023, doi: 10.1016/j.iotcps.2023.02.004.

[6] L. Kong and S. Jiao, “Edge-computing-driven Internet of Things : A Survey Edge-computing-driven Internet of Things : A Survey,” vol. 55, no. 8, 2026. doi : 10.1145/355530

[7] M. Whaiduzzaman et al., “A Review of Emerging Technologies for IoT-Based Smart Cities,” Sensors, vol. 22, no. 23, pp. 1–28, Dec. 2022, doi: 10.3390/s22239271.

[8] M. Wang and W. Deng, “Deep face recognition: A survey,” Neurocomputing, vol. 429, pp. 215–244, Mar. 2021, doi: 10.1016/j.neucom.2020.10.081.

[9] Y. Jing, X. Lu, and S. Gao, “3D face recognition: A comprehensive survey in 2022,” Comput. Vis. Media, vol. 9, no. 4, pp. 657–685, Dec. 2023, doi: 10.1007/s41095-022-0317-1.

[10] H. L. Gururaj, B. C. Soundarya, S. Priya, J. Shreyas, and F. Flammini, “A Comprehensive Review of Face Recognition Techniques, Trends, and Challenges,” IEEE Access, vol. 12, no. June, pp. 107903–107926, 2024, doi: 10.1109/ACCESS.2024.3424933.

[11] A. Nemavhola, C. Chibaya, and S. Viriri, “A Systematic Review of CNN Architectures, Databases, Performance Metrics, and Applications in Face Recognition,” Inf., vol. 16, no. 2, pp. 1–41, Feb. 2025, doi: 10.3390/info16020107.

[12] F. Barreto, J. Sarvaiya, and S. Patnaik, “Learning Representations for Face Recognition: A Review from Holistic to Deep Learning,” Adv. Technol. Innov., vol. 7, no. 4, pp. 279–294, Oct. 2022, doi: 10.46604/aiti.2022.8308.

[13] D. Wanyonyi and T. Celik, “Open-Source Face Recognition Frameworks: A Review of the Landscape,” IEEE Access, vol. 10, pp. 50601–50623, 2022, doi: 10.1109/ACCESS.2022.3170037.

[14] Y. Feng, S. Yu, H. Peng, Y.-R. R. Li, and J. Zhang, “Detect Faces Efficiently: A Survey and Evaluations,” IEEE Trans. Biometrics, Behav. Identity Sci., vol. 4, no. 1, pp. 1–18, Dec. 2022, doi: 10.1109/TBIOM.2021.3120412.

[15] [15] J. Nannung, A. Sa’ban Miru, Muliadi, and Gunawan, “Development of an IoT-Based Home Security System Prototype Using Multisensors and ESP32-CAM,” J. Embed. Syst. Secur. Intell. Syst., vol. 6, no. 2, pp. 168–177, Jun. 2025, doi: 10.59562/jessi.v6i2.8456.

[16] L. Parsons, G. Deng, and R. Ross, “Detecting image edges in IoT nodes without FPU,” Multimed. Tools Appl., vol. 83, no. 10, pp. 31161–31175, 2024, doi: 10.1007/s11042-023-16672-4.

[17] C. Contoli, L. Calisti, G. Di Fabrizio, N. Kania, A. Bogliolo, and E. Lattanzi, “A power-aware vision-based virtual sensor for real-time edge computing,” J. Real-Time Image Process., vol. 21, no. 4, pp. 1–12, Aug. 2024, doi: 10.1007/s11554-024-01482-0.

[18] A. Mileva, A. Velinov, L. Hartmann, S. Wendzel, and W. Mazurczyk, “Comprehensive analysis of MQTT 5.0 susceptibility to network covert channels,” Comput. Secur., vol. 104, p. 102207, May 2021, doi: 10.1016/j.cose.2021.102207.

[19] J. Dizdarevic, M. Michalke, and A. Jukan, “Engineering and Experimentally Benchmarking Open Source MQTT Broker Implementations,” May 2023, [Online]. Available: https://arxiv.org/abs/2305.13893.

DOI: https://doi.org/10.20961/jeeict.8.2.125587

Refbacks

  • There are currently no refbacks.