Federated Learning in Finger Biometrics:Opportunities and Challenges

Akshay Juneja; Himani Bhardwaj; Deepak Painuli; Divya Mishra; Priyanka Kumari

doi:10.70454/IJMRE.60302

Authors

Akshay Juneja Department of Computer Science and Engineering, College of Smart Computing, COER University, Roorkee, Uttarakhand, 247667, India Author
Himani Bhardwaj Department of Computer Science and Engineering, College of Smart Computing, COER University, Roorkee, Uttarakhand, 247667, India Author
Deepak Painuli Department of Computer Science and Engineering, College of Smart Computing, COER University, Roorkee, Uttarakhand, 247667, India Author
Divya Mishra Department of Computer Science and Engineering, College of Smart Computing, COER University, Roorkee, Uttarakhand, 247667, India. Author
Priyanka Kumari ECE Department, School of Engineering and Technology, CGC University, Mohall, Punjab, India Author

DOI:

https://doi.org/10.70454/IJMRE.60302

Keywords:

Biometrics, deep learning, fingerprint recognition, federated learning, finger vein recognition

Abstract

In past decade, various architectures for fingerprint and finger vein biometric systems have been designed. They are utilized in various universities, organizations, and institutions. But there is an increase in number of forgeries due to lack of security in single biometric implementation across the globe. Therefore, the researchers have proposed different multi-modal architectures that utilize integration of fingerprint and finger vein identification, including federated learning, to preserve user privacy and security. The fingerprint data is processed using discrete cosine transform method. The deep learning architectures along with reservoir sampling are used to classify genuine and fake users. The federated learning is used to classify authenticate users to avoid data island problem. It permits to share model weights and divides user data into personalized and shared parts to ensure users’ privacy. This paper highlights key challenges such as applicability and interpretability of optimization techniques. The interpretability is achieved using an expert system that uses its inference engine and knowledge base to identify the model’s behavior that gives a reasonable explanation. This article discusses the limitations and challenges of federal learning with finger vein recognition and fingerprint.

References

[1] Hammed, M., Wane, M. A., Shakily, K. A., Shaba, H., & Bad El-Latin, A. A. (2024). Deep cancelable multibiometric finger vein and fingerprint authentication with non-negative matrix factorization. IEEE Access.

[2] Alshardan, A., Kumar, A., Agamid, M., Mashie, M., Kalahari, S., Malherbe, A. A., & Alzahrani, Y. (2024). Multimodal biometric identification: leveraging convolution neural network (CNN) architectures and fusion techniques with fingerprint and finger vein data. Peer Computer Science, 10, e2440.

[3] Kyeremeh, G. K., Abdul-Al, M., Awaji, R., Ali, N. T., & Abd-Alhameed, R. A. (2025). Fusion of Hand Biometrics for Border Control Involving Fingerprint and Finger Vein. IEEE Access.

[4] Cadim, S. M., Paw, J. K. S., Take, Y. C., & Amen, S. (2024). Deep Learning Models for Biometric Recognition based on Face, Finger vein, Fingerprint, and Iris: A Survey. Journal of Smart Internet of Tidings, 2024(1), 117-157.

[5] Wang, Y., 2025. Federated Learning Applications in Fingerprint and Finger Vein Recognition. In ITM Web of Conferences (Vol. 70, p. 01023). EDP Sciences.

[6] Avenues, S. D., Sudharshan, B. S., Rishwanth, M., Suit, U., & Anita, G. (2025, June). Biometric Recognition using Fingerprint and Finger Vein. In Proceedings of the International Conference on Materials, Energy, Environment & Manufacturing Sciences & Computational Intelligence and Smart Communication (MEEMS-CISC 2024) (Vol. 269, p. 40). Springer Nature.

[7] Wang, S., Sheen, Y., & Yang, W. (2024). Touch less Finger Vein and Fingerprint Verification via Exploiting Attention-Based Cross-Domain Fusion. IEEE Transactions on Circuits and Systems for Video Technology.

[8] Juneau, A., Kumara, P., Bhatia, M., Tindal, H., Waxhaw, G., & Khan, P. (2024, December). Revival of Fingerprint Recognition-based Biometric Systems. In 2024 Eighth International Conference on Parallel, Distributed and Grid Computing (PDGC) (pp. 622-627). IEEE.

[9] Gomez-Barrera, M. (2025). Finger vein. In Encyclopedia of Cryptography, Security and Privacy (pp. 905-910). Cham: Springer Nature Switzerland.

[10] Li, E., Yang, L., Wang, K., Wang, Y., & Yin, Y. (2025). Vein Pattern Based Partial Finger Vein Alignment and Recognition. IEEE Transactions on Biometrics, Behavior, and Identity Science.

[11] Painful, D., Hardwar, S., & Chose, U. (2023). Optimized diagnosis of central nervous system (CNS) cancer using gene expression microarray & machine learning (ML) methods. European Chemical Bulletin, 12(10), 9757-9771.

[12] Painful, D., Hardwar, S., & Chose, U. (2025). Efficient Feature Selection and Hyper parameter Tuning for Improved Speech Signal-Based Parkinson’s Disease Diagnosis via Machine Learning Techniques. Health Technology Assessment in Action, 9(1).

[13] NeeluJyoti, D. M. D. N. D., & Painful, A. D. (2018). Fuzzy Expert System to diagnose Psoriasis Disease. International Journal of Computer Science and Information Security (IJCSIS), 16(9).

[14] A. Sony, M. Sandy, & Y. S. Rae. Privacy Preserving Fingerprint Classification Using Federated Learning. In International Conference on Deep Learning, Artificial Intelligence and Robotics (pp. 71-80). Cham: Springer Nature Switzerland (2023, December).