HYBRID SYSTEMS FOR HUMAN RECOGNITION FROM EYE IMAGES
DOI:
https://doi.org/10.47390/ydif-y2026v2i19/n05Keywords:
hybrid biometric system, iris, sclera, multimodal biometrics, fusion levels, deep learning, mobile biometrics.Abstract
The paper considers hybrid systems for human recognition from eye images that combine two traits of a single image — the iris and the scleral vascular pattern — as well as different classes of methods: handcrafted descriptors and deep neural networks. A classification of such systems by the level of information fusion is proposed: image, feature, matching score and decision level. For each level, its advantages, limitations, data alignment requirements and published implementations are identified. Published iris–sclera systems from 2010–2026 are analysed; score-level fusion is shown to dominate, while results of different authors are not comparable because of differences in datasets, metrics and evaluation protocols. A hybrid system structure with two-level fusion is proposed: deep features of the two modalities are fused at the feature level, and the final score additionally takes into account a classical iris code. A plan is formulated for comparing all fusion levels within a single protocol on the MOBIUS dataset captured with three smartphones, considering accuracy, template size and computational cost, which will allow a justified choice of the fusion level for mobile devices.
References
1. Adeniyi J.K. [et al.] Fusion of Siamese network-based sclera and iris detection: a multimodal biometrics approach using a sclera detection tracing algorithm // IET Information Security. – 2026. – DOI: 10.1049/ise2/5569382.
2. Chuang C.-W. [et al.] Deep-learning based joint iris and sclera recognition with YOLO network for identity identification // Journal of Advances in Information Technology. – 2021. – Vol. 12, No. 1. – P. 60–65. – DOI: 10.12720/jait.12.1.60-65.
3. Crihalmeanu S., Ross A. Multispectral scleral patterns for ocular biometric recognition // Pattern Recognition Letters. – 2012. – Vol. 33. – DOI: 10.1016/j.patrec.2011.11.006.
4. Das A. [et al.] A decision-level fusion strategy for multimodal ocular biometric in visible spectrum based on posterior probability // Proc. IEEE International Joint Conference on Biometrics (IJCB 2017). – 2017. – DOI: 10.1109/BTAS.2017.8272772.
5. Es-Sobbahi H. [et al.] Multimodal biometrics: a review of handcrafted and AI-based fusion approaches // IET Biometrics. – 2025. – DOI: 10.1049/bme2/5055434.
6. Haghighat M. [et al.] Discriminant correlation analysis: real-time feature level fusion for multimodal biometric recognition // IEEE Transactions on Information Forensics and Security. – 2016. – Vol. 11. – DOI: 10.1109/TIFS.2016.2569061.
7. Jain D. [et al.] Fusion of iris and sclera using phase intensive rubbersheet mutual exclusion for periocular recognition // Image and Vision Computing. – 2020. – DOI: 10.1016/j.imavis.2020.104024.
8. Kabir W. [et al.] A multi-biometric system based on feature and score level fusions // IEEE Access. – 2019. – Vol. 7. – DOI: 10.1109/ACCESS.2019.2914992.
9. Pathak M. [et al.] Multimodal eye biometric system based on contour based E-CNN and multi algorithmic feature extraction using SVBF matching // International Journal of Innovative Technology and Exploring Engineering. – 2019. – DOI: 10.35940/ijitee.I7729.078919.
10. Ross A., Jain A. Information fusion in biometrics // Pattern Recognition Letters. – 2003. – Vol. 24, No. 13. – P. 2115–2125. – DOI: 10.1016/S0167-8655(03)00079-5.
11. Sclera datasets: SBVPI, MOBIUS [Электронный ресурс] / University of Ljubljana. – URL: https://sclera.fri.uni-lj.si/datasets.html (дата обращения: 03.10.2026).
12. Tankasala S.P. [et al.] Visible spectrum, bi-modal ocular biometrics // Procedia Technology. – 2012. – Vol. 6. – DOI: 10.1016/j.protcy.2012.10.068.
13. Yu Y. [et al.] Segmentation and multimodal fusion in ocular biometric recognition: a review // Medinformatics. – 2026. – DOI: 10.47852/bonviewMEDIN62029908.
14. Zhang Q. [et al.] Deep feature fusion for iris and periocular biometrics on mobile devices // IEEE Transactions on Information Forensics and Security. – 2018. – Vol. 13. – DOI: 10.1109/TIFS.2018.2833033.
15. Zhou Z. [et al.] Multimodal eye recognition // Proc. SPIE. – 2010. – DOI: 10.1117/12.849731.
16. Zhou Z. [et al.] Quality fusion based multimodal eye recognition // Proc. IEEE International Conference on Systems, Man, and Cybernetics (SMC 2012). – 2012. – DOI: 10.1109/ICSMC.2012.6377912.
17. Zhou Z. [et al.] A comprehensive multimodal eye recognition // Signal, Image and Video Processing. – 2013. – DOI: 10.1007/s11760-013-0468-8.
This work is licensed under a 