PERSON RECOGNITION BY THE SCLERAL VASCULAR PATTERN (EYEPRINT) ON MOBILE DEVICES
Keywords:
biometrics, eyeprint, scleral vessels, iris, mobile biometrics, multimodal fusion, image quality assessment.Abstract
The paper addresses person recognition by the scleral vascular pattern (eyeprint): conjunctival and episcleral blood vessels captured in the visible spectrum by a standard smartphone camera without additional hardware. Methods for sclera segmentation, vessel extraction and template matching are analysed, from classical descriptors to hybrid convolutional-transformer architectures. The main limitations of the modality are identified: non-linear vessel deformation during eye movement, accuracy loss across devices and acquisition conditions, and demographic bias of segmentation models. In the visible spectrum the scleral vessels and the iris degrade under different conditions and therefore complement each other. A multimodal system is proposed in which quality scores of the ocular regions and the gaze angle control the fusion weights of the two modalities for each frame. An experimental protocol on the public MOBIUS, SBVPI and UBIRIS.v2 datasets with cross-device evaluation and ablation analysis is described. The hypothesis is that adaptive fusion lowers the equal error rate compared with single modalities and fixed-weight fusion, especially for dark irides and averted gaze.
References
1. Ant Financial acquires EyeVerify [Электронный ресурс] // NFCW. – 14.09.2016. – URL: https://www.nfcw.com/2016/09/14/347254/ant-financial-acquires-eyeverify/ (дата обращения: 03.10.2026).
2. Das A. [et al.] Sclera segmentation and joint recognition benchmarking competition: SSRBC 2023 // Proc. IEEE International Joint Conference on Biometrics (IJCB 2023). – 2023. – DOI: 10.1109/IJCB57857.2023.10448601.
3. Das S. [et al.] Sclera biometrics in restricted and unrestricted environment with cross dataset evaluation // Displays. – 2022. – DOI: 10.1016/j.displa.2022.102257.
4. Derakhshani R. [et al.] A new biometric modality based on conjunctival vasculature // Intelligent Engineering Systems through Artificial Neural Networks (ANNIE 2006). – ASME Press, 2006. – DOI: 10.1115/1.802566.paper74 .
5. EyeVerify technology uses smartphone cameras to easily verify ID by eye veins [Электронный ресурс] // Biometric Update. – 08.2015. – URL: https://www.biometricupdate.com/201508/eyeverify-technology-uses-smartphone-cameras-to-easily-verify-id-by-eye-veins (дата обращения: 03.10.2026).
6. Gottemukkula V. [et al.] Fusing iris and conjunctival vasculature: ocular biometrics in the visible spectrum // Proc. IEEE Conference on Technologies for Homeland Security (HST 2012). – 2012. – DOI: 10.1109/THS.2012.6459841.
7. Gottemukkula V. [et al.] Method for using visible ocular vasculature for mobile biometrics // IET Biometrics. – 2016. – Vol. 5. – DOI: 10.1049/iet-bmt.2014.0059.
8. Lee S. [et al.] Robust sclera recognition based on a local spherical structure // Expert Systems with Applications. – 2021. – DOI: 10.1016/j.eswa.2021.116081.
9. Proença H., Filipe S., Santos R., Oliveira J., Alexandre L.A. The UBIRIS.v2: a database of visible wavelength iris images captured on-the-move and at-a-distance // IEEE Transactions on Pattern Analysis and Machine Intelligence. – 2010. – Vol. 32, No. 8. – P. 1529–1535. – DOI: 10.1109/TPAMI.2009.66.
10. Rattani A., Derakhshani R. Ocular biometrics in the visible spectrum: a survey // Image and Vision Computing. – 2017. – Vol. 59. – P. 1–16. – DOI: 10.1016/j.imavis.2016.11.019.
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. Vitek M., Rot P., Štruc V., Peer P. A comprehensive investigation into sclera biometrics: a novel dataset and performance study // Neural Computing and Applications. – 2020. – Vol. 32. – DOI: 10.1007/s00521-020-04782-1.
14. Vitek M. [et al.] Exploring bias in sclera segmentation models: a group evaluation approach // IEEE Transactions on Information Forensics and Security. – 2023. – Vol. 18. – DOI: 10.1109/TIFS.2022.3216468.
15. Vitek M. [et al.] GazeNet: a lightweight multitask sclera feature extractor // Alexandria Engineering Journal. – 2025. – DOI: 10.1016/j.aej.2024.11.011.
16. Wang C. [et al.] Sclera-TransFuse: fusing vision transformer and CNN for accurate sclera segmentation and recognition // IEEE Transactions on Biometrics, Behavior, and Identity Science. – 2024. – DOI: 10.1109/TBIOM.2024.3415484.
17. Zhou Z., Du E.Y., Thomas N.L., Delp E.J. A new human identification method: sclera recognition // IEEE Transactions on Systems, Man, and Cybernetics. Part A: Systems and Humans. – 2012. – Vol. 42, No. 3. – P. 571–583. – DOI: 10.1109/TSMCA.2011.2170416.
