A decision-level fusion strategy for multimodal ocular biometric in visible spectrum based on posterior probability

Publication Type:
Conference Proceeding
Citation:
IEEE International Joint Conference on Biometrics, IJCB 2017, 2018, 2018-January pp. 794 - 798
Issue Date:
2018-01-29
Full metadata record
Files in This Item:
Filename Description Size
IJCB2017_for_review_2007_Format ss.pdfAccepted Manuscript521.46 kB
Adobe PDF
© 2017 IEEE. In this work, we propose a posterior probability-based decision-level fusion strategy for multimodal ocular biometric in the visible spectrum employing iris, sclera and peri-ocular trait. To best of our knowledge this is the first attempt to design a multimodal ocular biometrics using all three ocular traits. Employing all these traits in combination can help to increase the reliability and universality of the system. For instance in some scenarios, the sclera and iris can be highly occluded or for completely closed eyes scenario, the peri-ocular trait can be relied on for the decision. The proposed system is constituted of three independent traits and their combinations. The classification output of the trait which produces highest posterior probability is to consider as the final decision. An appreciable reliability and universal applicability of ocular trait are achieved in experiments conducted employing the proposed scheme.
Please use this identifier to cite or link to this item: