UPM Institutional Repository

Model of Bayesian tangent eye shape for eye capture


Nsaef, Asama Kuder and Jaafar, Azizah and Sliman, Layth and Sulaiman, Riza and O. K. Rahmat, Rahmita Wirza (2014) Model of Bayesian tangent eye shape for eye capture. In: 14th International Conference on Intelligent Systems Design and Applications (ISDA 2014), 28-30 Nov. 2014, Okinawa, Japan. (pp. 82-88).


Iris recognition system captures an image of an individual's eye. In addition, the process of segmentation, normalization and feature extraction is followed by the iris of an eye image in the system. Using the algorithms proposed by J. Daugman, Iris recognition system has significantly improved over the last decade, and it has been used in so many practical applications. However, some difficulties related to Iris position and movement are still to be improved. To overcome these difficulties one can enhance the image acquisition process. Obtaining a method in extracting quality of eye images automatically from the video stream is the main area of interest in this study. Besides, a Bayesian inference solution called Bayesian Tangent Eye Shape Model (BTESM) was suggested depending on estimation of tangent shape. During image acquisition, constraints on the position and motion of the subjects can be decreased owing to this approach. Owing to maximum a posteriori estimation, we can identify similarity transform coefficients as well as the eye shape parameters in BTESM. To apply the maximum a posteriori procedure, tangent Eye shape vector was considered the state of the model which is hidden and expectation maximization depending on searching algorithm was adopted. Hence, after being tested and matched to future studies, the acquisitioned eye image has been proved to be adequate for Iris recognition system.

Download File

Text (Abstract)
Model of Bayesian tangent eye shape for eye capture.pdf

Download (35kB) | Preview

Additional Metadata

Item Type: Conference or Workshop Item (Paper)
Divisions: Faculty of Computer Science and Information Technology
DOI Number: https://doi.org/10.1109/ISDA.2014.7066277
Publisher: IEEE
Keywords: Eye detection; Iris recognition based on video; Iris recognition on the move; Bayesian tangent eye shape model; Estimation of eye position
Depositing User: Nabilah Mustapa
Date Deposited: 08 Jul 2019 02:02
Last Modified: 08 Jul 2019 02:02
Altmetrics: http://www.altmetric.com/details.php?domain=psasir.upm.edu.my&doi=10.1109/ISDA.2014.7066277
URI: http://psasir.upm.edu.my/id/eprint/69610
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item