Iris Recognition Using an Enhanced Pre-Trained Backbone Based on Anti-Aliased CNNs
Published in IEEE Access, 2024
This work proposes an iris recognition approach that enhances a pretrained CNN backbone with anti-aliasing techniques to improve shift-invariance in feature extraction. The method retains the advantage of not requiring retraining on iris-specific data, while incorporating bit-shifting to handle variability due to pupil dilation. Evaluation on standard iris databases demonstrates improved recognition performance over baseline pretrained features.
How to cite
@article{zambrano2024irisantialias,
author = {Zambrano, Jorge E. and Pilataxi, J. I. and Perez, Claudio A. and Bowyer, Kevin W.},
journal = {IEEE Access},
title = {Iris Recognition Using an Enhanced Pre-Trained Backbone Based on Anti-Aliased CNNs},
year = {2024},
volume = {12},
pages = {94570--94583},
doi = {10.1109/ACCESS.2024.3425648}
}
` ``
---
Recommended citation: J. E. Zambrano, J. I. Pilataxi, C. A. Perez and K. W. Bowyer, "Iris Recognition Using an Enhanced Pre-Trained Backbone Based on Anti-Aliased CNNs," in IEEE Access, vol. 12, pp. 94570-94583, 2024, doi: 10.1109/ACCESS.2024.3425648.
Download Paper | Download Bibtex