An Adaptive CNNs Technology for Robust Iris Segmentation
Iris segmentation algorithms are of great significance in complete iris recognition systems, and directly affect the iris verification and recognition results. However, the conventional iris segmentation algorithms have poor adaptability and are not sufficiently robust when applied to noisy iris dat...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2019-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/8715779/ |
id |
doaj-da07dcbcc0134b15b84aa5fbc8039869 |
---|---|
record_format |
Article |
spelling |
doaj-da07dcbcc0134b15b84aa5fbc80398692021-03-29T22:27:34ZengIEEEIEEE Access2169-35362019-01-017645176453210.1109/ACCESS.2019.29171538715779An Adaptive CNNs Technology for Robust Iris SegmentationYing Chen0Wenyuan Wang1https://orcid.org/0000-0003-1439-2335Zhuang Zeng2https://orcid.org/0000-0002-4898-9546Yerong Wang3School of Software, Nanchang Hangkong University, Nanchang, ChinaSchool of Software, Nanchang Hangkong University, Nanchang, ChinaSchool of Software, Nanchang Hangkong University, Nanchang, ChinaSchool of Software, Nanchang Hangkong University, Nanchang, ChinaIris segmentation algorithms are of great significance in complete iris recognition systems, and directly affect the iris verification and recognition results. However, the conventional iris segmentation algorithms have poor adaptability and are not sufficiently robust when applied to noisy iris databases captured under unconstrained conditions. In addition, there are currently no large iris databases; thus, the iris segmentation algorithms cannot maximize the benefits of convolutional neural networks (CNNs). The main work of this paper is as follows: first, we propose an architecture based on CNNs combined with dense blocks for iris segmentation, referred to as a dense-fully convolutional network (DFCN), and adopt some popular optimizer methods, such as batch normalization (BN) and dropout. Second, because the public ground-truth masks of the CASIA-Interval-v4 and IITD iris databases do not include the labeled eyelash regions, we label these regions that occlude the iris regions using the Labelme software package. Finally, the promising results of experiments based on the CASIA-Interval-v4, IITD, and UBIRIS.V2 iris databases captured under different conditions reveal that the iris segmentation network proposed in this paper outperforms all of the conventional and most of the CNN-based iris segmentation algorithms with which we compared our algorithm's results in terms of various metrics, including the accuracy, precision, recall, f1 score, and nice1 and nice2 error scores, reflecting the robustness of our proposed network.https://ieeexplore.ieee.org/document/8715779/CNNsdense blockdense-fully convolutional networkiris segmentation |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Ying Chen Wenyuan Wang Zhuang Zeng Yerong Wang |
spellingShingle |
Ying Chen Wenyuan Wang Zhuang Zeng Yerong Wang An Adaptive CNNs Technology for Robust Iris Segmentation IEEE Access CNNs dense block dense-fully convolutional network iris segmentation |
author_facet |
Ying Chen Wenyuan Wang Zhuang Zeng Yerong Wang |
author_sort |
Ying Chen |
title |
An Adaptive CNNs Technology for Robust Iris Segmentation |
title_short |
An Adaptive CNNs Technology for Robust Iris Segmentation |
title_full |
An Adaptive CNNs Technology for Robust Iris Segmentation |
title_fullStr |
An Adaptive CNNs Technology for Robust Iris Segmentation |
title_full_unstemmed |
An Adaptive CNNs Technology for Robust Iris Segmentation |
title_sort |
adaptive cnns technology for robust iris segmentation |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2019-01-01 |
description |
Iris segmentation algorithms are of great significance in complete iris recognition systems, and directly affect the iris verification and recognition results. However, the conventional iris segmentation algorithms have poor adaptability and are not sufficiently robust when applied to noisy iris databases captured under unconstrained conditions. In addition, there are currently no large iris databases; thus, the iris segmentation algorithms cannot maximize the benefits of convolutional neural networks (CNNs). The main work of this paper is as follows: first, we propose an architecture based on CNNs combined with dense blocks for iris segmentation, referred to as a dense-fully convolutional network (DFCN), and adopt some popular optimizer methods, such as batch normalization (BN) and dropout. Second, because the public ground-truth masks of the CASIA-Interval-v4 and IITD iris databases do not include the labeled eyelash regions, we label these regions that occlude the iris regions using the Labelme software package. Finally, the promising results of experiments based on the CASIA-Interval-v4, IITD, and UBIRIS.V2 iris databases captured under different conditions reveal that the iris segmentation network proposed in this paper outperforms all of the conventional and most of the CNN-based iris segmentation algorithms with which we compared our algorithm's results in terms of various metrics, including the accuracy, precision, recall, f1 score, and nice1 and nice2 error scores, reflecting the robustness of our proposed network. |
topic |
CNNs dense block dense-fully convolutional network iris segmentation |
url |
https://ieeexplore.ieee.org/document/8715779/ |
work_keys_str_mv |
AT yingchen anadaptivecnnstechnologyforrobustirissegmentation AT wenyuanwang anadaptivecnnstechnologyforrobustirissegmentation AT zhuangzeng anadaptivecnnstechnologyforrobustirissegmentation AT yerongwang anadaptivecnnstechnologyforrobustirissegmentation AT yingchen adaptivecnnstechnologyforrobustirissegmentation AT wenyuanwang adaptivecnnstechnologyforrobustirissegmentation AT zhuangzeng adaptivecnnstechnologyforrobustirissegmentation AT yerongwang adaptivecnnstechnologyforrobustirissegmentation |
_version_ |
1724191560079245312 |