Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Feb 12:2014:670934.
doi: 10.1155/2014/670934. eCollection 2014.

Novel approaches to improve iris recognition system performance based on local quality evaluation and feature fusion

Ying Chen  1 , Yuanning Liu  2 , Xiaodong Zhu  2 , Huiling Chen  3 , Fei He  2 , Yutong Pang  2
Affiliations

Novel approaches to improve iris recognition system performance based on local quality evaluation and feature fusion

Ying Chen et al. ScientificWorldJournal. .

Abstract

For building a new iris template, this paper proposes a strategy to fuse different portions of iris based on machine learning method to evaluate local quality of iris. There are three novelties compared to previous work. Firstly, the normalized segmented iris is divided into multitracks and then each track is estimated individually to analyze the recognition accuracy rate (RAR). Secondly, six local quality evaluation parameters are adopted to analyze texture information of each track. Besides, particle swarm optimization (PSO) is employed to get the weights of these evaluation parameters and corresponding weighted coefficients of different tracks. Finally, all tracks' information is fused according to the weights of different tracks. The experimental results based on subsets of three public and one private iris image databases demonstrate three contributions of this paper. (1) Our experimental results prove that partial iris image cannot completely replace the entire iris image for iris recognition system in several ways. (2) The proposed quality evaluation algorithm is a self-adaptive algorithm, and it can automatically optimize the parameters according to iris image samples' own characteristics. (3) Our feature information fusion strategy can effectively improve the performance of iris recognition system.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Diagram of image preprocessing.
Figure 2
Figure 2
Steps involved in preprocessing. (a) Original iris image, (b) ROI selection, (c) Canny transformation, (d) pupil and limbic localization, (e) iris region segmentation, (f) iris region is divided into circular regions, (g) iris region is divided into 8 parts labeled with track0, track1,…, track7 from bottom to top, and (h) enhanced image after histogram equalization.
Figure 3
Figure 3
Analysis processing diagram.
Figure 4
Figure 4
The complex textural patterns of image.
Figure 5
Figure 5
Quality evaluation block diagram.
Figure 6
Figure 6
Bar chart of image quality evaluation. (a) CASIA-V1 image, (b) CASIA-V3 Interval image, (c) one-dimensional entropy, (d) two-dimensional entropy, (e) Haralick texture entropy, (f) Tamura texture directionality measurement, (g) Tamura texture contrast measurement, and (h) gray variance.
Figure 7
Figure 7
Architecture of the proposed fusion recognition system.
Figure 8
Figure 8
Sample images from CASIA-V1, CASIA-V3 Interval, MMU-V1, and JLUBRIRIS-V1 databases. (a) CASIA-V1, (b) CASIA-V3 Interval, (c) MMU-V1, and (d) JLUBRIRIS-V1.
Figure 9
Figure 9
RAR of different tracks. (a) (b) (c) are experimental results of TestGroupOne, (a) is RAR produced by 2D Gabor features, (b) is RAR produced by GLCM features, and (c) is RAR produced by combined features. (d) (e) (f) are experimental results of TestGroupTwo, (d) is RAR produced by 2D Gabor features, (e) is RAR produced by GLCM features, and (f) is RAR produced by combined features. (h) (i) (j) are experimental results of TestGroupThree, (h) is RAR produced by 2D Gabor features, (i) is RAR produced by GLCM features, and (j) is RAR produced by combined features.
Figure 10
Figure 10
Distance distributions of the intra-class and inter-class patterns for the four iris image databases without fusion. (a) CAISA-V1, (b) CASIA-V3 Interval, (c) MMU-V1 and (d) JLUBRIRIS-V1 iris image databases.
Figure 11
Figure 11
Distance distributions of the intra-class and inter-class patterns for the four iris image databases with fusion. (a) CAISA-V1, (b) CASIA-V3 Interval, (c) MMU-V1 and (d) JLUBRIRIS-V1.
Figure 12
Figure 12
ROC curves for (a) CASIA-V1, (b) CASIA-V3 Interval, (c) MMU-V1 and (d) JLUBRIRIS-V1 databases.
See this image and copyright information in PMC

References

    1. Broussard RP, Kennell LR, Ives RW. Identifying discriminatory information content within the iris. Biometric Technology for Human Identification V; March 2008;
    1. Williams GO. Iris recognition technology. IEEE Aerospace and Electronic Systems Magazine. 1997;12(4):23–29.
    1. Du YZ, Ives R, Bonney B, Etter D. Analysis of partial iris recognition. The International Society for Optical Engineering; 2005; Orlando, Fla, USA. pp. 31–40.
    1. Du Y, Bonney B, Ives R, Etter D, Schnltz R. Analysis of partial iris recognition using a 1-D approach. Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP ’05); March 2005; pp. II961–II964.
    1. Hollingsworth KP, Bowyer KW, Flynn PJ. The best bits in an Iris code. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2009;31(6):964–973. - PubMed

Publication types

LinkOut - more resources