Deep learning architecture for iris recognition based on optimal Gabor filters and deep belief network

被引:37
作者
He, Fei [1 ,2 ,3 ]
Han, Ye [4 ,5 ]
Wang, Han [1 ,3 ]
Ji, Jinchao [1 ,3 ]
Liu, Yuanning [4 ,5 ]
Ma, Zhiqiang [1 ,3 ]
机构
[1] Northeast Normal Univ, Sch Comp Sci & Informat Technol, Changchun, Peoples R China
[2] Northeast Normal Univ, Sch Environm, Changchun, Peoples R China
[3] Northeast Normal Univ, Inst Computat Biol, Changchun, Peoples R China
[4] Jilin Univ, Coll Comp Sci & Technol, Changchun, Peoples R China
[5] Jilin Univ, Key Lab Symbol Computat & Knowledge Engn, Minist Educ, Changchun, Peoples R China
基金
中国国家自然科学基金;
关键词
iris recognition; deep learning; Gabor filters; deep belief network; particle swarm optimization; LOCALIZATION; INFORMATION; PARAMETERS; DISTANCE; MACHINE; FUSION; SYSTEM; IMAGES; SPACE;
D O I
10.1117/1.JEI.26.2.023005
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Gabor filters are widely utilized to detect iris texture information in several state-of-the-art iris recognition systems. However, the proper Gabor kernels and the generative pattern of iris Gabor features need to be predetermined in application. The traditional empirical Gabor filters and shallow iris encoding ways are incapable of dealing with such complex variations in iris imaging including illumination, aging, deformation, and device variations. Thereby, an adaptive Gabor filter selection strategy and deep learning architecture are presented. We first employ particle swarm optimization approach and its binary version to define a set of data-driven Gabor kernels for fitting the most informative filtering bands, and then capture complex pattern from the optimal Gabor filtered coefficients by a trained deep belief network. A succession of comparative experiments validate that our optimal Gabor filters may produce more distinctive Gabor coefficients and our iris deep representations be more robust and stable than traditional iris Gabor codes. Furthermore, the depth and scales of the deep learning architecture are also discussed. (C) 2017 SPIE and IS&T
引用
收藏
页数:13
相关论文
共 69 条
[21]   Automated defect detection in textured surfaces using optimal elliptical Gabor filters [J].
Hu, Guang-Hua .
OPTIK, 2015, 126 (14) :1331-1340
[22]  
Huang YP, 2002, 2002 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND CYBERNETICS, VOLS 1-4, PROCEEDINGS, P450
[23]  
Hui Zhang, 2012, Proceedings of the 2012 IEEE 3rd International Conference on Software Engineering and Service Science (ICSESS), P131, DOI 10.1109/ICSESS.2012.6269422
[24]   Secondary iris recognition method based on local energy-orientation feature [J].
Huo, Guang ;
Liu, Yuanning ;
Zhu, Xiaodong ;
Dong, Hongxing .
JOURNAL OF ELECTRONIC IMAGING, 2015, 24 (01)
[25]   Iris localization using local histogram and other image statistics [J].
Ibrahim, Muhammad Talal ;
Khan, Tariq M. ;
Khan, Shahid A. ;
Khan, M. Aurangzeb ;
Guan, Ling .
OPTICS AND LASERS IN ENGINEERING, 2012, 50 (05) :645-654
[26]   Biometrics: A tool for information security [J].
Jain, Anil K. ;
Ross, Arun ;
Pankanti, Sharath .
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, 2006, 1 (02) :125-143
[27]   Iris localization in frontal eye images for less constrained iris recognition systems [J].
Jan, Farmanullah ;
Usman, Imran ;
Agha, Shahrukh .
DIGITAL SIGNAL PROCESSING, 2012, 22 (06) :971-986
[28]   Coupled Discriminative Feature Learning for Heterogeneous Face Recognition [J].
Jin, Yi ;
Lu, Jiwen ;
Ruan, Qiuqi .
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, 2015, 10 (03) :640-652
[29]   Supervised defect detection on textile fabrics via optimal Gabor filter [J].
Jing, Junfeng ;
Yang, Panpan ;
Li, Pengfei ;
Kang, Xuejuan .
JOURNAL OF INDUSTRIAL TEXTILES, 2014, 44 (01) :40-57
[30]   Iris Recognition using Robust Localization and Nonsubsampled Contourlet Based Features [J].
Khalighi, Sirvan ;
Pak, Fatemeh ;
Tirdad, Parisa ;
Nunes, Urbano .
JOURNAL OF SIGNAL PROCESSING SYSTEMS FOR SIGNAL IMAGE AND VIDEO TECHNOLOGY, 2015, 81 (01) :111-128