Iris recognition based on matching pursuits

被引:1
作者
Lee, Yueh-Shiun [1 ]
Ho, Jinn [3 ]
Hwang, Wen-Liang [3 ]
Huang, Chung-Lin [1 ,2 ]
机构
[1] Natl Tsing Hua Univ, Dept Elect Engn, Hsinchu, Taiwan
[2] Fo Guang Univ, Dept Informat, Lan, Taiwan
[3] Acad Sinica, Inst Sci Informat, Taipei, Taiwan
来源
2008 FOURTH INTERNATIONAL CONFERENCE ON INTELLIGENT INFORMATION HIDING AND MULTIMEDIA SIGNAL PROCESSING, PROCEEDINGS | 2008年
关键词
iris recognition; matching pursuit;
D O I
10.1109/IIH-MSP.2008.331
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We propose a novel Dynamic Programming Matching Pursuit (DPMP) algorithm for iris recognition. The method modifies the matching pursuit (MP) algorithm to select the most representative atoms for the iris recognition. In the experiments, we demonstrate that our system has (1) better performance for both personal identification and verification, and (2) a better ROC curve, (3) less computation than the conventional MP-based iris recognition.
引用
收藏
页码:183 / +
页数:2
相关论文
共 8 条
[1]   HIGH CONFIDENCE VISUAL RECOGNITION OF PERSONS BY A TEST OF STATISTICAL INDEPENDENCE [J].
DAUGMAN, JG .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1993, 15 (11) :1148-1161
[2]  
KODNO K, Patent No. 7099495
[3]   Personal identification based on iris texture analysis [J].
Ma, L ;
Tan, T ;
Wang, YH ;
Zhang, DX .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2003, 25 (12) :1519-1533
[4]  
MA L, 2004, IEEE T IP, V13
[5]  
MALAT S, 1933, IEEE T SIGNAL PR DEC
[6]  
SUN Z, 2005, IEEE T SMC, V35
[7]  
YUAN W, 2006, 6 WORLD C INT CONTR, V2, P10388
[8]  
CASIA IRIS IMAGE DAT