基于多方向Gabor滤波器和Adaboost的虹膜识别方法-Journalof.PDF

第33 卷第1期 东 北 大 学 学 报 ( 自 然 科 学 版 ) Vol33, No. 1 2 0 12 年 1 月 Journal of Northeastern U niversit ( Natural Science) Jan. 2 0 1 2 Gabor Adaboost 张祥德, 黎明奇, 王 琪, 孙艳蕊 ( , 110819) : , . , Gabor , Adaboost . CASIA - IrisV3- Lamp , 99. 85% ; NICE: , , , . : Gabor ; Adaboost ; ; ; : TP 391. 4 1 : A : 1005-3026( 20 12) 01-0039-04 A Multi-orientation Gabor Filter and Adaboost Based Iris Recognition Method ZHAN G Xiang-de, LI Ming-qi, WAN G Qi, S UN Yan-rui ( School of Sciences, Northeastern Universit , Shen ang 110819, China. Corresponding author: WANG Qi, E-mail: wangqimath @ 126. com) Abstract: A new method for iris feature extraction and recognition w as proposed based on the global and local texture information of iris. The normalized iris w as divided first. Then the mult-i orientation Gabor filters w ere used to encode the w hole iris and sub-block iris, and the feature vector w as constructed. Furthermore, the Adaboost algorithm was adopted to choose some features w hose recognition performances are good. T he results of the experiments carried out on CASIA-IrisV3-Lamp database indicate that the accurac of recognition reaches up to 9985% . M oreover, the algorithm has good recognition performance on the N ICE: database w hich contains a large amount of poor qualit images. It shows that the method not onl can take full advantage of the texture information of iris, but also can reduce efficientl the effect of noises. Key words: mult-i orientation Gabor filter; Adaboost algorithm; iris recognition; feature extraction; iris division , [ 3]