Title :
Efficient and robust approach of iris recognition through Fisher Linear Discriminant Analysis method and Principal Component Analysis method
Author :
Haq, Qazi Emad ul ; Javed, Muhammad Younus ; Sami ul Haq, Q.
Author_Institution :
Dept. of Comput. Eng., Nat. Univ. of Sci. & Technol. (NUST), Islamabad
Abstract :
Iris recognition has emerged as a vital and tested methodology for research investigations and routine security applications in the context of increasing security requirements. Thus biometrics has attained a very significant place in human verification and identification. In this paper, an efficient and precised methodology is brought out through using Fisher linear discriminant analysis method and principal component analysis method. These methodologies create different sections in low dimensional sub space. The suggested system in this research work contains four components i.e. preprocessing, segmentation, feature extraction and matching. The preprocessing part again consist of pupil localization, image refinement, iris localization and normalization procedures. The suggested algorithm in this research paper was tested on CASIA Iris image database. The soundness and time efficiency of the suggested algorithm proves it as perfect technique for real time applications.
Keywords :
biometrics (access control); feature extraction; image matching; image segmentation; principal component analysis; Fisher linear discriminant analysis; biometrics; feature extraction; human identification; human verification; image matching; image preprocessing; image refinement; image segmentation; iris localization; iris recognition; low dimensional sub space; normalization procedure; principal component analysis; pupil localization; Biometrics; Feature extraction; Humans; Image segmentation; Iris recognition; Linear discriminant analysis; Principal component analysis; Robustness; Security; Testing; Biometrics; FLDA; Iris Recognition; PCA;
Conference_Titel :
Multitopic Conference, 2008. INMIC 2008. IEEE International
Conference_Location :
Karachi
Print_ISBN :
978-1-4244-2823-6
Electronic_ISBN :
978-1-4244-2824-3
DOI :
10.1109/INMIC.2008.4777739