Proceedings of
5th International Conference on Advances in Computing, Communication and Information Technology CCIT 2017
"HUMAN IDENTIFICATION USING MULTI-REGION PCA FOR IRIS RECOGNITION"
Abstract: “Biometric identification is a constantly growing fie l d. Many physiological characteristics may be used for r eliable identification. T he most secure among them uses the iris pattern , because it is unique and stable from eye to eye and from person to person. The approach presented in this article guides through every step of iris - based human identification system from eye localization to iris extraction and identification using PCA features. The novelty of this paper consists of the application of PCA on multi - region iris images. The PCA identification based on multiple regions of interest (ROI) allows higher ident ification accuracy while extract ing muc h fewer features.”
Keywords: iris recognition; eigeniris; PCA; performance evaluation; UPOL database