Thin iris region recognition using the RED algorithm

Iris recognition system provides automatic identification of an individual based on a unique feature that located inside the iris. To detect these features Ridge Energy Detection (RED) algorithm, which is one of the most accurate and fast identification method to detect iris, features today. RED alg...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Engineering and Technology Journal 2016-05, Vol.34 (5A), p.1017-1028
Hauptverfasser: Umran, Safa S., al-Hilali, Aqil A.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Iris recognition system provides automatic identification of an individual based on a unique feature that located inside the iris. To detect these features Ridge Energy Detection (RED) algorithm, which is one of the most accurate and fast identification method to detect iris, features today. RED algorithm is applied to rectangle iris that generated from the normalization process. RED algorithm constructed a template contains the features of the iris by using two types of filter (horizontal and vertical). In this paper, two different rectangle iris templates were generated, the first rectangle iris template is generated as common way in this filed by taking rectangle iris template that contains full iris region, while the other is a novel rectangle iris template which contains a ring from the iris that near to the pupil. Both rectangle iris templates were applied to the RED algorithm in order to compare the accuracy and time between them. The novel rectangle iris template has 0 % fault in recognition and faster in extraction process by six times than the common rectangle iris template in extracting the features of iris when using RED algorithm and also three times faster in matching process when comparing with novel rectangle iris template than rectangle iris template that contains full iris region.
ISSN:1681-6900
2412-0758
2412-0758
DOI:10.30684/etj.34.5A.18