Title: Iris recognition: using a statistical model of shape and spatial relation for effective segmentation

Authors: Houda Khmila; Said Ettaieb; Nadia Feddaoui; Nadia Smaoui

Addresses: Control and Energy Management Laboratory, University of Sfax – ENIS, Soukra Road km 5, BP W, 3038, Sfax, Tunisia ' Research Laboratory of Image, Signal and Information Technology, University of Tunis El Manar, Rommana 1068, Tunis-B.P. n° 94, Tunisia ' Research Laboratory of Image, Signal and Information Technology, University of Tunis El Manar, Rommana 1068, Tunis-B.P. n° 94, Tunisia ' Control and Energy Management Laboratory, University of Sfax – ENIS, Soukra Road km 5, BP W, 3038, Sfax, Tunisia

Abstract: This paper presents a new segmentation method based both on active shape model (ASM) and spatial distance model to segment iris structures. The extracted iris is normalised by Daugman's rubber sheet model. Then the features are extracted by the 1D log Gabor filter, afterwards the hamming distance is used to compare the binary codes stored previously. The evaluation experiments are performed on a set of 300 right iris images from 41 persons of CASIA-IrisV4 database. We obtain a correct recognition rate of 100%. The experimental results have shown that the performance of the proposed approach is encouraging.

Keywords: biometrics; iris recognition; segmentation; ASM+D; identification.

DOI: 10.1504/IJDSSS.2018.090871

International Journal of Digital Signals and Smart Systems, 2018 Vol.2 No.1, pp.36 - 49

Received: 07 Feb 2017
Accepted: 11 Sep 2017
Published online: 30 Mar 2018 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article