Improving colour iris segmentation using a model selection techniquePattern Recognition Letters

About

Authors
Yang Hu, Konstantinos Sirlantzis, Gareth Howells
Year
2015
DOI
10.1016/j.patrec.2014.12.012
Subject
Signal Processing / Software / Artificial Intelligence / Computer Vision and Pattern Recognition

Similar

Optical determination of intracellular water in apoptotic cells

Authors:
Michael A. Model, Ethan Schonbrun
2013

An improved model of varicocele in rats using microsurgical technique

Authors:
B.B. Najari, R. Ramasamy, M. Katz, P.N. Schlegel, P.S. Li, M. Goldstein
2012

Iris Segmentation in Visible Wavelength Images using Circular Gabor Filters and Optimization

Authors:
Abduljalil Radman, Kasmiran Jumari, Nasharuddin Zainal
2014

Improved segmentation of low-contrast lesions using sigmoid edge model

Authors:
Amir Hossein Foruzan, Yen-Wei Chen
2015

Text

Accepted Manuscript

Improving colour iris segmentation using a model selection technique

Yang Hu , Konstantinos Sirlantzis, Gareth Howells

PII: S0167-8655(15)00008-2

DOI: 10.1016/j.patrec.2014.12.012

Reference: PATREC 6149

To appear in: Pattern Recognition Letters

Received date: 19 April 2014

Accepted date: 9 December 2014

Please cite this article as: Yang Hu , Konstantinos Sirlantzis, Gareth Howells, Improving colour iris segmentation using a model selection technique, Pattern Recognition Letters (2015), doi: 10.1016/j.patrec.2014.12.012

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN

US

CR

IP

T

Pattern Recognition Letters

Authorship Confirmation

Please save a copy of this file, complete and upload as the “Confirmation of Authorship” file.

As corresponding author I, Yang Hu, hereby confirm on behalf of all authors that: 1. This manuscript, or a large part of it, has not been published, was not, and is not being submitted to any other journal. 2. If presented at or submitted to or published at a conference(s), the conference(s) is (are) identified and substantial justification for re-publication is presented below. A copy of conference paper(s) is(are) uploaded with the manuscript. 3. If the manuscript appears as a preprint anywhere on the web, e.g. arXiv, etc., it is identified below. The preprint should include a statement that the paper is under consideration at Pattern Recognition Letters. 4. All text and graphics, except for those marked with sources, are original works of the authors, and all necessary permissions for publication were secured prior to submission of the manuscript. 5. All authors each made a significant contribution to the research reported and have read and approved the submitted manuscript.

SignatureYang Hu Date21 Nov 2014

List any pre-prints:

Relevant Conference publication(s) (submitted, accepted, or published): [1] Yang Hu, Konstantinos Sirlantzis, Gareth Howells. A robust algorithm for colour iris segmentation based on 1-norm regression. Accepted by International Joint Conference on Biometrics (IJCB), 2014.

Justification for re-publication:

This paper is an extension of the work in [1] with significantly new contributions. The differences are: (1) One more iris segmentation model (referred as `1-E) is proposed and investigated in this special issue submission. (2) The performance of all iris segmentation models is analysed in detail. (3) A model selection method is proposed, which significantly improves segmentation robustness and accuracy in comparison to existing methods. (4) Experiments are conducted on mobile device captured iris images and compared with results obtained from standard colour iris datasets.

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN

US

CR

IP

T

Research Highlights • Analysis of circle and ellipse based iris segmentation models. • A novel model selection method to improve colour iris segmentation. • Showing the effectiveness of HOG feature for model selection. • Analysis of the experimental results on both mobile and static camera data.

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN

US

CR

IP

T 1

Pattern Recognition Letters journal homepage: www.elsevier.com

Improving colour iris segmentation using a model selection technique

Yang Hua,∗∗, Konstantinos Sirlantzisa, Gareth Howellsa aSchool of Engineering and Digital Arts, Jennison Building, University of Kent, Canterbuty, CT2 7NT, UK

ABSTRACT

In this paper, we propose a novel method to improve the reliability and accuracy of colour iris segmentation for captures both from static and mobile devices. Our method is a fusion strategy based on selection among the segmentation outcomes of different segmentation methods or models. First, we present and analyse an iris segmentation framework which uses three different models to show that improvements can be obtained by selection among the outcomes generated by the three models. Then, we introduce a model selection method which defines the optimal segmentation based on a ring-shaped region around the outer segmentation boundary identified by each model. We use the

Histogram of Oriented Gradients (HOG) as features extracted from the ring-shaped region, and train a SVM-based classifier which provides the selection decision. Experiments on colour iris datasets, captured by mobile devices and static camera, show that the proposed method achieves an improve performance compared to the individual iris segmentation models and existing algorithms. c© 2015 Elsevier Ltd. All rights reserved. 1. Introduction

Iris recognition has been one of the most reliable techniques for biometric authentication due to the inherent stability, randomness and high degree of freedom of iris pattern. As a fundamental step of iris recognition, iris segmentation is an important prerequisite for iris recognition systems. The pioneering work by Daugman (Daugman, 1993) shows the effectiveness of an integro-differential operator for near-infrared (NIR) iris images captured in controlled environment. Following Daugman, a number of iris segmentation algorithms (Wildes, 1997; Daugman, 2007; Shah and Ross, 2009; Miyazawa et al., 2008; He et al., 2009; Ryan et al., 2008) have been proposed.

Despite excellent performance, the aforementioned algorithms are difficult to be deployed on mobile devices, such as smart phones, tablets and pads. The reason is that these algorithms are less effective for colour iris images captured by mobile devices. Compared with NIR images, colour iris images are much noisier due to specular reflection. Additionally, mobile devices usually work in less constraint environment, which leads to more noise factors such as illumination variance, eyelids occlusion and motion blur. Therefore, it is necessary to investigate an effective colour iris segmentation method. ∗∗Corresponding author: Tel.: +44-1227-824412; fax: +44-1227-456084; e-mail: yh94@kent.ac.uk (Yang Hu)