Implementation of unsupervised statistical methods for low-quality iris segmentation

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper, we explore the use of advanced statistical models for unsupervised segmentation of challenging eye images. A previous work has shown the superiority of Triplet Markov Field (TMF) over HMF for segmenting challenging eye region but TMF implementation is computationally very expensive. To enable faster processing while preserving performance, we investigate in this paper Hidden Markov Chain (HMC) and Pair wise Markov Chain (PMC). We developed novel adequate image scanning procedures and initialization steps for implementing these models and extensive experiments on challenging images of the ICE2005 database show that the use of HMC with the snail scan and Histogram Initialization enhances the quality of segmentation comparing to OSIRIS-V4 based on contour approach or TMF model.

Original languageEnglish
Title of host publicationProceedings - 10th International Conference on Signal-Image Technology and Internet-Based Systems, SITIS 2014
EditorsRichard Chbeir, Richard Chbeir, Kokou Yetongnon, Albert Dipanda
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages566-573
Number of pages8
ISBN (Electronic)9781479979783
DOIs
Publication statusPublished - 1 Jan 2014
Event10th International Conference on Signal-Image Technology and Internet-Based Systems, SITIS 2014 - Marrakech, Morocco
Duration: 23 Nov 201427 Nov 2014

Publication series

NameProceedings - 10th International Conference on Signal-Image Technology and Internet-Based Systems, SITIS 2014

Conference

Conference10th International Conference on Signal-Image Technology and Internet-Based Systems, SITIS 2014
Country/TerritoryMorocco
CityMarrakech
Period23/11/1427/11/14

Keywords

  • Hidden Markov Chain
  • Pairewise Markov Chain
  • unsupervised segmentation challenging eye image

Fingerprint

Dive into the research topics of 'Implementation of unsupervised statistical methods for low-quality iris segmentation'. Together they form a unique fingerprint.

Cite this