Implementation of unsupervised statistical methods for low-quality iris segmentation

Meriem Yahiaoui (1, 2, 3, 4) , Emmanuel Monfrini (5, 2, 3, 4) , Bernadette Dorizzi (5, 1, 2, 4)

1 TSP - EPH - Département Electronique et Physique
2 SAMOVAR - Services répartis, Architectures, MOdélisation, Validation, Administration des Réseaux
3 TSP - CITI - Communications, Images et Traitement de l'Information
4 CNRS - Centre National de la Recherche Scientifique
5 TIPIC-SAMOVAR - Traitement de l'Information Pour Images et Communications

In this paper, we explore the use of advanced statistical models for unsupervised segmentation of challeng- ing 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 compu- tationally very expensive. To enable faster processing while preserving performance, we investigate in this paper Hidden Markov Chain (HMC) and Pairwise 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 Intialization enhances the quality of segmentation comparing to OSIRIS-V4 based on contour approach or TMF model

Mots clés: Hidden Markov Chain; Pairwise Markov chain; Unsupervised segmentation challenging eye image

Domaines: Traitement du signal et de l'image [eess.SP]

Meriem Yahiaoui, Emmanuel Monfrini, Bernadette Dorizzi. Implementation of unsupervised statistical methods for low-quality iris segmentation. SITIS 2014 : 10th International Conference on Signal-Image Technology and Internet-Based Systems, Nov 2014, Marrakech, Morocco. pp.566 - 573, ⟨10.1109/SITIS.2014.46⟩. ⟨hal-01331567⟩