Novel iris segmentation method - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2012

Novel iris segmentation method

Résumé

Iris recognition is a proven, accurate means to identify people. Iris is a part of an eye image. To be possible the uses of this modality, it's indispensable to be detected and segmented. In this paper, we present our both methods: eye image preprocessing by method of Bias-Corrected Fuzzy C-Mean (BCFCM) and iris segmentation based on the active contours "Snake". The performance of iris recognition system highly depends on segmentation step. For instance, even an effective feature extraction method would not be able to obtain useful information from an iris image that is not segmented properly. This paper presents a straightforward method for segmenting the iris patterns. To prove the performance of our iris method segmentation, we have integrated it in an iris verification system. Experiments are performed using iris images obtained from CASIA V.1 database.
Fichier non déposé

Dates et versions

hal-02113701 , version 1 (29-04-2019)

Identifiants

  • HAL Id : hal-02113701 , version 1

Citer

Hanêne Guesmi, Hanène Trichili, Adel M. Alimi, Basel Solaiman. Novel iris segmentation method. ICMCS'12: 3rd IEEE International Conference Multimedia Computing and Systems, May 2012, Tanger, Morocco. pp.260 - 265. ⟨hal-02113701⟩
20 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More