Automatic classification of skin lesions using color mathematical morphology-based texture descriptors

Victor Gonzalez-Castro 1, 2, 3 Johan Debayle 1, 2, 3, * Yanal Wazaefi 4 Mehdi Rahim 4 Caroline Gaudy-Marqueste 5, 6 Jean-Jacques Grob 5, 6 Bernard Fertil 4
* Auteur correspondant
1 LGF-ENSMSE - Laboratoire Georges Friedel
2 PROPICE-ENSMSE - Département PROcédés Poudres, Interfaces, Cristallisation et Ecoulements
3 SPIN-ENSMSE - Centre Sciences des Processus Industriels et Naturels
4 LSIS - Laboratoire des Sciences de l'Information et des Systèmes
5 Département de dermatologie, CHU Marseille
6 TIMONE - Hôpital de la Timone [CHU - APHM]
Abstract : In this paper an automatic classification method of skin lesions from dermoscopic images is proposed. This method is based on color texture analysis based both on color mathematical morphology and Kohonen Self-Organizing Maps (SOM), and it does not need any previous segmentation process. More concretely, mathematical morphology is used to compute a local descriptor for each pixel of the image, while the SOM is used to cluster them and, thus, create the texture descriptor of the global image. Two approaches are proposed, depending on whether the pixel descriptor is computed using classical (i.e. spatially invariant) or adaptive (i.e. spatially variant) mathematical morphology by means of the Color Adaptive Neighborhoods (CANs) framework. Both approaches obtained similar areas under the ROC curve (AUC): 0.854 and 0.859 outperforming the AUC built upon dermatologists' predictions (0.792).
Keywords : Mathematical morphology Self-organizing maps Color texture description Melanoma Color adaptive neighborhoods
Type de document :
Communication dans un congrès
Fabrice Meriaudeau; Olivier Aubreton. Twelfth International Conference on Quality Control by Artificial Vision, Jun 2015, Le Creusot, France. SPIE.Digital Library, 9534 (953409), pp.[9534-3] ; doi:10.1117/12.2182592, 2015, SPIE proceedings - Medical Imaging and Biomedical imaging I. 〈http://proceedings.spiedigitallibrary.org/proceeding.aspx?articleid=2290076〉
Liste complète des métadonnées

Littérature citée [13 références]  Voir  Masquer  Télécharger
https://hal-emse.ccsd.cnrs.fr/emse-01163688
Contributeur : Fatima Lillouch <>
Soumis le : lundi 15 juin 2015 - 12:53:05
Dernière modification le : mardi 23 octobre 2018 - 14:36:10
Document(s) archivé(s) le : mardi 25 avril 2017 - 08:04:16

Fichier

V Gonzalez-Castro J Debayle QC...
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : emse-01163688, version 1

Collections

EMSE | UNIV-TLN | INSTITUT-TELECOM | LSIS | UNIV-AMU | LGF-ENSMSE | PROPICE-ENSMSE | LIS-LAB | UNIV-HESAM

Citation

Victor Gonzalez-Castro, Johan Debayle, Yanal Wazaefi, Mehdi Rahim, Caroline Gaudy-Marqueste, et al.. Automatic classification of skin lesions using color mathematical morphology-based texture descriptors. Fabrice Meriaudeau; Olivier Aubreton. Twelfth International Conference on Quality Control by Artificial Vision, Jun 2015, Le Creusot, France. SPIE.Digital Library, 9534 (953409), pp.[9534-3] ; doi:10.1117/12.2182592, 2015, SPIE proceedings - Medical Imaging and Biomedical imaging I. 〈http://proceedings.spiedigitallibrary.org/proceeding.aspx?articleid=2290076〉. 〈emse-01163688〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

639

Téléchargements de fichiers

519

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD