Graph-based skin lesion segmentation of multispectral dermoscopic images

Olivier Lézoray 1 Marinette Revenu 1 Michel Desvignes 2
1 Equipe Image - Laboratoire GREYC - UMR6072
GREYC - Groupe de Recherche en Informatique, Image, Automatique et Instrumentation de Caen
2 GIPSA-AGPIG - AGPIG
GIPSA-DIS - Département Images et Signal
Abstract : Accurate skin lesion segmentation is critical for automated early skin cancer detection and diagnosis. We present a novel method to detect skin lesion borders in multispectral der-moscopy images. First, hairs are detected on infrared images and removed by inpainting visible spectrum images. Second, skin lesion is pre-segmented using a clustering of a superpixel partition. Finally, the pre-segmentation is globally regular-ized at the superpixel level and locally regularized in a narrow band at the pixel level.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [18 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01080036
Contributor : Yvain Queau <>
Submitted on : Tuesday, November 4, 2014 - 12:09:30 PM
Last modification on : Thursday, February 7, 2019 - 5:27:14 PM
Document(s) archivé(s) le : Thursday, February 5, 2015 - 10:40:21 AM

File

Lezoray_ICIP14.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01080036, version 1

Citation

Olivier Lézoray, Marinette Revenu, Michel Desvignes. Graph-based skin lesion segmentation of multispectral dermoscopic images. 21st IEEE International Conference on Image Processing (ICIP 2014), Oct 2014, Paris, France. ⟨hal-01080036⟩

Share

Metrics

Record views

707

Files downloads

342