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

Victor Gonzalez-Castro; Johan Debayle; Yanal Wazaefi; Mehdi Rahim; Caroline Gaudy-Marqueste; Jean Jacques Grob; Bernard Fertil

doi:10.1117/12.2182592

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

Victor Gonzalez-Castro, Johan Debayle^*, Yanal Wazaefi, Mehdi Rahim, Caroline Gaudy-Marqueste, Jean Jacques Grob, Bernard Fertil

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

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).

Original language	English
Title of host publication	Proceedings of SPIE - The International Society for Optical Engineering
Publisher	SPIE
Volume	9534
ISBN (Print)	9781628416992
DOIs	https://doi.org/10.1117/12.2182592
Publication status	Published - 2015
Event	12th International Conference on Quality Control by Artificial Vision - Le Creusot, France Duration: 3 Jun 2015 → 5 Jun 2015

Conference

Conference	12th International Conference on Quality Control by Artificial Vision
Country/Territory	France
City	Le Creusot
Period	3/06/15 → 5/06/15

Keywords / Materials (for Non-textual outputs)

Color adaptive neighborhoods
Color texture description
Mathematical morphology
Melanoma
Self-organizing maps

Access to Document

10.1117/12.2182592

Cite this

@inproceedings{1020c5a77fc44bac99cea9b0f6c3eb92,

title = "Automatic classification of skin lesions using color mathematical morphology-based texture descriptors",

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 = "Color adaptive neighborhoods, Color texture description, Mathematical morphology, Melanoma, Self-organizing maps",

author = "Victor Gonzalez-Castro and Johan Debayle and Yanal Wazaefi and Mehdi Rahim and Caroline Gaudy-Marqueste and Grob, {Jean Jacques} and Bernard Fertil",

year = "2015",

doi = "10.1117/12.2182592",

language = "English",

isbn = "9781628416992",

volume = "9534",

booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

publisher = "SPIE",

address = "United States",

note = "12th International Conference on Quality Control by Artificial Vision ; Conference date: 03-06-2015 Through 05-06-2015",

}

Gonzalez-Castro, V, Debayle, J, Wazaefi, Y, Rahim, M, Gaudy-Marqueste, C, Grob, JJ & Fertil, B 2015, Automatic classification of skin lesions using color mathematical morphology-based texture descriptors. in Proceedings of SPIE - The International Society for Optical Engineering. vol. 9534, 953409, SPIE, 12th International Conference on Quality Control by Artificial Vision, Le Creusot, France, 3/06/15. https://doi.org/10.1117/12.2182592

TY - GEN

T1 - Automatic classification of skin lesions using color mathematical morphology-based texture descriptors

AU - Gonzalez-Castro, Victor

AU - Debayle, Johan

AU - Wazaefi, Yanal

AU - Rahim, Mehdi

AU - Gaudy-Marqueste, Caroline

AU - Grob, Jean Jacques

AU - Fertil, Bernard

PY - 2015

Y1 - 2015

N2 - 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).

AB - 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).

KW - Color adaptive neighborhoods

KW - Color texture description

KW - Mathematical morphology

KW - Melanoma

KW - Self-organizing maps

UR - http://www.scopus.com/inward/record.url?scp=84931310848&partnerID=8YFLogxK

U2 - 10.1117/12.2182592

DO - 10.1117/12.2182592

M3 - Conference contribution

AN - SCOPUS:84931310848

SN - 9781628416992

VL - 9534

BT - Proceedings of SPIE - The International Society for Optical Engineering

PB - SPIE

T2 - 12th International Conference on Quality Control by Artificial Vision

Y2 - 3 June 2015 through 5 June 2015

ER -

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

Abstract

Conference

Keywords / Materials (for Non-textual outputs)

Access to Document

Other files and links

Fingerprint

Cite this