Color Models for Skin Lesion Classification from Dermatoscopic Images

Iosif Mporas; Isidoros Perikos; Michael Paraskevas

doi:10.1007/978-981-15-1918-5_5

Color Models for Skin Lesion Classification from Dermatoscopic Images

Iosif Mporas, Isidoros Perikos, Michael Paraskevas

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

2 Citations (Scopus)

Abstract

In this paper, we present an architecture for classification of pigmented skin lesions from dermatoscopic images. The architecture is using image preprocessing for natural hair removal and image segmentation for extraction of the skin lesion area followed by computation of statistical values of colors as features. The color-based features were extracted from several well-known and widely used color models. Several classification algorithms were evaluated with the best performing classification algorithm being the AdaBoost with random forest classifier with classification accuracy equal to 73.08% when using RGB-based features only and 74.26% when combining RGB, HSV, and YIQ color model-based features.

Original language	English
Title of host publication	Advances in Integrations of Intelligent Methods - Post-workshop volume of the 8th International Workshop CIMA 2018, in conjunction with IEEE ICTAI 2018
Editors	Ioannis Hatzilygeroudis, Isidoros Perikos, Foteini Grivokostopoulou
Publisher	Springer Nature Link
Pages	85-98
Number of pages	14
ISBN (Print)	9789811519178
DOIs	https://doi.org/10.1007/978-981-15-1918-5_5
Publication status	Published - 1 Jan 2020
Event	8th International Workshop on Combinations of Intelligent Methods and Applications, CIMA 2018 held in conjunction with the 30th IEEE International Conference on Tools with Artificial Intelligence, IEEE ICTAI 2018 - Volos, Greece Duration: 5 Nov 2018 → 7 Nov 2018

Publication series

Name	Smart Innovation, Systems and Technologies
Volume	170
ISSN (Print)	2190-3018
ISSN (Electronic)	2190-3026

Conference

Conference	8th International Workshop on Combinations of Intelligent Methods and Applications, CIMA 2018 held in conjunction with the 30th IEEE International Conference on Tools with Artificial Intelligence, IEEE ICTAI 2018
Country/Territory	Greece
City	Volos
Period	5/11/18 → 7/11/18

Keywords

Color models
Dermatoscopy
Image classification

Access to Document

10.1007/978-981-15-1918-5_5

Cite this

Mporas, I., Perikos, I., & Paraskevas, M. (2020). Color Models for Skin Lesion Classification from Dermatoscopic Images. In I. Hatzilygeroudis, I. Perikos, & F. Grivokostopoulou (Eds.), Advances in Integrations of Intelligent Methods - Post-workshop volume of the 8th International Workshop CIMA 2018, in conjunction with IEEE ICTAI 2018 (pp. 85-98). (Smart Innovation, Systems and Technologies; Vol. 170). Springer Nature Link. https://doi.org/10.1007/978-981-15-1918-5_5

Mporas, Iosif ; Perikos, Isidoros ; Paraskevas, Michael. / Color Models for Skin Lesion Classification from Dermatoscopic Images. Advances in Integrations of Intelligent Methods - Post-workshop volume of the 8th International Workshop CIMA 2018, in conjunction with IEEE ICTAI 2018. editor / Ioannis Hatzilygeroudis ; Isidoros Perikos ; Foteini Grivokostopoulou. Springer Nature Link, 2020. pp. 85-98 (Smart Innovation, Systems and Technologies).

@inproceedings{be54a47ad4b44cb6942cefe763fb5a5e,

title = "Color Models for Skin Lesion Classification from Dermatoscopic Images",

abstract = "In this paper, we present an architecture for classification of pigmented skin lesions from dermatoscopic images. The architecture is using image preprocessing for natural hair removal and image segmentation for extraction of the skin lesion area followed by computation of statistical values of colors as features. The color-based features were extracted from several well-known and widely used color models. Several classification algorithms were evaluated with the best performing classification algorithm being the AdaBoost with random forest classifier with classification accuracy equal to 73.08% when using RGB-based features only and 74.26% when combining RGB, HSV, and YIQ color model-based features.",

keywords = "Color models, Dermatoscopy, Image classification",

author = "Iosif Mporas and Isidoros Perikos and Michael Paraskevas",

year = "2020",

month = jan,

day = "1",

doi = "10.1007/978-981-15-1918-5_5",

language = "English",

isbn = "9789811519178",

series = "Smart Innovation, Systems and Technologies",

publisher = "Springer Nature Link",

pages = "85--98",

editor = "Ioannis Hatzilygeroudis and Isidoros Perikos and Foteini Grivokostopoulou",

booktitle = "Advances in Integrations of Intelligent Methods - Post-workshop volume of the 8th International Workshop CIMA 2018, in conjunction with IEEE ICTAI 2018",

address = "Netherlands",

note = "8th International Workshop on Combinations of Intelligent Methods and Applications, CIMA 2018 held in conjunction with the 30th IEEE International Conference on Tools with Artificial Intelligence, IEEE ICTAI 2018 ; Conference date: 05-11-2018 Through 07-11-2018",

}

Mporas, I, Perikos, I & Paraskevas, M 2020, Color Models for Skin Lesion Classification from Dermatoscopic Images. in I Hatzilygeroudis, I Perikos & F Grivokostopoulou (eds), Advances in Integrations of Intelligent Methods - Post-workshop volume of the 8th International Workshop CIMA 2018, in conjunction with IEEE ICTAI 2018. Smart Innovation, Systems and Technologies, vol. 170, Springer Nature Link, pp. 85-98, 8th International Workshop on Combinations of Intelligent Methods and Applications, CIMA 2018 held in conjunction with the 30th IEEE International Conference on Tools with Artificial Intelligence, IEEE ICTAI 2018, Volos, Greece, 5/11/18. https://doi.org/10.1007/978-981-15-1918-5_5

Color Models for Skin Lesion Classification from Dermatoscopic Images. / Mporas, Iosif; Perikos, Isidoros; Paraskevas, Michael.
Advances in Integrations of Intelligent Methods - Post-workshop volume of the 8th International Workshop CIMA 2018, in conjunction with IEEE ICTAI 2018. ed. / Ioannis Hatzilygeroudis; Isidoros Perikos; Foteini Grivokostopoulou. Springer Nature Link, 2020. p. 85-98 (Smart Innovation, Systems and Technologies; Vol. 170).

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

TY - GEN

T1 - Color Models for Skin Lesion Classification from Dermatoscopic Images

AU - Mporas, Iosif

AU - Perikos, Isidoros

AU - Paraskevas, Michael

PY - 2020/1/1

Y1 - 2020/1/1

N2 - In this paper, we present an architecture for classification of pigmented skin lesions from dermatoscopic images. The architecture is using image preprocessing for natural hair removal and image segmentation for extraction of the skin lesion area followed by computation of statistical values of colors as features. The color-based features were extracted from several well-known and widely used color models. Several classification algorithms were evaluated with the best performing classification algorithm being the AdaBoost with random forest classifier with classification accuracy equal to 73.08% when using RGB-based features only and 74.26% when combining RGB, HSV, and YIQ color model-based features.

AB - In this paper, we present an architecture for classification of pigmented skin lesions from dermatoscopic images. The architecture is using image preprocessing for natural hair removal and image segmentation for extraction of the skin lesion area followed by computation of statistical values of colors as features. The color-based features were extracted from several well-known and widely used color models. Several classification algorithms were evaluated with the best performing classification algorithm being the AdaBoost with random forest classifier with classification accuracy equal to 73.08% when using RGB-based features only and 74.26% when combining RGB, HSV, and YIQ color model-based features.

KW - Color models

KW - Dermatoscopy

KW - Image classification

UR - http://www.scopus.com/inward/record.url?scp=85079091899&partnerID=8YFLogxK

U2 - 10.1007/978-981-15-1918-5_5

DO - 10.1007/978-981-15-1918-5_5

M3 - Conference contribution

AN - SCOPUS:85079091899

SN - 9789811519178

T3 - Smart Innovation, Systems and Technologies

SP - 85

EP - 98

BT - Advances in Integrations of Intelligent Methods - Post-workshop volume of the 8th International Workshop CIMA 2018, in conjunction with IEEE ICTAI 2018

A2 - Hatzilygeroudis, Ioannis

A2 - Perikos, Isidoros

A2 - Grivokostopoulou, Foteini

PB - Springer Nature Link

T2 - 8th International Workshop on Combinations of Intelligent Methods and Applications, CIMA 2018 held in conjunction with the 30th IEEE International Conference on Tools with Artificial Intelligence, IEEE ICTAI 2018

Y2 - 5 November 2018 through 7 November 2018

ER -

Mporas I, Perikos I, Paraskevas M. Color Models for Skin Lesion Classification from Dermatoscopic Images. In Hatzilygeroudis I, Perikos I, Grivokostopoulou F, editors, Advances in Integrations of Intelligent Methods - Post-workshop volume of the 8th International Workshop CIMA 2018, in conjunction with IEEE ICTAI 2018. Springer Nature Link. 2020. p. 85-98. (Smart Innovation, Systems and Technologies). doi: 10.1007/978-981-15-1918-5_5

Color Models for Skin Lesion Classification from Dermatoscopic Images

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this