Feature of malignant melanoma based on color information

Hisao Motoyama, Toshiyuki Tanaka, Masaru Tanaka, Hiroshi Oka

研究成果: Conference contribution

6 引用 (Scopus)

抄録

This research aims at finding the features that contribute to diagnose the illness as the malignant melanoma, based on the color information of images. We divide RGB space into 4096 pieces and choose the spaces including colors peculiar to malignant melanoma. By using these spaces, about 26% of images of melanoma could be taken out only by the color information. This is very useful to malignant melanoma distinction accuracy.

元の言語English
ホスト出版物のタイトルProceedings of the SICE Annual Conference
ページ399-402
ページ数4
出版物ステータスPublished - 2004
イベントSICE Annual Conference 2004 - Sapporo, Japan
継続期間: 2004 8 42004 8 6

Other

OtherSICE Annual Conference 2004
Japan
Sapporo
期間04/8/404/8/6

Fingerprint

Color

ASJC Scopus subject areas

  • Engineering(all)

これを引用

Motoyama, H., Tanaka, T., Tanaka, M., & Oka, H. (2004). Feature of malignant melanoma based on color information. : Proceedings of the SICE Annual Conference (pp. 399-402). [WPI-3-1]

Feature of malignant melanoma based on color information. / Motoyama, Hisao; Tanaka, Toshiyuki; Tanaka, Masaru; Oka, Hiroshi.

Proceedings of the SICE Annual Conference. 2004. p. 399-402 WPI-3-1.

研究成果: Conference contribution

Motoyama, H, Tanaka, T, Tanaka, M & Oka, H 2004, Feature of malignant melanoma based on color information. : Proceedings of the SICE Annual Conference., WPI-3-1, pp. 399-402, SICE Annual Conference 2004, Sapporo, Japan, 04/8/4.
Motoyama H, Tanaka T, Tanaka M, Oka H. Feature of malignant melanoma based on color information. : Proceedings of the SICE Annual Conference. 2004. p. 399-402. WPI-3-1
Motoyama, Hisao ; Tanaka, Toshiyuki ; Tanaka, Masaru ; Oka, Hiroshi. / Feature of malignant melanoma based on color information. Proceedings of the SICE Annual Conference. 2004. pp. 399-402
@inproceedings{993b0864ce3b43fbabebbed1449246a3,
title = "Feature of malignant melanoma based on color information",
abstract = "This research aims at finding the features that contribute to diagnose the illness as the malignant melanoma, based on the color information of images. We divide RGB space into 4096 pieces and choose the spaces including colors peculiar to malignant melanoma. By using these spaces, about 26{\%} of images of melanoma could be taken out only by the color information. This is very useful to malignant melanoma distinction accuracy.",
keywords = "Dermoscopy, Melanoma, RGB",
author = "Hisao Motoyama and Toshiyuki Tanaka and Masaru Tanaka and Hiroshi Oka",
year = "2004",
language = "English",
pages = "399--402",
booktitle = "Proceedings of the SICE Annual Conference",

}

TY - GEN

T1 - Feature of malignant melanoma based on color information

AU - Motoyama, Hisao

AU - Tanaka, Toshiyuki

AU - Tanaka, Masaru

AU - Oka, Hiroshi

PY - 2004

Y1 - 2004

N2 - This research aims at finding the features that contribute to diagnose the illness as the malignant melanoma, based on the color information of images. We divide RGB space into 4096 pieces and choose the spaces including colors peculiar to malignant melanoma. By using these spaces, about 26% of images of melanoma could be taken out only by the color information. This is very useful to malignant melanoma distinction accuracy.

AB - This research aims at finding the features that contribute to diagnose the illness as the malignant melanoma, based on the color information of images. We divide RGB space into 4096 pieces and choose the spaces including colors peculiar to malignant melanoma. By using these spaces, about 26% of images of melanoma could be taken out only by the color information. This is very useful to malignant melanoma distinction accuracy.

KW - Dermoscopy

KW - Melanoma

KW - RGB

UR - http://www.scopus.com/inward/record.url?scp=12744273430&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=12744273430&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:12744273430

SP - 399

EP - 402

BT - Proceedings of the SICE Annual Conference

ER -