Detailed Information

Cited 0 time in webofscience Cited 0 time in scopus
Metadata Downloads

트랜스포머 블록과 윤곽선 디코더를 활용한 딥러닝 기반의 피부 병변 분할 방법

Full metadata record
DC Field Value Language
dc.contributor.author김지훈-
dc.contributor.author박경리-
dc.contributor.author김해문-
dc.contributor.author문영식-
dc.date.accessioned2023-05-03T09:42:30Z-
dc.date.available2023-05-03T09:42:30Z-
dc.date.issued2022-04-
dc.identifier.issn2234-4772-
dc.identifier.issn2288-4165-
dc.identifier.urihttps://scholarworks.bwise.kr/erica/handle/2021.sw.erica/112749-
dc.description.abstract전문의는 피부암을 조기에 발견하기 위해 피부경을 사용하여 진단하지만 다양한 형태로 인해 피부 병변을 판단하는 데 어려움이 있다. 최근 높은 성능을 보인 딥러닝을 이용한 피부 병변 분할 방법이 제안되었지만 피부와 피부 병변 경계가 명확하지 않아서 피부 병변을 분할하는 데 문제점이 있었다. 이러한 문제를 개선하기 위해 제안하는 방법은 효과적으로 피부 병변을 분할하기 위해 트랜스포머 블록을 구성하였으며, 네트워크의 각 계층마다 윤곽선 디코더를 구성하여 피부 병변을 자세히 분할하였다. 실험 결과, 제안하는 방법은 기존의 방법보다 Dice coefficient 기준 0.041 ~ 0.071, Jaccard Index 기준 0.067 ~ 0.112의 성능 향상을 보인다.-
dc.description.abstractSpecialists diagnose skin cancer using a dermatoscopy to detect skin cancer as early as possible, but it is difficult to determine accurate skin lesions because skin lesions have various shapes. Recently, the skin lesion segmentation method using deep learning, which has shown high performance, has a problem in segmenting skin lesions because the boundary between healthy skin and skin lesions is not clear. To solve these issues, the proposed method constructs a transformer block to effectively segment the skin lesion, and constructs an edge decoder for each layer of the network to segment the skin lesion in detail. Experiment results have shown that the proposed method achieves a performance improvement of 0.041 ~ 0.071 for Dic Coefficient and 0.062 ~ 0.112 for Jaccard Index, compared with the previous method.-
dc.format.extent8-
dc.language한국어-
dc.language.isoKOR-
dc.publisher한국정보통신학회-
dc.title트랜스포머 블록과 윤곽선 디코더를 활용한 딥러닝 기반의 피부 병변 분할 방법-
dc.title.alternativeDeep Learning based Skin Lesion Segmentation Using Transformer Block and Edge Decoder-
dc.typeArticle-
dc.publisher.location대한민국-
dc.identifier.doi10.6109/jkiice.2022.26.4.533-
dc.identifier.bibliographicCitation한국정보통신학회논문지, v.26, no.4, pp 533 - 540-
dc.citation.title한국정보통신학회논문지-
dc.citation.volume26-
dc.citation.number4-
dc.citation.startPage533-
dc.citation.endPage540-
dc.identifier.kciidART002839015-
dc.description.isOpenAccessN-
dc.description.journalRegisteredClasskci-
dc.subject.keywordAuthor피부 병변 분할-
dc.subject.keywordAuthor피부 병변-
dc.subject.keywordAuthor윤곽선 디코더-
dc.subject.keywordAuthor트랜스포머 블록-
dc.subject.keywordAuthor딥러닝-
dc.subject.keywordAuthorSkin lesion segmentation-
dc.subject.keywordAuthorSkin lesion-
dc.subject.keywordAuthorEdge decoder-
dc.subject.keywordAuthorTransformer block-
dc.subject.keywordAuthorDeep learning-
dc.identifier.urlhttps://koreascience.kr/article/JAKO202212462665743.page-
Files in This Item
Go to Link
Appears in
Collections
COLLEGE OF COMPUTING > SCHOOL OF COMPUTER SCIENCE > 1. Journal Articles

qrcode

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.

Altmetrics

Total Views & Downloads

BROWSE