메뉴 건너뛰기
소속 기관 / 학교 인증
인증하면 논문, 학술자료 등을  무료로 열람할 수 있어요.
한국대학교, 누리자동차, 시립도서관 등 나의 기관을 확인해보세요
(국내 대학 90% 이상 구독 중)
고객센터 ENG
주제분류

논문 기본 정보

저자정보
(Seoul National University of Science and Technology) (Seoul National University of Science and Technology)
저널정보
한국방송·미디어공학회 한국방송미디어공학회 학술발표대회 논문집 한국방송·미디어공학회 2023 추계학술대회
오류 신고하기

피인용 0

검색

    초록·키워드

    Anomaly detection is a challenging problem extensively studied in diverse research and application domains. For example, it plays a crucial role in quality control and industrial inspection, where identifying defects and irregularities in products and processes is critically significant. To this end, in this paper, we introduce a novel approach based on deep learning for identifying anomalies in the MVTec AD dataset under a fully supervised setting. The proposed method involves training multiple convolutional neural networks on different subsets of the dataset and leveraging transfer learning and ensemble training techniques. This ensemble model demonstrates remarkable performance, achieving a macro F1 score of 0.8792.

    최근 본 자료 전체보기