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
(Sunmoon University) (Sunmoon University) (Sunmoon University) (Sunmoon University) (Sunmoon University)
저널정보
한국정보기술학회 Proceedings of The International Workshop on Future Technology The Proceedings of ADINTECH 2024
오류 신고하기

피인용 0

검색

    초록·키워드

    Reliability and stability of AI models are crucial issues especially in technology fields that deal with human life (e.g., self-driving cars). This study presents a metamorphic testing method for more accurately and effectively assessing the robustness of AI models, considering the unique characteristics of AI models, which require a different approach from traditional software testing methods. Experiments involved various testing purpose of image distortions such as rotation, blurring, brightness adjustment, noise addition, and tilting. We defined and measured the robustness of a given AI model, calculating the robustness value(R-value). This provides a foundation for understanding the stability of AI-based software and for ensuring its safe and reliable use in real-life applications.

    최근 본 자료 전체보기