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
(Guangdong Nanfang Institute of Technology)
저널정보
대한전자공학회 IEIE Transactions on Smart Processing & Computing IEIE Transactions on Smart Processing & Computing Vol.14 No.3
오류 신고하기

피인용 0

검색

    초록·키워드

    In response to the problem of high image similarity in traditional tourism image retrieval, a tourism image retrieval method based on attention mechanism improved deep residual shrinkage network was proposed. The study first utilized deactivation mechanisms and activation functions to improve the deep residual shrinkage network, and applied it to feature extraction in tourism images. Then, based on the improvement of the network, attention mechanism was introduced to construct a retrieval model. The results showed that the average accuracy and consistency of the retrieval model in image dimensionality reduction were 95.18% and 89.55%, respectively. At the same time, the training loss value and image retrieval accuracy in both datasets were 0.21, 0.16, and 94.52% and 90.88%, respectively. And the precision and recall of the image were 92.61% and 96.48%, respectively, which were better than the comparison method. This indicated that the retrieval model could effectively improve the feature representation and retrieval performance of tourism images, while enhancing its robustness to noise and interference. The research aimed to provide a new and effective solution for tourism image retrieval.

    최근 본 자료 전체보기

      UCI(KEPA) : I410-151-25-02-093686240