ETRI-Knowledge Sharing Plaform

ENGLISH

성과물

논문 검색
구분 SCI
연도 ~ 키워드

상세정보

학술지 Horizon Detection in Maritime Images using Scene Parsing Network
Cited 18 time in scopus Download 14 time Share share facebook twitter linkedin kakaostory
저자
정치윤, 양현승, 문경덕
발행일
201806
출처
Electronics Letters, v.54 no.12, pp.760-762
ISSN
0013-5194
출판사
IET
DOI
https://dx.doi.org/10.1049/el.2018.0989
협약과제
18PS1200, 자율운항 선박을 위한 운항관제 인공지능 시스템 원천기술 개발, 문경덕
초록
A method for horizon detection in maritime scenes using a scene parsing network is proposed. First, each pixel from an input image is segmented into corresponding semantic categories using a scene parsing network, which relies on a deep neural network. Then, the boundary information related to the horizon and the sea is extracted. Scene segmentation allows the proposed method to identify the horizon, regardless of whether the boundary between the sea and sky is smooth or blurry, or whether the image contains many line elements like the horizon. Moreover, least squares and median filtering are iteratively used to retrieve an accurate estimation of the horizon line. Experimental results demonstrate the superior accuracy of the proposed method to identify the horizon when compared to state-of-the-art methods.
KSP 제안 키워드
Deep neural network(DNN), Least Squares(LS), Median Filtering, Scene Parsing, Scene Segmentation, Semantic categories, accurate estimation, boundary information, horizon detection, state-of-The-Art