ETRI-Knowledge Sharing Plaform

ENGLISH

성과물

논문 검색
구분 SCI
연도 ~ 키워드

상세정보

학술대회 The Analysis of Image Acquisition Method for Anti-UAV Surveillance using Cameras Image
Cited 4 time in scopus Download 4 time Share share facebook twitter linkedin kakaostory
저자
김영일, 여건민, 박성희, 정운철, 송순용, 허태욱
발행일
202010
출처
International Conference on Information and Communication Technology Convergence (ICTC) 2020, pp.549-554
DOI
https://dx.doi.org/10.1109/ICTC49870.2020.9289164
협약과제
20NR1100, 소음 및 영상신호 결합기반 무인기 검출 기술 개발, 김영일
초록
As the number of cases where human life and property are threatened by illegal use of UAV is rapidly increasing, a technology for detecting illegal UAV using deep learning algorithm is being researched. In order for the deep learning algorithm to work, an appropriate UAV object image size and quality are required. Unfortunately, in detecting a small UAV using a camera sensor, when the object image of the camera becomes small, the detection distance of the UAV is reduced due to the difficulty of detecting the UAV. Therefore, a large number of cameras are required to expand the detection distance. This paper suggests the methodology for configuring UAV surveillance area for image acquisition in terms of reducing the required number of cameras. Four methods for constructing the UAV surveillance area considering the optical limitations of the camera are proposed, and compared the proposed methods using a simple thin lens camera model.
키워드
Deep Learning, UAV detection, UAV surveillance area
KSP 제안 키워드
Acquisition method, Camera model, Camera sensor, Detection distance, Object image, Small UAVs, Thin lens, UAV detection, UAV surveillance, deep learning(DL), deep learning algorithm