지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학위논문
- 저자정보
- 지도교수
- 조정훈
- 발행연도
- 2018
- 저작권
- 경북대학교 논문은 저작권에 의해 보호받습니다.
이용수5
이 논문의 연구 히스토리 (2)
- 이 논문의 후속연구가 궁금하신가요?
- 연관 학술논문 또는 학술발표를 통해 보다 발전된 연구결과를 확인하실 수 있습니다.
- 이 논문의 연구 히스토리 확인하기
초록· 키워드
오류제보하기
The research trend of the ADAS / autonomous camera sensor is developing a technology in which one camera performs multiple functions at the same time. In order to process multiple functions simultaneously in a micro-embedded system such as a black box, the real time image processing is the key issue. For this purpose, it is very important to apply a high-speed chip and efficient software algorithm. The proposed adaptive ROI is an efficient algorithm which minimizes unnecessary processing area and dynamically selects the optimal ROI position for the object to be detected.
Through the proposed method, we confirmed the performance improvement of vehicle black box based lane detection and vehicle detection algorithm. The adaptive ROI of lane detection based on vanishing point is confirmed the improvement on detection rate in various driving environment. Also, the adaptive region of interest for vehicle detection time has been further reduced by adaptively resetting the region of interest through the vehicle information detected in the previous frame. Especially, in case of detection of forward vehicle, the average processing speed is improved to 3.5 times compared with the existing area of interest, with an average of 14.35 fps. This algorithm has low implementation complexity and it is applied to system using low-end processor vehicle black box achieving real time processing performance of ADAS system.
Through the proposed method, we confirmed the performance improvement of vehicle black box based lane detection and vehicle detection algorithm. The adaptive ROI of lane detection based on vanishing point is confirmed the improvement on detection rate in various driving environment. Also, the adaptive region of interest for vehicle detection time has been further reduced by adaptively resetting the region of interest through the vehicle information detected in the previous frame. Especially, in case of detection of forward vehicle, the average processing speed is improved to 3.5 times compared with the existing area of interest, with an average of 14.35 fps. This algorithm has low implementation complexity and it is applied to system using low-end processor vehicle black box achieving real time processing performance of ADAS system.
목차
I. 서 론 …… 1II. 관련 연구 …… 42.1 첨단 운전자 보조 시스템 …… 42.2 블랙박스 시스템 …… 62.3 관심영역 설정 …… 8III. 고정 관심영역의 차선 검출 및 차량검출 …… 103.1 차선 검출 …… 113.2 차량 검출 ……153.3 적응적 관심영역의 필요성 ……20IV. 제안하는 방법 …… 234.1 차선 검출을 위한 적응적 관심영역 …… 254.2 차량 검출을 위한 적응적 관심영역 …… 30V. 실험 및 실험 결과 …… 355.1 실험 환경 및 데이터…… 355.2 실험 방법 및 성능 측정 방법 …… 385.3 실험 결과 …… 39Ⅵ. 결 론 …… 44참고문헌 …… 45Abstract …… 49
최근 본 자료
댓글(0)
0