AVM을 위한 비선형 패턴 기반 어안 카메라 교정 = Nonlinear-Pattern Based Fisheye Camera Calibration for AVM
Recently, around view monitor(AVM), which prevent accidents between vehicles or with other objects(e.g., persons or animals) by utilizing a wide-angle camera, have been developed to provide the driver with information about the vehicle’s surrounding area that is difficult to detect, in advance. The core part of this system is camera calibration, which removes the projective transformations and nonlinear transformations that occur inevitably in the images acquired by a camera that uses the fisheye lens. Camera calibration usually estimates the intrinsic and extrinsic parameters that were previously unknown. This requires the features existing in the pre-known real world 3D space. Further, images including these features acquired by the camera to be calibrated are required.
The features of the conventional calibrators utilize lines, circular dots, and chessboards. In these calibrators, the features are arranged to be equidistant between features(EBF) throughout the calibrator area. In the conventional chessboard calibrator, the features are all identical black and white squares, and the distances between the immediately adjacent feature points(DbFP) are constant. A feature point is where the corners of two black squares meet. All identical squares transform all the different curvilinear quadrilaterals into the calibrator’s image acquired by the fisheye camera. Further, DbFP are not constant but decrease nonlinearly from the center to the periphery. Often, the corners of two black curvilinear quadrilaterals near the periphery do not meet at a unique point due to aliasing effect of imaging sensor. Thus, it is quite difficult to detect the exact locations of the feature points, and even if they were detected, the locations would be inaccurate. In such cases, it would not be possible to calibrate the camera fully automatically, and even if such a calibration were performed, its performance would be degraded. Recently, H-pattern planar calibrator that utilizes the parking lines on a parking space has been developed to calibrate the fisheye camera of the vehicle AVM. However, the intrinsic parameters were not estimated, and the Hough transform, which is known to have a heavy burden of computational complexity and long processing time, was used for estimating the vanishing points in order to determine the camera’s extrinsic parameters. Futhermore, when the heights from ground to the fisheye cameras mounted on a vehicle are different each other, additional efforts should be required to adjust their scale difference among calibrated images.
To solve these problems, this thesis proposes a new nonlinear-pattern based planar calibrator for an easy detection of the feature points in a calibrator’s image acquired by a fisheye camera. The features of the proposed calibrator are not squares but curvilinear quadrilaterals having different sizes. Further, the DbFP are not constant but increase nonlinearly from the center to the periphery. Thus, the proposed calibrator becomes a non-equidistant between features(NeBF) calibrator. Interestingly, all different curvilinear quadrilaterals transform almost all identical squares into the calibrator’s image acquired by the fisheye camera. Moreover, the DbFP become constant, and all the corners of the two black squares always meet exactly at unique points in the calibrator’s image. Accordingly, all the feature points in the image can be easily and correctly detected without any miss or false detections, despite the aliasing effect of imaging sensor. Further, the optimum number and locations of feature points on the proposed calibrator were statistically determined for ensuring the best camera calibration performance. The results were used for developing a calibrator to calibrate the fisheye camera of the vehicle AVM.
The performances of the camera calibration of the proposed NeBF calibrator, the conventional EBF chessboard calibrator, and H-pattern calibrator were compared and analyzed. As a result of the evaluation, the proposed NeBF calibrator was found to be superior to the conventional chessboard calibrator and H-pattern calibrator in terms of the ease of feature point detection, location accuracy of the detected feature points, and calibration performance of the camera.
서지정보 내보내기(Export)
닫기소장기관 정보
닫기권호소장정보
닫기오류접수
닫기오류 접수 확인
닫기음성서비스 신청
닫기음성서비스 신청 확인
닫기이용약관
닫기학술연구정보서비스 이용약관 (2017년 1월 1일 ~ 현재 적용)
학술연구정보서비스(이하 RISS)는 정보주체의 자유와 권리 보호를 위해 「개인정보 보호법」 및 관계 법령이 정한 바를 준수하여, 적법하게 개인정보를 처리하고 안전하게 관리하고 있습니다. 이에 「개인정보 보호법」 제30조에 따라 정보주체에게 개인정보 처리에 관한 절차 및 기준을 안내하고, 이와 관련한 고충을 신속하고 원활하게 처리할 수 있도록 하기 위하여 다음과 같이 개인정보 처리방침을 수립·공개합니다.
주요 개인정보 처리 표시(라벨링)
목 차
3년
또는 회원탈퇴시까지5년
(「전자상거래 등에서의 소비자보호에 관한3년
(「전자상거래 등에서의 소비자보호에 관한2년
이상(개인정보보호위원회 : 개인정보의 안전성 확보조치 기준)개인정보파일의 명칭 | 운영근거 / 처리목적 | 개인정보파일에 기록되는 개인정보의 항목 | 보유기간 | |
---|---|---|---|---|
학술연구정보서비스 이용자 가입정보 파일 | 한국교육학술정보원법 | 필수 | ID, 비밀번호, 성명, 생년월일, 신분(직업구분), 이메일, 소속분야, 웹진메일 수신동의 여부 | 3년 또는 탈퇴시 |
선택 | 소속기관명, 소속도서관명, 학과/부서명, 학번/직원번호, 휴대전화, 주소 |
구분 | 담당자 | 연락처 |
---|---|---|
KERIS 개인정보 보호책임자 | 정보보호본부 김태우 | - 이메일 : lsy@keris.or.kr - 전화번호 : 053-714-0439 - 팩스번호 : 053-714-0195 |
KERIS 개인정보 보호담당자 | 개인정보보호부 이상엽 | |
RISS 개인정보 보호책임자 | 대학학술본부 장금연 | - 이메일 : giltizen@keris.or.kr - 전화번호 : 053-714-0149 - 팩스번호 : 053-714-0194 |
RISS 개인정보 보호담당자 | 학술진흥부 길원진 |
자동로그아웃 안내
닫기인증오류 안내
닫기귀하께서는 휴면계정 전환 후 1년동안 회원정보 수집 및 이용에 대한
재동의를 하지 않으신 관계로 개인정보가 삭제되었습니다.
(참조 : RISS 이용약관 및 개인정보처리방침)
신규회원으로 가입하여 이용 부탁 드리며, 추가 문의는 고객센터로 연락 바랍니다.
- 기존 아이디 재사용 불가
휴면계정 안내
RISS는 [표준개인정보 보호지침]에 따라 2년을 주기로 개인정보 수집·이용에 관하여 (재)동의를 받고 있으며, (재)동의를 하지 않을 경우, 휴면계정으로 전환됩니다.
(※ 휴면계정은 원문이용 및 복사/대출 서비스를 이용할 수 없습니다.)
휴면계정으로 전환된 후 1년간 회원정보 수집·이용에 대한 재동의를 하지 않을 경우, RISS에서 자동탈퇴 및 개인정보가 삭제처리 됩니다.
고객센터 1599-3122
ARS번호+1번(회원가입 및 정보수정)