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
- 이은주
- 발행연도
- 2018
- 저작권
- 한밭대학교 논문은 저작권에 의해 보호받습니다.
이용수0
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
This dissertation presents a method to perform accurate camera self-calibration based on image quality assessment. We suggest the use of only high-quality images for the camera self-calibration process. We define high-quality images as images that contain little or no blur and have a high contrast level. Assuming constant noise, we perform image quality assessment using pre-processing techniques such as histogram analysis to measure the contrast level and statistical analysis such as energy and standard deviation of high-frequency components of discrete cosine transform. We used maximum energy and standard deviation to define a sharp image and eliminate blur images. Additionally, we suggest the use of Scale Invariant Feature Transform (SIFT) for the detection of good features, which are invariant to scale, illumination and rotation. The good feature descriptors that have minimum distances are used to compute homography projection matrix to get the good perspective view of images. Our experimental results show less or no distortion in the perspective view of images. The suggested method achieves accurate image perspective views, comparable to conventional camera self-calibration method with SIFT features.
목차
1. Introduction 11.1 Motivation 11.2 Related Research 21.3 Environmental Variables 41.4 Suggestion 41.5 Organization of the Dissertation 62. The Camera Model and State of Art 82.1 Projective Geometry 82.1.1 Points and Homogeneous Coordinates 82.1.2 The Projective Plane 92.1.3 The Projective Space 112.2 The Camera Model 132.2.1 Intrinsic Parameters 132.2.2 Extrinsic Parameters 162.3 State of Art 172.3.1 Camera Calibration 182.3.2 Camera Self-calibration 183. Accurate Camera Self Calibration based on Image Quality Assessment 203.1 Image Quality Assessment 233.1.1 Contrast Assessment by Histogram Analysis 233.1.2 Blur Detection based on Discrete Cosine Transform 253.2 Camera Self-calibration 283.2.1 Feature Detection 293.2.1.1 Harris Corner Detector 303.2.1.2 SIFT Feature Detector 343.2.2 Feature Matching 373.2.2.1 NCC (Normalized Cross Correlation) 373.3 RANSAC algorithm for getting Homography matrix 383.3.1 Direct Linear Transformation 383.3.2 RANSAC algorithm 404. Experiments and Discussion 435. Conclusion 49References 50
최근 본 자료
댓글(0)
0