25권 5호 555-562 2017 [한국자동차공학회 논문집 ]
| 제목 | 원형 광패턴을 활용한 능동방식 3차원 좌표 복원 알고리즘 |
|---|---|
| 분야 | 전기ㆍ전자ㆍ통신 |
| 언어 | Korean |
| 저자 | 이덕우(영산대학교) |
| Key Words | 3D coordinates reconstruction(3차원 좌표 복원), Structured light patterns(구조광 패턴), Concentric circular patterns(동심원 패턴), 3D camera(3차원 카메라) 3D object(3차원 물체) |
| 초록 |
The measurement of the 3D coordinates of an object surface from a 2D image has been significant in the areas of computer vision, pattern recognition, and machine learning, among others. In particular, it has gained much attention in practical fields, such as 3D camera system, automotive vision system, etc. This paper suggests an approach for the reconstruction of a 3D object surface from a 2D image that was captured by a camera. The camera captures the object overlaid with a set of concentric, circular-shaped light patterns. In the experiment, the geometric information of the light patterns, and the intrinsic and extrinsic parameters provide a mathematical model to achieve a 3D reconstruction result. The reconstruction is carried out on the basis of the relationship between the projected circular patterns and the deformed patterns by a 3D object surface. This paper primarily deals with establishing a mathematical proof of concept for 3D reconstruction by using structured circular light patterns. |
| 원문(PDF) | 다운로드 |