• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2011.07a
• /
• pp.39-41
• /
• 2011

본 논문에서는 깊이 영상의 해상도와 품질을 향상시키는 방법을 제안한다. 일반적으로 2D-plus-Depth 구조의 3D 콘텐츠에서는 깊이 영상의 품질이 매우 중요하다. 최근 들어 Time-of-Flight (TOF) 방식의 깊이 센서가 깊이 영상 획득에 많이 사용되고 있는데 TOF 깊이 센서가 제공하는 깊이 영상은 저해상도이기 때문에 고해상도 3D 콘텐츠를 제작하기 위해서는 깊이 영상의 해상도를 상향 변환하는 것이 필수적이다. 또한 고품질의 깊이 영상을 얻기 위해서는 물체 간의 경계를 정교하게 보존하는 것이 중요하다. 최근에는 깊이 영상의 해상도 상향 변환을 위해서 Joint Bilateral Upsampling(JBU) 방식이 많이 사용되고 있다. 본 논문은 깊이 영상의 해상도를 높임에 있어서 우선 보간법을 수행하여 영상의 상향 변환 시에 생긴 빈 홀들의 값을 채워준 후 Bilateral Filtering을 수행함으로써 성능을 높인다. 일반적으로 영상을 상향 변환을 할 때 다양한 방법들이 있는데 본 논문에서는 Nearest Neighborhood(NN), Gaussian과 경계 보존 보간법, 경계 보존 보간법과 Fast Curvature Based Interpolation(FCBI)를 결합한 보간법을 사용하였다. 실험 결과 제안 방법이 기존 방법보다 우수한 성능을 가짐을 보여준다. 또한 경계 보존 보간법과 FCBI를 결합한 보간법을 이용해서 상향 변환을 수행한 결과가 다른 보간법들에 의한 결과보다 우수하다는 점을 알 수 있다.

• Journal of the Society of Naval Architects of Korea
• /
• v.44 no.5
• /
• pp.517-525
• /
• 2007

This paper presents how to approximate an optimal shape of roll bending process in the fabrication of a curved shell plate. The roll bending process usually makes the cylindrical or conic shape from an initial flat plate. It means that the final shape is developable or its surface representation has zero Gaussian curvature. The fabrication shape is important in order to find process parameters of roil bending. An optimal concept is used to determine the developable fabrication shape which is in the closest proximity to the design surface or the given shell plate and is subject to developability. The results and the efficiency of this algorithm are evaluated by applying to some shell plates. Furthermore, the fabrication shape will be fundamental information for other process parameters of roll bending such as the vertical displacement of the center roller and the rolling directions.

• The KIPS Transactions:PartB
• /
• v.11B no.3
• /
• pp.297-302
• /
• 2004

In this paper, the 3-D shape description for the objects with the cone ridge and valley surfaces, and the corresponding threshold value selection for surface classification are considered. The existing method based on the mean and Gaussian curvatures(H and K) of differential geometries cannot properly describe cone primitives, which are some of the most common objects in the real world. Also the existing method for surface classification based on the sign values of H and K has Problems in practical applications. For this, cone surface shapes are classified cone ridges and cone valleys are derived from surfaces using the fact that H values are constant case of cylinder surfaces and variable for cone surfaces, respectively. Also threshold value selection for surface classification from a statistical point of view is proposed. The effectiveness of the proposed methods are verified through experiments.

• Korean Journal of Optics and Photonics
• /
• v.17 no.2
• /
• pp.149-158
• /
• 2006

We have analyzed various tolerances of the multi-configurative microscopic system for inspecting the wire-bonding of a reed frame by using the Gaussian bracket method and the equivalent lens method. The tolerances for the curvature and the thickness, which are axial symmetric tolerances, are given by varying the back focal length within a fecal depth under diffraction-limited conditions. Moreover, by using the trial and error method, the axial non-symmetric tolerances for decenter and tilt are established by assigning the 5% variation of MTF(modulation transfer function) at the spatial frequency of 50 lp/mm and at the field angle of 0.7 field. As the tolerances with the most probable distribution are distributed within the range of the decay rate of less than 5% independent of the probability distribution of tolerances, we can achieve completely the desired design performances of the multi-configurative microscopic system by using the various ranges of these tolerances.

• Korean Journal of Optics and Photonics
• /
• v.17 no.5
• /
• pp.420-429
• /
• 2006

In the zoom lens of digital still cameras with the variation of the image plane generated by various symmetric error factors such as curvature, thickness and refractive index error of each lens surface about the optic axis, we induce a theoretical condition to fix constantly the image plane by translating the compensator group of the zoom lens by using the Gaussian bracket. We confirm the validity of this condition by using three examples of general zoom lens types with 3, 4, and 5 groups, respectively. When these error factors are randomly changed within the range of tolerance according to the Monte Carlo method, we verify that the distributions of the degree of moving of the compensator are normal distributions at three zoom lens types. From capability analysis using these results, we theoretically propose the method estimating the standard deviation, that is, sigma-level, as a function of the maximum movement of the compensator.