• 한국광학회지
• /
• 제19권2호
• /
• pp.95-102
• /
• 2008

패럴랙스배리어의 장점은 2D와 3D 콘텐츠를 자동적으로 스위칭 하여 디스플레이 할 수 있다는 것이다. 그러나 관찰자가 움직일 경우 배리어를 통과하는 영상이 바뀌게 되며, 수평으로 어느 범위 이상을 움직이게 되면 영상이 역전되어 다른 영상을 보게 된다. 이러한 제한점을 해결하고자 시점추적이나 시점을 증가시키는 것과 같은 무안경 스테레오 디스플레이가 소개되고 있다. 본 논문에서는 일정한 위치에서만 관찰할 수 있는 고정식 패럴랙스배리어의 단점을 보완할 수 있는 가변형 패럴랙스배리어 방식을 제안하였으며, 가변형 배리어의 제작을 위하여 640 라인 FPC로 구성된 새로운 교차커넥터를 설계하였다. 또한, 일반적으로 사용되는 웹 카메라를 이용하여 시점추적시스템을 구현하였고 실험 결과를 보였다.

• 한국축산식품학회지
• /
• 제33권2호
• /
• pp.198-204
• /
• 2013

This study aimed to apply the Parallux system to detect (fluoro)quinone antibiotics residues in raw bovine milk. The immunogen enabled the generation of a specific antiserum with a titer of 1/40,000. The $Parallax^{TM}$ kit using the antibody displayed $IC_{50}$ value of 10 to 150 ppb for (fluoro)quinolone antibiotics. $Parallax^{TM}$ kit was also sensitive for the detection of incurred (fluoro)quinolone at Korean Maximum Residual Levels in raw bovine milk as the result of dose response test. Cross reactivities of the antibody with the common (fluoro)quinolones were determined to be norfloxacin, 100%; enrofloxacin, 100%; ciprofloxacin, 100%; danofloxacin, 100%; nalidixic acid, 40%. Lower detection limit (LOD) values of the $Parallax^{TM}$ kit in raw bovine milk were determined to be norfloxacin, 4 ppb; enrofloxacin, 5 ppb; danofloxacin, 5 ppb; ciprofloxacin, 5 ppb and nalidixic acid, 10 ppb. The $Parallax^{TM}$ kit was run 8 times with five different concentrations of norfloxacin to determine the coefficient of variation (CV, %) of intra-assay, which was between 2.7% and 11.8%. To confirm the precision among kit batches for the inter-assay, five different batch kits were tested with 2 different concentration of norfloxacin. The CVs of the inter assay were 4.2% at 50 ppb, and 7.2% at 10 ppb norfloxacin, respectively.

• 천문학논총
• /
• 제30권2호
• /
• pp.119-120
• /
• 2015

We report results of the measurement of the trigonometric parallax of an $H_2O$ maser source in IRAS 22555+6213 with the VLBI Exploration of Radio Astrometry (VERA). The annual parallax was determined to be $0.278{\pm}0.019$ mas, corresponding to a distance of $3.66^{+0.30}_{-0.26}kpc$. Our results confirm that IRAS 22555+6213 is located in the Perseus arm. We computed the peculiar motion of IRAS 22555+6213 to be ($U_{src}$, $V_{src}$, $W_{src}$) = ($0{\pm}1$, $-32{\pm}1$, $9{\pm}1$) $km\;s^{-1}$, where $U_{src}$, $V_{src}$, and $W_{src}$ are directed toward the Galactic center, in the direction of Galactic rotation and toward the Galactic north pole, respectively. IRAS 22555+6213, NGC7538 and Cepheus A lie along the same line of sight, and are within $2^{\circ}$ on the sky. Their parallax distances, with which we derived their absolute position in the Milky Way, show that IRAS 22555+6213 and NGC7538 are associated with the Perseus arm, while Cepheus A is located in the Local arm. We compared the kinematic distances of IRAS 22555+6213 derived with flat and non-flat rotation curve with its parallax distance and found the kinematic distance derived from the non-flat rotation assumption ($-32km\;s^{-1}$ lag) to be consistent with the parallax distance.

• 한국군사과학기술학회지
• /
• 제11권6호
• /
• pp.63-73
• /
• 2008

The helicopter position, heading data and the direction finding data of ESM are essentially required to compensate the parallax and analyze the direction finding accuracy of heliborne ESM in flight test phase. In the case of the long test range compared with small platform like as LYNX helicopter and Jisim Island test site, the parallax compensation for direction finding accuracy calculation and GPS position error can be neglected. In this paper, the direction finding accuracy on the basis of helicopter propeller was calculated by coordinate changing between helicopter and transmitting antenna from WGS84 coordinate to navigation coordinate using helicopter position and direction finding data.

• 한국정보통신학회논문지
• /
• 제10권4호
• /
• pp.712-717
• /
• 2006

본 논문에서는 기존의 집적영상에서 요소 영상을 변형하여 수평시차만을 갖는 새로운 집적영상 기법을 제시하고 광학적으로 구현하였다. 즉, 기존 방식으로 픽업된 2차원 요소 영상에서 한 열의 수평 요소 영상을 수직방향으로 확대하여 전체 요소 영상을 얻고 이를 통해 3차원 입체 영상을 재현하게 된다. 제시하는 수평시차 기반의 집적영상 기법은 수직시차가 제거되어 전송을 위한 요소 영상의 정보량을 크게 줄일 수 있다. 제시된 방식의 유용성을 보이기 위하여 광학적 구현 실험을 수행하고 그 결과를 제시하였다.

• Journal of Information Display
• /
• 제10권4호
• /
• pp.188-194
• /
• 2009

Conventional stereoscopic (3D) displays using binocular parallax generate unnatural conflicts between convergence and accommodation. These conflicts can affect the observer's ability to fuse binocular images and may cause visual fatigue. In this study, time series changes in visual fatigue and depth sensation when viewing stereoscopic images with changing parallax were examined. In particular, the physiological changes, including the subjective symptoms of visual fatigue, when viewing five parallax conditions, were examined. Then a comparative analysis of the 2D and 3D conditions was performed based on the visual function. To obtain data regarding the visual function, the time series changes in the spontaneous-blinking rate before and during the viewing of 3D images were measured. The time series change results suggest that 2D and 3D images cause significantly different types of visual fatigue over the range of binocular disparity.

• Current Optics and Photonics
• /
• 제1권2호
• /
• pp.95-100
• /
• 2017

The transmittance of the three dimensional (3D) transparent display is an important factor and can be enhanced by adding a see-through area to the displayed 3D image in order to transmit an ambient light with maximum transparency. However, there is a side effect that the perceived 3D resolution can be degraded due to the see-through area. In this paper, we propose an advanced method to resolve the above trade-off relation between the transparency and the 3D resolution by using an active parallax barrier (PB) with a see-through area. The experimental results are also presented to prove the proposed principle.

• Journal of the Optical Society of Korea
• /
• 제16권2호
• /
• pp.121-126
• /
• 2012

We propose a novel technique that records the horizontal-parallax-only (HPO) hologram of a real object using optical scanning holography (OSH). The proposing HPO-OSH is composed of a conventional OSH and an electronics low pass filter. When we scan an object along vertical direction before horizontal direction, the electronic low pass filter filters the vertical fringes with preserving horizontal fringes and gives an HPO hologram. To the best of our knowledge, this is the first time to record the HPO hologram using OSH without either truncation of the scanning beam or digital post processing.

• 한국정보디스플레이학회:학술대회논문집
• /
• 한국정보디스플레이학회 2009년도 9th International Meeting on Information Display
• /
• pp.1099-1102
• /
• 2009

Conventional stereoscopic (3D) displays using binocular parallax generate unnatural conflicts between convergence and accommodation. Those conflicts can affect the ability to fuse binocular images and may cause visual fatigue. This study examined time-series changes in visual fatigue and depth sensation while viewing stereoscopic images with changing parallax. We examined the physiological changes, including the subjective symptoms of visual fatigue, when viewing five parallax conditions. The time-series results suggest that 2D and 3D images produce significantly different types of visual fatigue over the range of binocular disparity.

• 대한임베디드공학회논문지
• /
• 제16권5호
• /
• pp.179-186
• /
• 2021

For stereo matching based on deep learning, the design of network structure is crucial to the calculation of matching cost, and the time-consuming problem of convolutional neural network in image processing also needs to be solved urgently. In this paper, a method of stereo matching using sparse loss volume in parallax dimension is proposed. A sparse 3D loss volume is constructed by using a wide step length translation of the right view feature map, which reduces the video memory and computing resources required by the 3D convolution module by several times. In order to improve the accuracy of the algorithm, the nonlinear up-sampling of the matching loss in the parallax dimension is carried out by using the method of multi-category output, and the training model is combined with two kinds of loss functions. Compared with the benchmark algorithm, the proposed algorithm not only improves the accuracy but also shortens the running time by about 30%.