• 대기
• /
• 제26권4호
• /
• pp.697-709
• /
• 2016

This study estimates and evaluates the extreme value of 30 m-resolution daily maximum and minimum temperatures over South Korea, using inverse distance weighting (IDW), parameter-elevation regression on independent slopes model (PRISM) and generalized extreme value (GEV) method. The three experiments are designed and performed to find the optimal estimation strategy to obtain extreme value. First experiment (EXP1) applies GEV firstly to automated surface observing system (ASOS) to estimate extreme value and then applies IDW to produce high-resolution extreme values. Second experiment (EXP2) is same as EXP1, but using PRISM to make the high-resolution extreme value instead of IDW. Third experiment (EXP3) firstly applies PRISM to ASOS to produce the high-resolution temperature field, and then applies GEV method to make high resolution extreme value data. By comparing these 3 experiments with extreme values obtained from observation data, we find that EXP3 shows the best performance to estimate extreme values of maximum and minimum temperatures, followed by EXP1 and EXP2. It is revealed that EXP1 and EXP2 have a limitation to estimate the extreme value at each grid point correctly because the extreme values of these experiments with 30 m-resolution are calculated from only 60 extreme values obtained from ASOS. On the other hand, the extreme value of EXP3 is similar to observation compared to others, since EXP3 produces 30m-resolution daily temperature through PRISM, and then applies GEV to that result at each grid point. This result indicates that the quality of statistically produced high-resolution extreme values which are estimated from observation data is different depending on the combination and procedure order of statistical methods.

• 대기
• /
• 제22권1호
• /
• pp.1-12
• /
• 2012

The Atmospheric motion vectors (AMVs) derived using infrared (IR) channel imagery of geostationary satellites have been utilized widely for real-time weather analysis and data assimilation into global numerical prediction model. As the horizontal resolution of sensors on-board satellites gets higher, it becomes possible to identify atmospheric motions induced by convective clouds ($meso-{\beta}$ and $meso-{\gamma}$ scales). The National Institute of Meteorological Research (NIMR) developed the high resolution visible (HRV) AMV algorithm to detect mesoscale atmospheric motions including ageostrophic flows. To retrieve atmospheric motions smaller than $meso-{\beta}$ scale effectively, the target size is reduced and the visible channel imagery of geostationary satellite with 1 km resolution is used. For the accurate AMVs, optimal conditions are decided by investigating sensitivity of algorithm to target selection and correction method of height assignment. The results show that the optimal conditions are target size of 32 km ${\times}$ 32 km, the grid interval as same as target size, and the optimal target selection method. The HRV AMVs derived with these conditions depict more effectively tropical cyclone OMAIS than IR AMVs and the mean speed of HRV AMVs in OMAIS is slightly faster than that of IR AMVs. Optimized mesoscale AMVs are derived for 6 months (Feb. 2010-Jun. 2010) and validated with radiosonde observations, which indicates NIMR's HRV AMV algorithm can retrieve successfully mesoscale atmospheric motions.

• 한국음향학회지
• /
• 제22권1호
• /
• pp.35-41
• /
• 2003

일반적인 음성인식 방법에서는 주파수 전대역에서 추출한 특징벡터를 사용하므로, 각 주파수 부대역은 최종인식 결과에 동등하게 기여한다. 본 논문에서는 주파수 부대역별로 독립적인 특징을 추출하고, 음성인식에 효과적이 되도록 부대역의 켑스트럼 해상도를 조절하는 방법을 제안한다. 주파수 부대역별로 독립적인 특징을 추출하는 멀티밴드 음성인식접근을 사용하여 부대역 특징벡터의 차원을 변화시킨다. 최적의 벡터 차원 조합을 찾기 위하여 음성인식률과 군집화 품질을 사용한다. TIDIGITS 연결 숫자음을 사용한 실험결과에서, 제안한 방법은 전대역 특징추출에 비해 적은 계산량으로도 숫자열 인식률은 99.12%, 백분율 정확도 (percent correct)는 99.775%, 백분율 정밀도 (percent accuracy)는 99.705%를 얻었으며, 이는 전대역 특징벡터에 비해 상대적 오류율을 각각 38%, 32%, 37% 감소시킨 결과이다.

• 한국산학기술학회논문지
• /
• 제18권2호
• /
• pp.590-599
• /
• 2017

국내에서 하수관로 탐사장치는 200만 화소 이상의 고해상도 디지털 카메라를 이용한 제품이 개발되어 있으나 30만 화소 이하의 장치가 대부분 사용되고 있다. 특히, 10만화소 이하의 장치가 아직도 많이 사용되고 있어, 영상처리를 위한 환경이 매우 열악하다. 본 연구에서 다루는 하수관 영상은 매우 저해상도($240{\times}320$ = 76,800화소)로 균열탐지가 매우 어렵다. 국내에서 이러한 저해상도 하수관거 영상이 대부분이기 때문에, 이를 연구대상으로 선택하였다. 이러한 저해상도 영상으로 하수 관거의 균열을 자동으로 탐지하는 기법을 디지털 영상처리 기술을 이용하여 연구하였다. 총8단계를 거쳐 균열을 자동으로 탐지하는 프로그램을 개발하였으며, 기본적으로 Matlab 프로그램의 함수를 이용하였다. 2단계에서 최적의 임계값을 찾는 알고리즘과 5단계에서 균열을 판단하는 알고리즘을 개발하였다. 2단계는 자막이 흰색이기 때문에 자막이 없는 원래 영상보다 Otsu's 임계값(threshold)이 높게 계산이 되는 점에 착안하여 Otsu 임계값을 시작으로 0.01씩 감소시키면서 최적의 임계값을 찾는 방법 알고리즘이며, 5단계는 길이가 10mm(40픽셀) 이상이고 폭이 1mm(4픽셀) 이상으로 판단하여, 균열을 탐지하는 알고리즘이다. 해석 결과 매우 저해상도 영상임에도 불구하고 균열 탐지 결과가 우수한 것으로 판단된다.

• Imaging Science in Dentistry
• /
• 제32권2호
• /
• pp.75-79
• /
• 2002

Purpose : To evaluate the effect of exposure time and image resolution on fractal dimension calculations for determining the optimal range of these two variances. Materials and Methods : Thirty-one radiographs of the mandibular angle area of sixteen human dry mandibles were taken at different exposure times (0.01, 0.08, 0.16, 0.25, 0.40, 0.64, and 0.80 s). Each radiograph was digitized at 1200 dpi, 8 bit, 256 gray level using a film scanner. We selected an Region of Interest (ROI) that corresponded to the same region as in each radiograph, but the resolution of ROI was degraded to 1000, 800, 600, 500, 400, 300, 200, and 100 dpi. The fractal dimension was calculated by using the tile-counting method for each image, and the calculated values were then compared statistically. Results: As the exposure time and the image resolution increased, the mean value of the fractal dimension decreased, except the case where exposure time was set at 0.01 seconds (α = 0.05). The exposure time and image resolution affected the fractal dimension by interaction (p<0.001). When the exposure time was set to either 0.64 seconds or 0.80 seconds, the resulting fractal dimensions were lower, irrespective of image resolution, than at shorter exposure times (α = 0.05). The optimal range for exposure time and resolution was determined to be 0.08- 0.40 seconds and from 400-1000 dpi, respectively. Conclusion : Adequate exposure time and image resolution is essential for acquiring the fractal dimension using tile-counting method for evaluation of the mandible.

• 대한원격탐사학회지
• /
• 제33권1호
• /
• pp.25-35
• /
• 2017

Spatial downscaling with fine resolution auxiliary variables has been widely applied to predict precipitation at fine resolution from coarse resolution satellite-based precipitation products. The spatial downscaling framework is usually based on the decomposition of precipitation values into trend and residual components. The fine resolution auxiliary variables contribute to the estimation of the trend components. The main focus of this study is on quantitative analysis of impacts of trend component estimates on predictive performance in spatial downscaling. Two regression models were considered to estimate the trend components: multiple linear regression (MLR) and geographically weighted regression (GWR). After estimating the trend components using the two models,residual components were predicted at fine resolution grids using area-to-point kriging. Finally, the sum of the trend and residual components were considered as downscaling results. From the downscaling experiments with time-series Tropical Rainfall Measuring Mission (TRMM) 3B43 precipitation data, MLR-based downscaling showed the similar or even better predictive performance, compared with GWR-based downscaling with very high explanatory power. Despite very high explanatory power of GWR, the relationships quantified from TRMM precipitation data with errors and the auxiliary variables at coarse resolution may exaggerate the errors in the trend components at fine resolution. As a result, the errors attached to the trend estimates greatly affected the predictive performance. These results indicate that any regression model with high explanatory power does not always improve predictive performance due to intrinsic errors of the input coarse resolution data. Thus, it is suggested that the explanatory power of trend estimation models alone cannot be always used for the selection of an optimal model in spatial downscaling with fine resolution auxiliary variables.

• 한국ITS학회 논문지
• /
• 제22권5호
• /
• pp.137-147
• /
• 2023

LiDAR는 자율 주행뿐만 아니라 다양한 산업 현장에 적용되어 대상의 크기와 거리를 측정하는 데 사용되고 있다. 이에 더하여 이 센서는 반사된 빛의 양을 바탕으로 반사 강도 영상 또한 제공한다. 이는 측정 대상의 형상에 대한 정보를 제공하여 센서 데이터 처리에 긍정적인 효과를 일으킨다. LiDAR는 고해상도가 될수록 높은 성능을 보장하지만 이는 센서 비용의 증가를 야기하는데, 이 점은 반사 강도 영상에도 해당된다. 높은 해상도의 반사 강도 영상을 취득하기 위해서는 고가의 장비 사용이 필수적이다. 따라서 본 연구에서는 저해상도의 반사 강도 영상을 고해상도의 영상으로 개선하는 인공지능을 개발하였다. 이를 위해서 본 연구에서는 최적의 초해상화 신경망 모델을 위한 파라미터 분석을 수행하였다. 또한, 초해상화 알고리즘을 2,500여 장의 반사 강도 영상에 적용하여 훈련과 검증을 하였다. 결과적으로 반사 강도 영상의 해상도를 향상시켰다. 바라건대 본 연구의 결과가 향후 자율 주행 분야에 적용되어 주행환경 인식과 장애물 탐지 성능 향상에 기여할 수 있기를 기대하는 바이다.

• Journal of the Korean Statistical Society
• /
• 제35권2호
• /
• pp.201-212
• /
• 2006

The technique of foldover is usually used by experimenters to de-alias the effects that are interesting in follow-up experiment. Employing a $2^{k-p}$ design with resolution III or higher, Li and Lin (2003) developed an algorithm and used computer programs to search its corresponding optimal foldover design for selected 16-run and 32-run experiments. Based on the minimum aberration criterion, the strong combined-optimal design, defined by Li and Lin, is the better choice of the initial design. In this article, we apply the technique of blocking to find the strong combined-optimal designs. Furthermore, we will tabulate all 16-run and 32-run strong combined-optimal designs and their corresponding core foldover plans for practical use. Some new designs that have not appeared in the other literature but constructed by the technique of blocking are also proposed in this article.

• 지구물리와물리탐사
• /
• 제7권1호
• /
• pp.19-24
• /
• 2004

Frequency-space domain full-wave tomography is a promising technique for delineating detailed subsurface structure with high resolution. However, this method requires criteria for the selection of a set of optimal temporal frequency components, to achieve stability in the sequence of inversion processes together with computational efficiency. We propose a method of selecting optimal temporal frequencies, based on wavenumber continuity. The proposed method is tested numerically and is shown to be able to select an optimal set of frequency components that are sufficient to image the anomalies.

• Journal of the Optical Society of Korea
• /
• 제9권1호
• /
• pp.26-31
• /
• 2005

We describe the design and the implementation of video-rate reflection confocal scanning microscopy (CSM) using an acousto-optical deflector (AOD) for the fast horizontal scan and a galvanometer mirror (GM) for the slow vertical scan. Design parameters of the optical system are determined for optimal resolution and contrast. The OSLO simulations show that the performances of CSM are not changed with deflection angle and the wavefront errors of the system are less than 0.012λ. To evaluate the performances of designed CSM, we do a series of tests, measuring lateral and axial resolution, real time image acquisition. Due to a higher axial resolution compared with conventional microscopy, CSM can detect the surface of sub-micrometer features. We detect 138㎚ line shape pattern with a video-rate (30 frm/sec). And 10㎚ axial resolution is archived. The lateral resolution of the topographic images will be further enhanced by differential confocal microscopy (DCM) method and computational algorithms.