• Journal of Korea Multimedia Society
• /
• v.24 no.2
• /
• pp.178-185
• /
• 2021

Three-dimensional high-resolution magnetic resonance imaging (MRI) provides fine-level anatomical information for disease diagnosis. However, there is a limitation in obtaining high resolution due to the long scan time for wide spatial coverage. Therefore, in order to obtain a clear high-resolution(HR) image in a wide spatial coverage, a super-resolution technology that converts a low-resolution(LR) MRI image into a high-resolution is required. In this paper, we propose a super-resolution technique through filter learning based on information on the surrounding gradient information in 3D space from 3D MRI images. In the learning step, the gradient features of each voxel are computed through eigen-decomposition from 3D patch. Based on these features, we get the learned filters that minimize the difference of intensity between pairs of LR and HR images for similar features. In test step, the gradient feature of the patch is obtained for each voxel, and the filter is applied by selecting a filter corresponding to the feature closest to it. As a result of learning 100 T1 brain MRI images of HCP which is publicly opened, we showed that the performance improved by up to about 11% compared to the traditional interpolation method.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.12C
• /
• pp.1194-1200
• /
• 2006

In this paper, we propose an adaptive MAP (Maximum A Posteriori) high-resolution image reconstruction algorithm using local statistics. In order to preserve the edge information of an original high-resolution image, a visibility function defined by local statistics of the low-resolution image is incorporated into MAP estimation process, so that the local smoothness is adaptively controlled. The weighted non-quadratic convex functional is defined to obtain the optimal solution that is as close as possible to the original high-resolution image. An iterative algorithm is utilized for obtaining the solution, and the smoothing parameter is updated at each iteration step from the partially reconstructed high-resolution image is required. Experimental results demonstrate the capability of the proposed algorithm.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2009.10a
• /
• pp.1338-1340
• /
• 2009

High resolution three-dimensional integral imaging display is proposed. Each time-multiplexed image is projected with different incident angle on same array of elemental lenses. Those images are collected at different positions in focal plane of lens array, and thus the number of the point light sources increases and their spacing decreases. Therefore, proposed method can create high resolution 3D images.

• Atmosphere
• /
• v.26 no.4
• /
• 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.

• Journal of KIISE:Software and Applications
• /
• v.34 no.3
• /
• pp.266-274
• /
• 2007

This paper proposes a new reconstruction method of high-resolution facial image from a low-resolution facial image based on top-down machine learning and recursive error back-projection. A face is represented by a linear combination of prototypes of shape and that of texture. With the shape and texture information of each pixel in a given low-resolution facial image, we can estimate optimal coefficients for a linear combination of prototypes of shape and those that of texture by solving least square minimizations. Then high-resolution facial image can be obtained by using the optimal coefficients for linear combination of the high-resolution prototypes. In addition, a recursive error back-projection procedure is applied to improve the reconstruction accuracy of high-resolution facial image. The encouraging results of the proposed method show that our method can be used to improve the performance of the face recognition by applying our method to reconstruct high-resolution facial images from low-resolution images captured at a distance.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.5
• /
• pp.85-92
• /
• 2022

In this paper, we propose a technique that can express low-resolution hair and fur simulations in high-resolution without noise using ConvNet and geometric images of strands in the form of lines. Pairs between low-resolution and high-resolution data can be obtained through physics-based simulation, and a low-resolution-high-resolution data pair is established using the obtained data. The data used for training is used by converting the position of the hair strands into a geometric image. The hair and fur network proposed in this paper is used for an image synthesizer that upscales a low-resolution image to a high-resolution image. If the high-resolution geometry image obtained as a result of the test is converted back to high-resolution hair, it is possible to express the elastic movement of hair, which is difficult to express with a single mapping function. As for the performance of the synthesis result, it showed faster performance than the traditional physics-based simulation, and it can be easily executed without knowing complex numerical analysis.

• The Bulletin of The Korean Astronomical Society
• /
• v.44 no.1
• /
• pp.52.1-52.1
• /
• 2019

The Helioseismic and Magnetic Imager (HMI) is the instrument of Solar Dynamics Observatory (SDO) to study the magnetic field and oscillation at the solar surface. The HMI image is not enough to analyze very small magnetic features on solar surface since it has a spatial resolution of one arcsec. Super resolution is a technique that enhances the resolution of a low resolution image. In this study, we use a method for enhancing the solar image resolution using a Deep-learning model which generates a high resolution HMI image from a low resolution HMI image (4 by 4 binning). Deep learning networks try to find the hidden equation between low resolution image and high resolution image from given input and the corresponding output image. In this study, we trained a model based on a very deep residual channel attention networks (RCAN) with HMI images in 2014 and test it with HMI images in 2015. We find that the model achieves high quality results in view of both visual and measures: 31.40 peak signal-to-noise ratio(PSNR), Correlation Coefficient (0.96), Root mean square error (RMSE) is 0.004. This result is much better than the conventional bi-cubic interpolation. We will apply this model to full-resolution SDO/HMI and GST magnetograms.

• Korean Journal of Remote Sensing
• /
• v.27 no.6
• /
• pp.717-728
• /
• 2011

An efficient method for the unsupervised classification of high resolution imagery is suggested in this paper. It employs pixel-linking and merging based on the adjacency graph. The proposed algorithm uses the neighbor lines of 8 directions to include information in spatial proximity. Two approaches are suggested to employ neighbor lines in the linking. One is to compute the dissimilarity measure for the pixel-linking using information from the best lines with the smallest non. The other is to select the best directions for the dissimilarity measure by comparing the non-homogeneity of each line in the same direction of two adjacent pixels. The resultant partition of pixel-linking is segmented and classified by the merging based on the regional and spectral adjacency graphs. This study performed extensive experiments using simulation data and a real high resolution data of IKONOS. The experimental results show that the new approach proposed in this study is quite effective to provide segments of high quality for object-based analysis and proper land-cover map for high resolution imagery of urban area.

• Proceedings of the KSRS Conference
• /
• 2007.10a
• /
• pp.468-470
• /
• 2007

This paper is to present an approach for generating coarse resolution (MODIS data) fraction images of forested region in Korea peninsula using forest type area fraction derived from high resolution data (ASTER data) in regional forest area. A 15-m spatial resolution multi-spectral ASTER image was acquired under clear sky conditions on September 22, 2003 over the forested area near Seoul, Korea and was used to select each end-member that represent a pure reflectance of component of forest such as different forest, bare soil and water. The area fraction of selected each end-member and a 500-m spatial resolution MODIS reflectance product covering study area was applied to a linear mixture inversion model for calculating the fraction image of forest component across the South Korea. We found that the area fraction values of each end-member observed from high resolution image data could be used to separate forest cover in low resolution image data.

• Proceedings of The Korean Society of Agricultural and Forest Meteorology Conference
• /
• 2016.09a
• /
• pp.25-28
• /
• 2016

Recently, because of the weather forecasts through the low-resolution data has been limited, the demand of the high-resolution data is sharply increasing. Therefore, in this study, we restore the ultra-high resolution synthetic precipitation and temperature data for 2000-2014 due to small-scale topographic effect using the QPM (Quantitative Precipitation Model)/QTM (Quantitative Temperature Model). First, we reproduce the detailed precipitation and temperature data with 1km resolution using the distribution of Automatic Weather System (AWS) data and Automatic Synoptic Observation System (ASOS) data, which is about 10km resolution with irregular grid over South Korea. Also, we recover the precipitation and temperature data with 1km resolution using the MERRA reanalysis data over North Korea, because there are insufficient observation data. The precipitation and temperature from restored current climate reflect more detailed topographic effect than irregular AWS/ASOS data and MERRA reanalysis data over the Korean peninsula. Based on this analysis, more detailed prospect of regional climate is investigated.