• The KIPS Transactions:PartB
• /
• v.8B no.5
• /
• pp.557-564
• /
• 2001

In general, several moving regions with different motions coexist in a block located on motion boundaries in the block-based motion estimation. In this case the motion compensation error(MCEs) are different with the moving regions. This is inclined to deteriorate the quality of motion compensated images because of the inaccurate motions estimated from the conventional mean absolute error(MAE) based matching function in which the matching error per pixel is accumulate throughout the block. In this paper, we divided a block into the regions according to their motions using the motion information of the spatio-temporally neighboring blocks and calculate the average MCF for each moving mentioned. From the simulation results, we showed the improved performance of the proposed method by comparing the results from other methods such as the full search method and the edge oriented block matching algorithm. Especially, we improved the quality of the motion compensated images of blocks on motion boundaries.

• The KIPS Transactions:PartD
• /
• v.9D no.6
• /
• pp.1009-1016
• /
• 2002

We design and implement a new case-based recommender system using implicit rating information for a digital content site. Our system consists of the User Profile Generation module, the Similarity Evaluation and Recommendation module, and the Personalized Mailing module. In the User Profile Generation Module, we define intra-attribute and inter-attribute weight deriver from own's past interests of a user stored in the access logs to extract individual preferences for a content. A new similarity function is presented in the Similarity Evaluation and Recommendation Module to estimate similarities between new items set and the user profile. The Personalized Mailing Module sends individual recommended mails that are transformed into platform-independent XML document format to users. To verify the efficiency of our system, we have performed experimental comparisons between the proposed model and the collaborative filtering technique by mean absolute error (MAE) and receiver operating characteristic (ROC) values. The results show that the proposed model is more efficient than the traditional collaborative filtering technique.

• Journal of KIISE:Computing Practices and Letters
• /
• v.9 no.5
• /
• pp.521-529
• /
• 2003

In this paper we present an improved method by using demographic information for overcoming the similarity miss-calculation from the sparsity problem in collaborative filtering recommendation systems. The similarity between a pair of users is only determined by the ratings given to co-rated items, so items that have not been rated by both users are ignored. To solve this problem, we add virtual neighbor's rating using demographic information of neighbors for improving prediction accuracy. It is one kind of extentions of traditional collaborative filtering methods using the peason correlation coefficient. We used the Grouplens movie rating data in experiment and we have compared the proposed method with the collaborative filtering methods by the mean absolute error and receive operating characteristic values. The results show that the proposed method is more efficient than the collaborative filtering methods using the pearson correlation coefficient about 9% in MAE and 13% in sensitivity of ROC.

• Korean Journal of Materials Research
• /
• v.32 no.8
• /
• pp.345-353
• /
• 2022

In this study, using deep learning, super-resolution images of transmission electron microscope (TEM) images were generated for nanomaterial analysis. 1169 paired images with 256 × 256 pixels (high resolution: HR) from TEM measurements and 32 × 32 pixels (low resolution: LR) produced using the python module openCV were trained with deep learning models. The TEM images were related to DyVO₄ nanomaterials synthesized by hydrothermal methods. Mean-absolute-error (MAE), peak-signal-to-noise-ratio (PSNR), and structural similarity (SSIM) were used as metrics to evaluate the performance of the models. First, a super-resolution image (SR) was obtained using the traditional interpolation method used in computer vision. In the SR image at low magnification, the shape of the nanomaterial improved. However, the SR images at medium and high magnification failed to show the characteristics of the lattice of the nanomaterials. Second, to obtain a SR image, the deep learning model includes a residual network which reduces the loss of spatial information in the convolutional process of obtaining a feature map. In the process of optimizing the deep learning model, it was confirmed that the performance of the model improved as the number of data increased. In addition, by optimizing the deep learning model using the loss function, including MAE and SSIM at the same time, improved results of the nanomaterial lattice in SR images were achieved at medium and high magnifications. The final proposed deep learning model used four residual blocks to obtain the characteristic map of the low-resolution image, and the super-resolution image was completed using Upsampling2D and the residual block three times.

• Dementia and Neurocognitive Disorders
• /
• v.21 no.4
• /
• pp.138-146
• /
• 2022

Background and Purpose: Magnetic resonance imaging (MRI) helps with brain development analysis and disease diagnosis. Brain volumes measured from different ages using MRI provides useful information in clinical evaluation and research. Therefore, we trained machine learning models that predict the brain age gap of healthy subjects in the East Asian population using T1 brain MRI volume images. Methods: In total, 154 T1-weighted MRIs of healthy subjects (55-83 years of age) were collected from an East Asian community. The information of age, gender, and education level was collected for each participant. The MRIs of the participants were preprocessed using FreeSurfer(https://surfer.nmr.mgh.harvard.edu/) to collect the brain volume data. We trained the models using different supervised machine learning regression algorithms from the scikit-learn (https://scikit-learn.org/) library. Results: The trained models comprised 19 features that had been reduced from 55 brain volume labels. The algorithm BayesianRidge (BR) achieved a mean absolute error (MAE) and r squared (R²) of 3 and 0.3 years, respectively, in predicting the age of the new subjects compared to other regression methods. The results of feature importance analysis showed that the right pallidum, white matter hypointensities on T1-MRI scans, and left hippocampus comprise some of the essential features in predicting brain age. Conclusions: The MAE and R² accuracies of the BR model predicting brain age gap in the East Asian population showed that the model could reduce the dimensionality of neuroimaging data to provide a meaningful biomarker for individual brain aging.

• Atmosphere
• /
• v.21 no.4
• /
• pp.373-389
• /
• 2011

This research focuses on the sensitivity of the WRF(Weather Research and Forecasting) Model according to three different land cover data(USGS(United States Geological Survey), MODIS(Moderate Resolution Imaging Spectroradiometer)30s+USGS, and KLC (Korea Land Cover)) for an event of sea breeze, occurred over the Gangwon Yeongdong region on 13 May 2009. Based on the observation, the easterly into Gangneung, due to the sea-breeze circulation, was identified between 1000 LST and 1640 LST. It did not reach beyond the Taebaek Mountain Range and thus the easterly was not observed near Daegwallyeong. On the other hand, the numerical simulations utilizing land cover data of USGS, MODIS30s+USGS, and KLC showed easterlies beyond the Taebaek Mountain Range up to Daegwallyeong. In addition, rather different penetration distances of each easterly, and different timings of beginning and ending of sea breeze were identified among the simulations. The Bias, MAE(Mean Absolute Error) and RMSE(Root Mean Square Error) of the wind from WRF simulation using MODIS30s+USGS land cover data were the least among the simulations particularly over Gangwon Yeongdong coastal area(Sokcho, Gangneung and Donghae), while those of the wind over the Gangwon Mountain area(Daegwallyeong and Jinbu) from the simulation using KLC land cover data were the least among them. The wind field over Gangwon Yeongdong coastal area from the simulation using USGS land cover data was rather poor among them.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.39 no.12
• /
• pp.953-963
• /
• 2015

There is an increasing need to understand the thermal-hydraulic phenomena, including the critical heat flux (CHF), in narrow rectangular channels and consider these in system design. The CHF mechanism under a saturated flow boiling condition involves the depletion of the liquid film of an annular flow. To predict this type of CHF, the previous representative liquid film dryout models (LFD models) were studied, and their shortcomings were reviewed, including the assumption that void fraction or quality is constant at the boundary condition for the onset of annular flow (OAF). A new LFD model was proposed based on the recent constitutive correlations for the droplet deposition rate and entrainment rate. In addition, this LFD model was applied to predict the CHF in vertical narrow rectangular channels that were uniformly heated. The predicted CHF showed good agreement with 284 pieces of experimental data, with a mean absolute error of 18. 1 % and root mean square error of 22.9 %.

• Structural Engineering and Mechanics
• /
• v.53 no.5
• /
• pp.939-956
• /
• 2015

In this study, in order to propose an efficient model to predict the torque capacity of steel fiber reinforced concrete (SFRC) beams, the existing experimental data related to torsional response of beams is reviewed. It is observed that existing data neglects the effects of some parameters on the variation of torque capacity. Thus, an experimental research was also conducted to obtain the effects of neglected parameters. In the experimental study, a total of seventeen SFRC beams are tested against torsion. The parameters considered in the experiments are concrete compressive strength, steel fiber aspect ratio, volumetric ratio of steel fibers and longitudinal reinforcement ratio. The effect of each parameter is discussed in terms of torque versus unit angle of twist graphs. The data obtained from this experimental research is also combined with the data got from previous studies and employed in artificial neural network (ANN) analysis to estimate the ultimate torque capacity of SFRC beams. In addition to parameters considered in the experiments, aspect ratio of beam cross-section, yield strengths of both transverse and longitudinal reinforcements, and transverse reinforcement ratio are also defined as parameters in ANN analysis due to their significant effects observed in previous studies. Assessment of the accuracy of ANN analysis in estimating the ultimate torque capacity of SFRC beams is performed by comparing the analytical and experimental results. Comparisons are conducted in terms of root mean square error (RMSE), mean absolute error (MAE) and coefficient of efficiency ($E_f$). The results of this study revealed that addition of steel fibers increases the ultimate torque capacity of reinforced concrete beams. It is also found that ANN is a powerful method and a feasible tool to estimate ultimate torque capacity of both normal and high strength concrete beams within the range of input parameters considered.

• Journal of Climate Change Research
• /
• v.6 no.4
• /
• pp.331-344
• /
• 2015

A long-term gridded historical data at 3 km spatial resolution has been generated for practical regional applications such as hydrologic modelling. However, overly high or low values have been found at some grid points where complex topography or sparse observational network exist. In this study, the Inverse Distance Weighting (IDW) method was applied to properly smooth the overly predicted values of Improved GIS-based Regression Model (IGISRM), called the IDW-IGISRM grid data, at the same resolution for daily precipitation, maximum temperature and minimum temperature from 2001 to 2010 over South Korea. We tested various effective distances in the IDW method to detect an optimal distance that provides the highest performance. IDW-IGISRM was compared with IGISRM to evaluate the effectiveness of IDW-IGISRM with regard to spatial patterns, and quantitative performance metrics over 243 AWS observational points and four selected stations showing the largest biases. Regarding the spatial pattern, IDW-IGISRM reduced irrational overly predicted values, i. e. producing smoother spatial maps that IGISRM for all variables. In addition, all quantitative performance metrics were improved by IDW-IGISRM; correlation coefficient (CC), Index Of Agreement (IOA) increase up to 11.2% and 2.0%, respectively. Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) were also reduced up to 5.4% and 15.2% respectively. At the selected four stations, this study demonstrated that the improvement was more considerable. These results indicate that IDW-IGISRM can improve the predictive performance of IGISRM, consequently providing more reliable high-resolution gridded data for assessment, adaptation, and vulnerability studies of climate change impacts.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.14-14
• /
• 2015

공간보간기법은 미계측지역의 강수예측을 위해 통상적으로 사용되는 방법 중의 하나이다. 이 연구에서는 기상청에서 제공하고 있는 RCP 8.5 시나리오에 의한 남한상세 강수자료 중 지형이 복잡한 도서지역에서 제공되지 않는 데이터 누락격자에 대하여 최적의 공간보간기법을 선정하여 강수자료를 생성할 수 있도록 하였다. 적합한 보간기법을 선정하기 위해 데이터 누락지역에 대한 분석을 수행하였고, 최신 행정구역도에 맞추어 $1km{\times}1km$ 격자를 한반도 전체지역에 맞추어 생성된 격자를 사용하였다. ESRI사의 ArcGIS 프로그램을 이용하여 공간보간기법을 적용하였다. 사용된 보간법은 역거리가중치법(IDW), 정규크리깅(Ordinary Kriging), 보편크리깅(Universal Kriging), 스플라인(Spline)이며 가장 적합한 공간보간기법을 선정하기 위해 기후변화시나리오에 의한 데이터 중 해안선 주변 특정격자에서의 값을 누락시켜 공간보간기법을 통해 생성된 값과 기후변화 시나리오에 의한 값을 정량적으로 비교하였다. 공간보간기법의 적합도 평가를 위해 MAE(Mean Absolute Error), MSE(Mean Squared Error), PBIAS(Percent of BIAS), G(goodness of prediction) 분석을 수행하였고, 산점도 분석을 통해 실제값과 보간값의 오차율 평가를 병행하여 최적 공간보간기법을 결정하였다. 사용된 강수데이터는 RCP 8.5 시나리오에서 2015~2019년 중 강수가 높게 나타난 8월 자료를 이용하였다. 해안선 지역의 강수량 추정시 역거리 가중치법과 크리깅방법은 일부 지점에서 과다 추정되는 경향이 있고, 스플라인 방법이 전체적인 총 강수량이 기후변화시나리오에 의한 실제값과 유사한 것으로 나타났다. 실제값과 보간값의 교차검증을 수행한 결과 정규크리깅 기법이 가장 높은 정확도를 보였으며, 전체적으로 실제값과 유사한 범위내의 강수량이 생성되는 것으로 나타났다.