• Journal of The Korean Society of Agricultural Engineers
• /
• v.66 no.1
• /
• pp.1-13
• /
• 2024

This study assessed the efficacy of improving the accuracy of reservoir water level prediction models by employing automated machine learning models and efficient cross-validation methods for time-series data. Considering the inherent complexity and non-linearity of time-series data related to reservoir water levels, we proposed an optimized approach for model selection and training. The performance of twelve models was evaluated for the Obong Reservoir in Gangneung, Gangwon Province, using the TPOT (Tree-based Pipeline Optimization Tool) and four cross-validation methods, which led to the determination of the optimal pipeline model. The pipeline model consisting of Extra Tree, Stacking Ridge Regression, and Simple Ridge Regression showed outstanding predictive performance for both training and test data, with an R² (Coefficient of determination) and NSE (Nash-Sutcliffe Efficiency) exceeding 0.93. On the other hand, for predictions of water levels 12 hours later, the pipeline model selected through time-series split cross-validation accurately captured the change pattern of time-series water level data during the test period, with an NSE exceeding 0.99. The methodology proposed in this study is expected to greatly contribute to the efficient generation of reservoir water level predictions in regions with high rainfall variability.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.9
• /
• pp.2605-2625
• /
• 2024

Siamese-based segmentation and tracking algorithms improve accuracy and stability for video object segmentation and tracking tasks simultaneously. Although effective, variability in target appearance and background clutter can still affect segmentation accuracy and further influence the performance of tracking. In this paper, we present a memory propagation-based target-aware and mask-attention decision network for robust object segmentation and tracking. Firstly, a mask propagation-based attention module (MPAM) is constructed to explore the inherent correlation among image frames, which can mine mask information of the historical frames. By retrieving a memory bank (MB) that stores features and binary masks of historical frames, target attention maps are generated to highlight the target region on backbone features, thus suppressing the adverse effects of background clutter. Secondly, an attention refinement pathway (ARP) is designed to further refine the segmentation profile in the process of mask generation. A lightweight attention mechanism is introduced to calculate the weight of low-level features, paying more attention to low-level features sensitive to edge detail so as to obtain segmentation results. Finally, a mask fusion mechanism (MFM) is proposed to enhance the accuracy of the mask. By utilizing a mask quality assessment decision network, the corresponding quality scores of the "initial mask" and the "previous mask" can be obtained adaptively, thus achieving the assignment of weights and the fusion of masks. Therefore, the final mask enjoys higher accuracy and stability. Experimental results on multiple benchmarks demonstrate that our algorithm performs outstanding performance in a variety of challenging tracking tasks.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.16 no.4
• /
• pp.297-303
• /
• 2014

The effect of solar irradiance has been used to estimate daily maximum temperature, which make it possible to reduce the error inherent to lapse-rate based elevation difference correction in mountainous terrain. Still, recent observations indicated that the effect of solar radiation would need correction for estimation of daily maximum temperature. It was attempted to examine what would cause the variability of solar irradiance effect in determination of daily maximum temperature under natural field conditions and to suggest improved methods for estimation of the temperature distribution over mountainous regions. Temperature at 1500 and the wind speed for 1100 to 1500 were obtained at 10 validation sites with various topographical features including slope and aspect within a mountainous $50km^2$ catchment for 2012-2013. Lapse-rate corrected temperature estimates on clear days were compared with these observations, which would represent the differential irradiance effect among sloped surfaces. Results indicated a negative correlation between the mean wind speed and the estimation error. A simple scheme was derived from relationship between wind speed and estimation error for daily temperature to correct the effect of solar radiation. This scheme was incorporated into an existing model to estimate daily maximum temperature based on the effect of solar radiation. At 10 validation sites on clear days, estimates of 1500 LST temperature with and without the correction scheme were compared. It was found that a substantial improvement was achieved when the correction scheme was applied in terms of bias correction as well as error size reduction at all sites.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.15 no.1
• /
• pp.13-24
• /
• 2013

Various elastic wave-based site investigation methods have been used to characterize subsurface ground because the dynamic properties can be correlated with various geotechnical parameters. Although the inherent spatial variability of the geotechnical parameters affects the P-wave propagation characteristics, ground heterogeneity has not been considered as an influential factor. Thus, the effect of heterogeneous ground on the travel-time shift and wavefront characteristics of elastic waves through stochastic numerical analyses is investigated in this study. The effects of the relative correlation lengths and relative propagation distances on the travel-time shift of P-waves considering various intensities of ground heterogeneity were investigated. Heterogeneous ground fields of stiffness (e.g., the coefficient of variation = 10 ~ 40%) were repeatedly realized in numerical finite difference grids using the turning band method. Monte Carlo simulations were undertaken to simulate P-wave propagation in heterogeneous ground using a finite difference method-based numerical approach. The results show that the disturbance of the wavefront becomes more significant with stronger heterogeneity and induces travel-time delays. The relative correlation lengths and propagation distances are systematically related to the travel-time shift.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.27 no.6
• /
• pp.566-576
• /
• 2016

Inverse synthetic aperture radar(ISAR) image generated from bistatic radar(Bi-ISAR) represents two-dimensional scattering distribution of a target, and the Bi-ISAR can be used for bistatic target identification. However, Bi-ISAR has large variability in scattering mechanisms depending on bistatic configurations and do not represent exact range-Doppler information of a target due to inherent distortion. Thus, an efficient training DB construction is the most important factor in target identification using Bi-ISARs. Recently, a database construction method based on realistic flight scenarios of a target, which provides a reliable identification performance for the monostatic target identification, was applied to target identification using high resolution range profiles(HRRPs) generated from bistatic radar(Bi-HRRPs), to construct efficient training DB under bistatic configurations. Consequently, high identification performance was achieved using only small amount of training Bi-HRRPs, when the target is a considerable distance away from the bistatic radar. Thus, flight scenarios based training DB construction is applied to target identification using Bi-ISARs. Then, the capability and efficiency of the method is analyzed.

• The Korean Journal of Nuclear Medicine
• /
• v.37 no.3
• /
• pp.162-170
• /
• 2003

The probabilistic anatomical maps are used to localize the functional neuro-images and morphological variability. The quantitative indicator is very important to inquire the anatomical position of an activated legion because functional image data has the low-resolution nature and no inherent anatomical information. Although previously developed MNI probabilistic anatomical map was enough to localize the data, it was not suitable for the Korean brains because of the morphological difference between Occidental and Oriental. In this study, we develop a probabilistic anatomical map for Korean normal brain. Normal 75 blains of T1-weighted spoiled gradient echo magnetic resonance images were acquired on a 1.5-T GESIGNA scanner. Then, a standard brain is selected in the group through a clinician searches a brain of the average property in the Talairach coordinate system. With the standard brain, an anatomist delineates 89 regions of interest (ROI) parcellating cortical and subcortical areas. The parcellated ROIs of the standard are warped and overlapped into each brain by maximizing intensity similarity. And every brain is automatically labeledwith the registered ROIs. Each of the same-labeled region is linearly normalize to the standard brain, and the occurrence of each legion is counted. Finally, 89 probabilistic ROI volumes are generated. This paper presents a probabilistic anatomical map for localizing the functional and structural analysis of Korean normal brain. In the future, we'll develop the group specific probabilistic anatomical maps of OCD and schizophrenia disease.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.32 no.4
• /
• pp.262-272
• /
• 2020

A large number of studies on wave breaking have been carried out, and many experimental data have been documented. Moreover, on the basis of various experimental data set, many empirical or semi-empirical formulas based primarily on regression analysis have been proposed to quantitatively estimate wave breaking for engineering applications. However, wave breaking has an inherent variability, which imply that a linear statistical approach such as linear regression analysis might be inadequate. This study presents an alternative nonlinear method using an neural network, one of the machine learning methods, to estimate breaking wave height and breaking depth. The neural network is modeled using Tensorflow, a machine learning open source platform distributed by Google. The neural network is trained by randomly selecting the collected experimental data, and the trained neural network is evaluated using data not used for learning process. The results for wave breaking height and depth predicted by fully trained neural network are more accurate than those obtained by existing empirical formulas. These results show that neural network is an useful tool for the prediction of wave breaking.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.39 no.1
• /
• pp.165-174
• /
• 2019

As a nonstationarity is observed in hydrological data, various studies on nonstationary frequency analysis for hydraulic structure design have been actively conducted. Although the inherent diversity in the atmosphere-ocean system is known to be related to the nonstationary phenomena, a nonstationary frequency analysis is generally performed based on the linear trend. In this study, a nonstationary frequency analysis was performed using climate indices as covariates to consider the climate variability and the long-term trend of the extreme rainfall. For 11 weather stations where the trend was detected, the long-term trend within the annual maximum rainfall data was extracted using the ensemble empirical mode decomposition. Then the correlation between the extracted data and various climate indices was analyzed. As a result, autumn-averaged AMM, autumn-averaged AMO, and summer-averaged NINO4 in the previous year significantly influenced the long-term trend of the annual maximum rainfall data at almost all stations. The selected seasonal climate indices were applied to the generalized extreme value (GEV) model and the best model was selected using the AIC. Using the model diagnosis for the selected model and the nonstationary GEV model with the linear trend, we identified that the selected model could compensate the underestimation of the rainfall quantiles.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.23 no.4
• /
• pp.391-404
• /
• 2021

This study developed a long-term prediction model for the potential yield of garlic based on a genetic algorithm (GA) by utilizing global climate reanalysis data. The GA is used for digging the inherent signals from global climate reanalysis data which are both directly and indirectly connected with the garlic yield potential. Our results indicate that both deterministic and probabilistic forecasts reasonably capture the inter-annual variability of crop yields with temporal correlation coefficients significant at 99% confidence level and superior categorical forecast skill with a hit rate of 93.3% for 2 × 2 and 73.3% for 3 × 3 contingency tables. Furthermore, the GA method, which considers linear and non-linear relationships between predictors and predictands, shows superiority of forecast skill in terms of both stability and skill scores compared with linear method. Since our result can predict the potential yield before the start of farming, it is expected to help establish a long-term plan to stabilize the demand and price of agricultural products and prepare countermeasures for possible problems in advance.

• Journal of the Korean Geotechnical Society
• /
• v.25 no.10
• /
• pp.87-97
• /
• 2009

The seismic site coefficients are often used with the seismic hazard maps to develop the design response spectrum at the surface. The site coefficients are most commonly developed deterministically, while the seismic hazarde maps are derived probabilistically. There is, hence, an inherent incompatibility between the two approaches. However, they are used together in the seismic design codes without a clear rational basis. To resolve the fundamental imcompatibility between the site coefficients and hazard maps, this study uses a novel probabilistic seismic hazard analysis (PSHA) technique that simulates the results of a standard PSHA at a rock outcrop, but integrates the site response analysis function to capture the site amplification effects within the PSHA platform. Another important advantage of the method is its ability to model the uncertainty, variability, and randomness of the soil properties. The new PSHA was used to develop fully probabilistic site coefficients for site classes of the seismic design code and another sets of site classes proposed in Korea. Comparisons highlight the pronounced discrepancy between the site coefficients of the seismic design code and the proposed coefficients, while another set of site coefficients show differences only at selected site classes.