• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.1
• /
• pp.85-92
• /
• 2020

Deep learning shows outstanding performance in image and video analysis, such as object classification, object detection and semantic segmentation. In this paper, it is analyzed that the performances of deep learning models can be affected by characteristics of train dataset. It is proposed as a method for selecting activation function and optimization algorithm of deep learning to classify facial expression. Classification performances are compared and analyzed by applying various algorithms of each component of deep learning model for CK+, MMI, and KDEF datasets. As results of simulation, it is shown that genetic algorithm can be an effective solution for optimizing components of deep learning model.

• Proceedings of the Korean Radioactive Waste Society Conference
• /
• 2004.06a
• /
• pp.105-114
• /
• 2004

This study investigated the thermal behavior of the nuclear graphite waste generated from the decommissioning of the Korean nuclear research reactor, The first part study investigated the decomposition rate of the nuclear graphite waste up to $1000^{\circ}C$ under various oxygen partial pressures using a thermo-gravimetric analyzer (TGA). Tested graphite waste sample not easily destroyed in the oxygen-deficient condition. However, the gas-solid oxidation reaction was found to be very effective in the presence of oxygen. No significant amount of the product of incomplete combustion was formed even in the limited oxygen concentration of 4% $O_2$. The influence of temperature and oxygen partial pressure was evaluated by the theoretical model analysis of the thermo-gravimetric data. The activation energy and the reaction order of graphite oxidation were evaluated as 128 kJ/mole and 1.1, respectively. The second part of this study investigated the behavior of radioactive elements under graphite oxidation atmosphere using thermodynamic equilibrium model. $^{22}Na$, $^{134}Cs$ and $^{137}Cs$ were found be the semi-volatile elements. Since volatile uranium species can be formulated at high temperatures above $1050^{\circ}C$, the temperature of incinerator furnace should be minimized. Other corrosion/activation products, fission products and uranium were found to be the non-volatile species.

• BMB Reports
• /
• v.38 no.4
• /
• pp.373-380
• /
• 2005

For measles viruses, fusion on the cell membrane is an important initial step in the entry into the infected cells. The recent research indicated that hemagglutinin firstly leads the conformational changes in the fusion protein then co-mediates the membrane fusion. In the work, we use the co-immunoprecipitation and pull-down techniques to identify the interactions among fusion protein, hemagglutinin and signaling lymphocyte activation molecule (SLAM), which reveal that the three proteins can form a functional complex to mediate the SLAM-dependent fusion. Moreover, under the confocal microscope, fusion protein and hemagglutinin protein can show the cocapping mediated by the SLAM. So fusion protein not only is involved in the fusion but also might directly interact with the SLAM to be a new fusion-trimer model, which might account for the infection mechanism of measles virus.

• Carbon letters
• /
• v.8 no.3
• /
• pp.199-206
• /
• 2007

The activated carbon produced from rubber wood sawdust by chemical activation using phosphoric acid have been utilized as an adsorbent for the removal of Cu(II) from aqueous solution in the concentration range 5-40 mg/l. Adsorption experiments were carried out in a batch process and various experimental parameters such as effect of contact time, initial copper ion concentration, carbon dosage, and pH on percentage removal have been studied. Adsorption results obtained for activated carbon from rubber wood sawdust were compared with the results of commercial activated carbon (CAC). The adsorption on activated carbon samples increased with contact time and attained maximum value at 3 h for CAC and 4 h for PAC. The adsorption results show that the copper uptake increased with increasing pH, the optimum efficiency being attained at pH 6. The precipitation of copper hydroxide occurred when pH of the adsorbate solution was greater than 6. The equilibrium data were fitted using Langmuir and Freundlich adsorption isotherm equation. The kinetics of sorption of the copper ion has been analyzed by two kinetic models, namely, the pseudo first order and pseudo second order kinetic model. The adsorption constants and rate constants for the models have been determined. The process follows pseudo second order kinetics and the results indicated that the Langmuir model gave a better fit to the experimental data than the Freundlich model. It was concluded that activated carbon produced using phosphoric acid has higher adsorption capacity when compared to CAC.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• v.2
• /
• pp.101-106
• /
• 2006

The geoidal undulations are needed for determining the orthometric heights from the Global Positioning System GPS-derived ellipsoidal heights. There are several methods for geoidal undulation determination. The paper presents a method employing a simple architecture Two Layer Multiquadric-Biharmonic Artificial Neural Network (TLMB-ANN) to approximate an area of 4200 square kilometres quasigeoid surface with GPS-levelling data. Hardy’s Multiquadric-Biharmonic functions is used as the hidden layer neurons’ activation function and Levenberg-Marquardt algorithm is used to train the artificial neural network. In numerical examples five surfaces were compared: the gravimetric geometry hybrid quasigeoid, Support Vector Machine (SVM) model, Hybrid Fuzzy Neural Network (HFNN) model, Traditional Three Layer Artificial Neural Network (ANN) with tanh activation function and TLMB-ANN surface approximation. The effectiveness of TLMB-ANN surface approximation depends on the number of control points. If the number of well-distributed control points is sufficiently large, the results are similar with those obtained by gravity and geometry hybrid method. Importantly, TLMB-ANN surface approximation model possesses good extrapolation performance with high precision.

• Preventive Nutrition and Food Science
• /
• v.1 no.1
• /
• pp.41-45
• /
• 1996

This study was conducted to develop the fermentation kinetic model for the prediction of acidity and pH changes in Kimchi as a function of fermentation temperatures. The fitness of the model was evaluated using traditional two-step method and an alternative non-linear regression method. The changes in acidity and pH during fermentation followed the pattern of the first order reaction of a two-step method. As the fermentation temperature increased from 4$^{\circ}C$ to 28, the reaction rates of acidity and pH were increased 8.4 and 7.6 times, respectively. The activation energies of acidity and pH were 16.125 and 16.003kcal/mole. The average activation energies of acidity and pH using a non-linear method were 16.006 by the first order and 15.813 kcal/mole by the zero order, respectively. The non-linear procedure had better fitting 개 experimental data of the acidity and pH than two-step method. The shelf-lives based on the time to reach the 1.0% of acidity were 33.1day at 4$^{\circ}C$ and 2.8 day 28$^{\circ}C$.

• The Journal of Information Systems
• /
• v.28 no.1
• /
• pp.115-132
• /
• 2019

Purpose The purpose of this study is to explain the agrophotovolatic system built by the Korea South-East Power Company and to propose methods to activate the agrophotovolatic system for the development of the renewable energy industry. Design/methodology/approach We conducted a three-step simulation in order to design a photovoltaic module, and we built the agrophotovolatic system based on the results of the simulation. Then, we analyzed the monthly generation of power and the rice harvests produced on farmland using the photovoltaic module. Based on the results of the analysis, we proposed institutional improvements to increase the use of the agrophotovolatic system, and we proposed new business models to increase the participation of farmers and business persons. Findings When we compared the agrophotovolastic system with the general photovoltaic system, we found that the agrophotovoltaic system had higher utilization rates and power generation. An analysis of rice produced on farmland using the photovoltaic module showed that more than 80% of the rice produced on general farmland was harvested. We suggested activation plans that involved the revision of the farmland law and the introduction of renewable energy certificate (REC). We also proposed a land lease model and a farmer participation model as two new business models, and we conducted economic evaluations and sensitivity analyses for both models.

• ETRI Journal
• /
• v.46 no.3
• /
• pp.538-549
• /
• 2024

A conventional codec aims to increase the compression efficiency for transmission and storage while maintaining video quality. However, as the number of platforms using machine vision rapidly increases, a codec that increases the compression efficiency and maintains the accuracy of machine vision tasks must be devised. Hence, the Moving Picture Experts Group created a standardization process for video coding for machines (VCM) to reduce bitrates while maintaining the accuracy of machine vision tasks. In particular, in-loop filters have been developed for improving the subjective quality and machine vision task accuracy. However, the high computational complexity of in-loop filters limits the development of a high-performance VCM architecture. We analyze the effect of an in-loop filter on the VCM performance and propose a suboptimal VCM method based on the selective activation of in-loop filters. The proposed method reduces the computation time for video coding by approximately 5% when using the enhanced compression model and 2% when employing a Versatile Video Coding test model while maintaining the machine vision accuracy and compression efficiency of the VCM architecture.

• The Journal of the Korea Contents Association
• /
• v.21 no.3
• /
• pp.616-625
• /
• 2021

Deep neural networks are an approximation method that approximates an arbitrary function to a linear model and then repeats additional approximation using a nonlinear active function. In this process, the method of evaluating the performance of approximation uses the loss function. Existing in-depth learning methods implement approximation that takes into account loss functions in the linear approximation process, but non-linear approximation phases that use active functions use non-linear transformation that is not related to reduction of loss functions of loss. This study proposes parametric activation functions that introduce scale parameters that can change the scale of activation functions and location parameters that can change the location of activation functions. By introducing parametric activation functions based on scale and location parameters, the performance of nonlinear approximation using activation functions can be improved. The scale and location parameters in each hidden layer can improve the performance of the deep neural network by determining parameters that minimize the loss function value through the learning process using the primary differential coefficient of the loss function for the parameters in the backpropagation. Through MNIST classification problems and XOR problems, parametric activation functions have been found to have superior performance over existing activation functions.

• Applied Chemistry for Engineering
• /
• v.27 no.1
• /
• pp.56-61
• /
• 2016

In this study, the effect of natural minerals on the reaction kinetics for lignite-$CO_2$ gasification was investigated. After physical mixing of lignite from Meng Tai area with 5 wt% of each natural mineral catalysts among Dolomite, Silica sand, Olivine and Kaolin, $CO_2$ gasification was performed using TGA at each 800, $850^{\circ}C$ and $900^{\circ}C$. The experimental data was analyzed with volumetric reaction model (VRM), shrinking core model (SCM) and modified volumetric reaction model (MVRM). MVRM was the most suitable among three models. As increasing the reaction temperature, the reaction rate constant became higher. With natural mineral catalysts, the reaction rate constant was higher and activation energy was lower than that of without catalysts. The lowest activation energy, 114.90 kJ/mol was obtained with silica sand. The highest reaction rate constant at $850^{\circ}C$ and $900^{\circ}C$ and lower reaction rate constant at $800^{\circ}C$ were obtained with Kaolin. Conclusively, the better catalytic performance could be observed with Kaolin than that of using other catalysts when the reaction temperature increased.