• Healthcare Informatics Research
• /
• v.24 no.4
• /
• pp.309-316
• /
• 2018

Objectives: Both the valence and arousal components of affect are important considerations when managing mental healthcare because they are associated with affective and physiological responses. Research on arousal and valence analysis, which uses images, texts, and physiological signals that employ deep learning, is actively underway; research investigating how to improve the recognition rate is needed. The goal of this research was to design a deep learning framework and model to classify arousal and valence, indicating positive and negative degrees of emotion as high or low. Methods: The proposed arousal and valence classification model to analyze the affective state was tested using data from 40 channels provided by a dataset for emotion analysis using electrocardiography (EEG), physiological, and video signals (the DEAP dataset). Experiments were based on 10 selected featured central and peripheral nervous system data points, using long short-term memory (LSTM) as a deep learning method. Results: The arousal and valence were classified and visualized on a two-dimensional coordinate plane. Profiles were designed depending on the number of hidden layers, nodes, and hyperparameters according to the error rate. The experimental results show an arousal and valence classification model accuracy of 74.65 and 78%, respectively. The proposed model performed better than previous other models. Conclusions: The proposed model appears to be effective in analyzing arousal and valence; specifically, it is expected that affective analysis using physiological signals based on LSTM will be possible without manual feature extraction. In a future study, the classification model will be adopted in mental healthcare management systems.

• Korean Journal of Materials Research
• /
• v.29 no.5
• /
• pp.328-335
• /
• 2019

Nb-doped $TiO_2$(NTO) coated NiCrAl alloy foam for hydrogen production is prepared using ultrasonic spray pyrolysis deposition(USPD) method. To optimize the size and distribution of NTO particles based on good physical and chemical stability, we synthesize particles by adjusting the weight ratio of the Nb precursor solution(5 wt%, 10 wt% and 15 wt%). The morphological, chemical bonding, and structural properties of the NTO coated NiCrAl alloy foam are investigated by X-ray diffraction(XRD), X-ray photo-electron spectroscopy(XPS), and Field-Emission Scanning Electron Microscopy(FESEM). As a result, the samples of controlled Nb weight ratio exhibit a common diffraction pattern at ${\sim}25.3^{\circ}$, corresponding to the(101) plane, and have chemical bonding(O-Nb=O) at 534 eV. The NTO particles with the optimum weight ratio of N (10 wt%) show a uniform distribution with a size of ~18.2-21.0 nm. In addition, they exhibit the highest corrosion resistance even in the electrochemical stability estimation. As a result, the introduction of NTO coated NiCrAl alloy foam by USPD improves the chemical stability of the NiCrAl alloy foam by protecting the direct electrochemical reaction between the foam and the electrolyte. Thus, the optimized NTO coating can be proposed for excellent protection of NiCrAl alloy foam for hydrocarbon-based steam methane reforming(SMR).

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.12
• /
• pp.1624-1630
• /
• 2019

With the increase in there are smart devices penetration around the world, services related to exercise checks are attracting attention. However, there is existing exercise amount measurement service does not use the altitude information, or because the use of an algorithm that does not corrected the GPS altitude error is not accurate movement distance provided have a problem. Therefore, in this paper, to improve the existing problems, Kalman-filter-based user movement distance measurement algorithm is designed and implementation of improved by using the Kalman-filter based GPS and barometric altimeter sensor fusion algorithm to improve the altitude value the accuracy and of calculate the coordinate plane distance. As a result of comparing the designed and implementation of algorithm with the existing algorithms, it is confirmed that the proposed algorithm improves the accuracy by about 2.17%.

• Journal of Advanced Navigation Technology
• /
• v.24 no.6
• /
• pp.627-632
• /
• 2020

In this paper, we have studied a development of a humidity sensor using a microwave sensor based on a microstrip defected ground structure coated with polyvinyl alcohol. A high-sensitivity microwave sensor, which is sensitive to the changes in the permittivity of the material under test, is designed by adding an interdigital capacitor-shaped defected ground structure to the ground plane of a microstrip line. Polyvinyl alcohol, a polymer material whose permittivity varies depending on humidity, is coated with a thin thickness on the defected ground structure of the proposed microwave sensor, and the changes in the resonance frequency and magnitude of the transmission coefficient for the microwave sensor according to humidity are measured. When relative humidity increases from 40% to 80% in 10% increments at a temperature of 25 degrees using a temperature/humidity chamber, the resonant frequency of the transmission coefficient decreases from 1.475 GHz to 1.449 GHz, and the magnitude is increased from -32.90 dB to -25.67 dB.

• Journal of Advanced Navigation Technology
• /
• v.26 no.1
• /
• pp.29-34
• /
• 2022

In this paper, the design of a triple dipole quasi-yagi antenna operating in the 2.45 GHz and 5 GHz wireless LAN frequency bands and the 3.5 GHz WiMAX frequency band was studied. The proposed quasi-Yagi antenna consists of three dipoles connected in series with a V-shaped ground plane. The longest half-bow-tie-shaped dipole resonates in the 2.45 GHz band, whereas the medium-length dipole resonates at 3.5 GHz. The shortest dipole resonates in the 5 GHz band. By adjusting the length and width of the dipoles and the spacings between the dipoles, a triple-band directional antenna operating in the 2.45 GHz, 3.5 GHz, and 5 GHz bands are designed, and fabricated on an FR4 substrate with a size of 45 mm × 55 mm. It was confirmed that the fabricated antenna operates in the designed triple bands of 2.32-2.57 GHz, 3.26-3.69 GHz, and 4.50-6.56 GHz for a voltage standing wave ratio less than 2. Gain is maintained above 3 dBi in the three bands.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.5
• /
• pp.489-495
• /
• 2022

It makes difference to LSTM D/L(Deep Learning) results for language model construction as the corpus preprocess changes. An LSTM model was trained with a famouse literaure poems(Ki Hyung-do's work) for training corpus in the study. You get the two wordvector sets for two corpus sets of the original text and eraised word ending text each once D/L training completed. It's been inspected of the similarity/analogy operation results, the positions of the wordvectors in 2D plane and the generated texts by the language models for the two different corpus sets. The suggested words by the silmilarity/analogy operations are changed for the corpus sets but they are related well considering the corpus characteristics as a literature work. The positions of the wordvectors are different for each corpus sets but the words sustained the basic meanings and the generated texts are different for each corpus sets also but they have the taste of the original style. It's supposed that the D/L language model can be a useful tool to enjoy the literature in object and in diverse with the analysis results shown in the study.

• Journal of Electrochemical Science and Technology
• /
• v.14 no.4
• /
• pp.388-396
• /
• 2023

Photoelectrochemical (PEC) water splitting is a vital source of clean and sustainable hydrogen energy. Moreover, the large-scale H₂ production is currently necessary, while long-term stability and high PEC activity still remain important issues. In this study, a GaN-based photoelectrode was modified by an additional NH₃ treatment (900℃ for 10 min) and its PEC behavior was monitored. The bare GaN exhibited a highly crystalline wurtzite structure with the (002) plane and the optical bandgap was approximately 3.2 eV. In comparison, the NH₃-treated GaN film exhibited slightly reduced crystallinity and a small improvement in light absorption, resulting from the lattice stress or cracks induced by the excessive N supply. The minor surface nanotexturing created more surface area, providing electroactive reacting sites. From the surface XPS analysis, the formation of an N-Ga-O phase on the surface region of the GaN film was confirmed, which suppressed the charge recombination process and the positive shift of E_FB. Therefore, these effects boosted the PEC activity of the NH₃-treated GaN film, with J values of approximately 0.35 and 0.78 mA·cm^-2 at 0.0 and 1.23 V_RHE, respectively, and an onset potential (V_on) of -0.24 V_RHE. In addition, there was an approximate 50% improvement in the J value within the highly applied potential region with a positive shift of V_on. This result could be explained by the increased nanotexturing on the surface structure, the newly formed defect/trap states correlated to the positive V_on shift, and the formation of a GaO_xN_1-x phase, which partially blocked the charge recombination reaction.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.6
• /
• pp.145-150
• /
• 2023

The subway is a public transportation that many people use every day. Line 2 especially has the most crowded stations during the day. However, the risk of crush accidents is increasing due to high congestion during rush hour and this reduces the safety and comfort of passengers. Subway congestion prediction is helpful to forestall problems caused by high congestion. Therefore, this study proposes machine learning classification models that predict subway congestion during commuting time. To predict congestion in Line 2 based in machine learning, we investigate variables that affect subway congestion through previous research and collect a dataset of subway congestion on Line 2 during rush hour from PUBLIC DATA PORTAL. The proposed model is expected to establish the subway operation plane to make passengers safe and satisfied.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.13 no.2
• /
• pp.73-84
• /
• 1993

In this research, configuration design optimization of plane truss structure has been tested by using decomposition technique. In the first level, the problem of transferring the nonlinear programming problem to linear programming problem has been effectively solved and the number of the structural analysis necessary for doing the sensitivity analysis can be decreased by developing stress constraint into member stress approximation according to the design space approach which has been proved to be efficient to the sensitivity analysis. And the weight function has been adopted as cost function in order to minimize structures. For the design constraint, allowable stress, buckling stress, displacement constraint under multi-condition and upper and lower constraints of the design variable are considered. In the second level, the nodal point coordinates of the truss structure are used as coordinating variable and the objective function has been taken as the weight function. By treating the nodal point coordinates as design variable, unconstrained optimal design problems are easy to solve. The decomposition method which optimize the section areas in the first level and optimize configuration variables in the second level was applied to the plane truss structures. The numerical comparisons with results which are obtained from numerical test for several truss structures with various shapes and any design criteria show that convergence rate is very fast regardless of constraint types and configuration of truss structures. And the optimal configuration of the truss structures obtained in this study is almost the identical one from other results. The total weight couldbe decreased by 5.4% - 15.4% when optimal configuration was accomplished, though there is some difference.

• Journal of Digital Convergence
• /
• v.13 no.4
• /
• pp.303-310
• /
• 2015

Game technologies are now applied in various engineering areas such as the simulation of surgical operation or the implementation of a cyber model house. One of the essential and important technology in these applications is cutting of the 3D polygon model in real time. Real-time cutting technology is an essential technology needed to implement the simulation of a building demolition or a car assembly for training or educational purpose. Previous cutting method using the conventional BSP-Tree structure has some limitations in that they divide the whole world including the 3D model and its environment, only into two parts with respect to an infinite plane. In this paper, we show a technique cutting the 3D model in a finite extent in order to solve this problem. Specifically, we restricted the cut surface in a finite rectangular area and constructed the mesh for the divided surface. To show the usefulness of our partial cutting technique, an example of the dismantling process simulation of a nuclear reactor polygon model was illustrated.