• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.5
• /
• pp.103-110
• /
• 2008

A reliable extraction of welded area is the precedent task before the detection of weld defects in industrial radiography. This paper describes an attempt to detect and extract the welded features of steel tubes from the computed radiography(CR) images. The statistical properties are first analyzed on over 160 sample radiographic images which represent either weld or non-weld area to identify the differences between them. The analysis is then proceeded by pattern classification to determine the clustering parameters. These parameters are the width, the functional match, and continuity. The observed weld image is processed line by line to calculate these parameters for each flexible moving window in line image pixel set. The local statistic line-clustering method is used as the classifier to recognize each window data as weld or non-weld cluster. The sequential procedure is to track the edge lines between two distinct regions by iterative calculation of threshold, and it results in extracting the weld feature. Our methodology is concluded to be effective after experiment with CR weld images.

• Journal of the Korean Chemical Society
• /
• v.60 no.1
• /
• pp.9-15
• /
• 2016

The theoretical investigation has been performed to predict detonation velocity, detonation pressure, and thermodynamic stability of HMX/LLM-116 cocrystal. All possible geometries of HMX, LLM-116, and cocrystal have been optimized at the B3LYP/cc-pVTZ level of theory. The binding energy for the trigger bond and cluster has been calculated to predict the thermodynamic stability. The MP2 binding energies were obtained using single point energy calculation at the B3LYP optimized geometries, and the density has been calculated from monte carlo integration. The detonation velocity and detonation pressure have been calculated using Kamlet-Jacobs equation, while enthalpy has been predicted at the CBS-Q level of theory.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.9
• /
• pp.1731-1737
• /
• 2017

K Nearest Neighbor Join (KNN Join) is a simple yet effective method in machine learning. It is widely used in small dataset of the past time. As the number of data increases, it is infeasible to run this model on an actual application by a single machine due to memory and time restrictions. Nowadays a popular batch process model called MapReduce which can run on a cluster with a large number of computers is widely used for large-scale data processing. Hadoop is a framework to implement MapReduce, but its performance can be further improved by a new framework named Spark. In the present study, we will provide a KNN Join implement based on Spark. With the advantage of its in-memory calculation capability, it will be faster and more effective than Hadoop. In our experiments, we study the influence of different factors on running time and demonstrate robustness and efficiency of our approach.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.17 no.7
• /
• pp.691-696
• /
• 2004

The electronic state and chemical bonding of $\beta$-MnO$_2$ with transition metal dopants were theoretically investigated by DV-X$_{\alpha}$ (the discrete variational X$_{\alpha}$) method, which is a sort of the first principles molecular orbital method using the Hartree-Fock-Slater approximation. The calculations were performed with a $_Mn_{14}$ MO$_{56}$ )$^{-52}$ (M = transition metals) cluster model. The electron energy level, the density of states (DOS), the overlap population, the charge density distribution, and the net charges, were calculated. The energy level diagram of MnO$_2$ shows the different band structure and electron occupancy between the up spin states and down spin states. The dopant levels decrease between the conduction band and the valence band with the increase of the atomic number of dopants. The covalency of chemical bonding was shown to increase and ionicity decreased in increasing the atomic number of dopants. Calculated results were discussed on the basis of the interaction between transition metal 3d and oxygen 2p orbital. In conclusion it is expected that when the transition metals are added to MnO$_2$ the band gap decreases and the electronic conductivity increases with the increase of the atomic number of dopants. the atomic number of dopants.

• Journal of the Mineralogical Society of Korea
• /
• v.24 no.4
• /
• pp.289-300
• /
• 2011

Atomistic origins of carbon solubility into silicates are essential to understand the effect of carbon on the properties of silicates and evolution of the Earth system through igneous and volcanic processes. Here, we investigate the atomic structure and NMR properties of dissolved carbon in enstatite using Raman spectroscopy and quantum chemical calculations. Raman spectrum for enstatite synthesized with 2.4. wt% of amorphous carbon at 1.5 GPa and $1,400^{\circ}C$ shows vibrational modes of enstatite, but does not show any vibrational modes of $CO_2$ or ${CO_3}^{2-}$. The result indicates low solubility of carbon into enstatite at a given pressure and temperature conditions. Because $^{13}C$ NMR chemical shift is sensitive to local atomic structure around carbon and we calculated $^{13}C$ NMR chemical shielding tensors for C substituted enstatite cluster as well as molecular $CO_2$ using quantum chemical calculations to give insights into $^{13}C$ NMR chemical shifts of carbon in enstatite. The result shows that $^{13}C$ NMR chemical shift of $CO_2$ is 125 ppm, consistent with previous studies. Calculated $^{13}C$ NMR chemical shift of C is ~254 ppm. The current calculation will alllow us to assign potential $^{13}C$ NMR spectra for the enstatite dissolved with carbon and thus may be useful in exploring the atomic environment of carbon.

• The korean journal of orthodontics
• /
• v.39 no.2
• /
• pp.112-119
• /
• 2009

Objective: To develop a mixed dentition analysis method in consideration of the normal variation of tooth sizes. Methods: According to the tooth-size of the maxillary central incisor, maxillary 1st molar, mandibular central incisor, mandibular lateral incisor, and mandibular 1st molar, 307 normal occlusion subjects were clustered into the smaller and larger tooth-size groups. Multiple regression analyses were then performed to predict the sizes of the canine and premolars for the 2 groups and both genders separately. For a cross validation dataset, 504 malocclusion patients were assigned into the 2 groups. Then multiple regression equations were applied. Results: Our results show that the maximum errors of the predicted space for the canine, 1st and 2nd premolars were 0.71 and 0.82 mm residual standard deviation for the normal occlusion and malocclusion groups, respectively. For malocclusion patients, the prediction errors did not imply a statistically significant difference depending on the types of malocclusion nor the types of tooth-size groups. The frequency of prediction error more than 1 mm and 2 mm were 17.3% and 1.8%, respectively. The overall prediction accuracy was dramatically improved in this study compared to that of previous studies. Conclusions: The computer aided calculation method used in this study appeared to be more efficient.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.33 no.4
• /
• pp.1581-1589
• /
• 2013

The aim of this study is to suggest a methodology for calculating the peak-hour ratio of passengers at urban railway stations by reflecting the characteristics of passenger demand patterns and the land use inventory of stations. To achieve this, urban railway stations in Seoul are divided into three groups by using factor analysis and cluster analysis. For each station group, we calculate five and four variables related to the passenger demand patterns and the land use inventory of stations, respectively, as well as the peak-hour ratios of passengers. Among these nine variables, average daily passengers and the location quotient (LQ) index for business services are selected as the classification criteria for station groups based on statistical tests. Using the two variables, a group allocation process is suggested to estimate the peak-hour ratio of passengers for a newly-constructed station. Evaluation results based on thirteen stations show that the proposed methodology produces lower errors than the currently-used guideline does. The results of this study contribute to establishing efficiently construction and operation plans for newly-constructed stations.

• Journal of the Korean Vacuum Society
• /
• v.17 no.5
• /
• pp.387-393
• /
• 2008

Using a first-principles total-energy method, we investigate structural and energy changes on Ag/Si(111)$\sqrt{3}{\times}\sqrt{3}$($\sqrt{3}-Ag$ hereafter) as the number of the additional Ag adatoms increases. The Ag coverage varies from 0.02 to 0.14 ML. Most Ag adatoms occupy the ST site, which is the center of small triangles of the substrate Ag layer that is composed of small and large triangles. One of the interesting adsorption features is that the adatoms immerse below the substrate layer. The total energy calculations show that the clusters become the most stable when the number of Ag atoms is three. This three-Ag cluster becomes the building block of the $\sqrt{21}{\times}\sqrt{21}$ phase that shows a large surface conductivity. The simulated STM images show that the adatoms look dark in filled-state images while bright in empty-state images. This suggests that the adatoms donate their charge to the substrate. The simulated STM images agree well with the experimental images.

• Applied Chemistry for Engineering
• /
• v.17 no.4
• /
• pp.349-356
• /
• 2006

We have investigated adsorption and desorption properties of CO adsorption on silica supported Ru/Ni alloys at various Ru/Ni mole content ratio as well as CO partial pressures using Fourier transform infrared spectrometer (FT-IR). For Ru-$SiO_{2}$ sample, four bands were observed at $2080.0cm^{-1}$, $2021.0{\sim}2030.7cm^{-1}$, $1778.9{\sim}1799.3cm^{-1}$, $1623.8cm^{-1}$ on adsorption and three bands were observed at $2138.7cm^{-1}$, $2069.3cm^{-1}$, $1988.3{\sim}2030.7cm^{-1}$ on vacumn desorption. For Ni-$SiO_{2}$ sample, four bands were observed at $2057.7cm^{-1}$, $2019.1{\sim}2040.3cm^{-1}$, $1862.9{\sim}1868.7cm^{-1}$, $1625.7cm^{-1}$ on adsorption and two bands were observed at $2009.5{\sim}2040.3cm^{-1}$, $1828.4{\sim}1868.7cm^{-1}$ on vacumn desorption. These absorption bands correspond with those of the previous reports approximately. For Ru/Ni(9/1, 8/2, 7/3, 6/4, 5/5; mole content ratio)-$SiO_{2}$ samples, three bands were observed at $2001.8{\sim}2057.7cm^{-1}$, $1812.8{\sim}1926.5cm^{-1}$, $1623.8{\sim}1625.7cm^{-1}$ on adsorption and three bands were observed at $2140.6cm^{-1}$, $2073.1cm^{-1}$, $1969.0{\sim}2057.7cm^{-1}$ on vacumn desorption. The spectrum pattern observed for Ru/Ni-$SiO_{2}$ sample at 9/1 Ru/Ni mole content ratio on CO adsorption and on vacumn desorption is almost like the spectrum pattern observed for Ru-$SiO_{2}$ sample. But the spectrum patterns observed for Ru/Ni-$SiO_{2}$ samples under 8/2 Ru/Ni mole content ratio on CO adsorption and vacumn desorption are almost like the pattern observed for $Ni-SiO_{2}$ sample. It may be suggested surfaces of alloy clusters on the Ru/Ni-$SiO_{2}$ samples contain more Ni components than the mole content ratio of the sample considering the above phenomena. With Ru/Ni-$SiO_{2}$ samples the absorption band shifts may be ascribed to variations of surface concentration, strain variation due to atomic size difference, variation of bonding energy and electronic densities, and changes of surface geometries according to surface concentration variation. Studies for CO adsorption on Ru/Ni alloy cluster surface by LEED and Auger spectroscopy, interation between Ru/Ni alloy cluster and $SiO_{2}$, and MO calculation for the system would be needed to look into the phenomena.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.3
• /
• pp.41-48
• /
• 1965

This study was aimed at knowing the magnitude of year variation in rice heading dates under the different seasonal cultures, and to estimate the heading date in advance. Using six rice varieties such as Kwansan, Suwon＃82, Suwon ＃144, Norin＃17, Yukoo＃132 and Paltal, the early, ordinary and late seasonal cultures had been carried out at Paddy Crop Division, Crop Experiment Station at Suwon for the six-year period 1959 to 1964. In addition the data of the standard rice cultures at the Provincial Offices of Rural Development for the 12-year period 1953 to 1954, were analyzed for the purpose of clarifying a relationship between variation of rice heading dates and some of meteorological data related to the locations and years. The results of this study are as follows: 1. Year variation of rice heading dates was as high as 14 to 21 days in the early seasonal culture and 7 to 14 days in the ordinary seasonal culture, while as low as one to seven days in the late seasonal culture which was the lowest among three cultures. The magnitude of variation depended greatly on variety, cultural season and location. 2. It was found out that there was a close negative correlation between the accumulated average air temperature for 40 days from 31 days after seeding and number of days to heading in the early seasonal culture. Accordingly, it was considered possible to predict the rice heading date through calculation of the accumulated average air temperature for the above period and then the linear regression(Y=a+bx). On the other hand, an estimation of the heading date in the late seasonal culture requires for the further studies. In the ordinary seasonal culture, no significant correlation between the accumulated average air temperature and number of days to heading was obtained in the six-year experiments conducted at Suwon. There was a varietal difference in relationship between the accumulated average air temperature for 70 days from seeding and number of days to heading in the standard cultures at the provincial offices of rural development. Some of varieties showed a significant correlation between two factors while the others didn't show any significant correlation. However, there was no regional difference in this relationship.