• International Journal of Concrete Structures and Materials
• /
• v.11 no.1
• /
• pp.59-68
• /
• 2017

Mercury intrusion and nitrogen sorption porosimetry were employed to investigate the pore structure of calcium sulfoaluminate ($C{\bar{S}}A$) and portland cement pastes with cement-to-water ratio (w/c) of 0.40, 0.50, and 0.60. A unimodal distribution of pore size was drawn for $C{\bar{S}}A$ cement pastes, whereas a bimodal distribution was established for the portland cement pastes through analysis of mercury intrusion porosimetry. For the experimental results generated by nitrogen sorption porosimetry, the $C{\bar{S}}A$ cement pastes have a smaller and coarser pore volume than cement paste samples under the same w/c condition. The relative dynamic modulus and percentage weight loss were used for investigation of the concrete durability in freeze-thaw condition. When coarse aggregate with good freeze-thaw durability was mixed, air entrained portland cement concrete has the same durability in terms of relative dynamic modulus as $C{\bar{S}}A$ cement concrete in a freeze-thaw environment. The $C{\bar{S}}A$ cement concrete with poor performance of durability in a freeze-thaw environment demonstrates the improved durability by 300 % over portland cement concrete. The $C{\bar{S}}A$ concrete with good performance aggregate also exhibits less surface scaling in a freeze-thaw environment, losing 11 % less mass after 297 cycles.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.1
• /
• pp.90-95
• /
• 2009

This paper presents a new feature extraction algorithm which can deal with the problems of linear discriminant analysis, widely used for linear dimensionality reduction. The scatter matrices included in linear discriminant analysis are defined by the distances between each datum and its class mean, and those between class means and mean of whole data. Use of these scatter matrices can cause computational problems and the limitation on the number of features. In addition, these definition assumes that the data distribution is unimodal and normal, for the cases not satisfying this assumption the appropriate features are not achieved. In this paper we define a new scatter matrix which is based on the differently weighted distances between individual data, and presents a feature extraction algorithm using this scatter matrix. With this new method. the mentioned problems of linear discriminant analysis can be avoided, and the features appropriate for discriminating data can be achieved. The performance of this new method is shown by experiments.

• Journal of the Society of Cosmetic Scientists of Korea
• /
• v.39 no.3
• /
• pp.225-231
• /
• 2013

Oil in water nanoemulsion of astaxanthin was prepared by high pressure homogenization. The emulsifying conditions including emulsifier type, concentration and astaxanthin concentration were optimized. Stability of nanoemulsion was measured using zetasizer, freeze-fracture scanning electron microscope (FF-SEM), particle analyzer and colorimeter. The mean diameter of the dispersed particles containing astaxanthin ranged from 160 to 190 nm. Size distribution was unimodal and extended from 40 to 200 nm. The nanoemulsion prepared by glyceryl citrate/lactate/linoleate/oleate had smaller particle size and narrow size distribution. Stable incorporation of astaxanthin in nanoemulsion was performed and checked using high performance liquid chromatography (HPLC), freeze-fracture scanning electron microscope (FF-SEM). Physical stability of nanoemulsion was not significantly changed during storage at both light and thermal condition for a month with zeta potential value of -41 mV meaning stable colloid.

• Particle and aerosol research
• /
• v.9 no.1
• /
• pp.7-21
• /
• 2013

Twelve-hour size-resolved atmospheric aerosols were measured to determine size distributions of water-soluble organic carbon(WSOC) during daytime and nighttime, and to investigate sources and formation pathways of WSOC in individual particle size classes. Mass, WSOC, ${NO_3}^-$, $K^+$, and $Cl^-$ at day and night showed mostly bimodal size distributions, peaking at the size range of $0.32-0.55{\mu}m$(condensation mode) and $3.1-6.2{\mu}m$(coarse mode), respectively, with a predominant condensation mode and a minor coarse mode. While ${NH_4}^+$ and ${SO_4}^{2-}$ showed unimodal size distributions which peaked between 0.32 and $0.55{\mu}m$. WSOC was enriched into nuclei mode particles(< $0.1{\mu}m$) based on the WSOC-to-mass and WSOC-to-water soluble species ratios. The sources and formation mechanisms of WSOC were inferred in reference to the size distribution characteristics of inorganic species(${SO_4}^{2-}$, ${NO_3}^-$, $K^+$, $Ca^{2+}$, $Na^+$, and $Cl^-$) and carbon monoxide. Nuclei mode WSOC was likely associated with primary combustion sources during daytime and nighttime. Among significant sources contributing to the condensation mode WSOC were homogeneous gas-phase oxidation of VOCs, primary combustion emissions, and fresh(or slightly aged) biomass burning aerosols. The droplet mode WSOC could be attributed to aqueous oxidation of VOCs in clouds, cloud-processed biomass burning aerosols, and small contributions from primary combustion sources. From the correlations between WSOC and soil-related particles, and between WSOC and sea-salt particles, it is suggested that the coarse mode WSOC during daytime is likely to condense on the soil-related particles($K^+$ and $Ca^{2+}$), while the WSOC in the coarse fraction during nighttime is likely associated with the sea-salt particles($Na^+$).

• Geotechnical Engineering
• /
• v.2 no.3
• /
• pp.67-78
• /
• 1986

This paper presents how to determine the system reliability of a typical shallow foundation constituted four potential correlated failure modes of hearing capacity (BCM), consolidation settlement (CSM), moment (MFM), and tension shear (PCM). Through the idenfication of the distinct and different modes and evaluation of range of system reliability, the obtained conclusions are as follows; 1. The CSM and the PCM are the lowest and highest of reliability indices of single performance function, and the BCM and the MFM are medium of them. 2. For the correlated failure modes, the hi-modal bounds Is narrower and lower of failure probability than the unimodal bounds. Not to be overestimated, therefore, the system reliability should be based on the second-order bounds using correlated performance functions.

• Journal of Information Processing Systems
• /
• v.16 no.1
• /
• pp.6-29
• /
• 2020

Biometrics identification using multiple modalities has attracted the attention of many researchers as it produces more robust and trustworthy results than single modality biometrics. In this paper, we present a novel multimodal recognition system that trains a deep learning network to automatically learn features after extracting multiple biometric modalities from a single data source, i.e., facial video clips. Utilizing different modalities, i.e., left ear, left profile face, frontal face, right profile face, and right ear, present in the facial video clips, we train supervised denoising auto-encoders to automatically extract robust and non-redundant features. The automatically learned features are then used to train modality specific sparse classifiers to perform the multimodal recognition. Moreover, the proposed technique has proven robust when some of the above modalities were missing during the testing. The proposed system has three main components that are responsible for detection, which consists of modality specific detectors to automatically detect images of different modalities present in facial video clips; feature selection, which uses supervised denoising sparse auto-encoders network to capture discriminative representations that are robust to the illumination and pose variations; and classification, which consists of a set of modality specific sparse representation classifiers for unimodal recognition, followed by score level fusion of the recognition results of the available modalities. Experiments conducted on the constrained facial video dataset (WVU) and the unconstrained facial video dataset (HONDA/UCSD), resulted in a 99.17% and 97.14% Rank-1 recognition rates, respectively. The multimodal recognition accuracy demonstrates the superiority and robustness of the proposed approach irrespective of the illumination, non-planar movement, and pose variations present in the video clips even in the situation of missing modalities.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.26 no.3
• /
• pp.266-278
• /
• 1993

Ecological and taxonomical studies on the copepods were reviewed in order to dake data bases on the species composition, abundance, distributional pattern, and seasonal variations of copepods in the South Sea of Korea. Total 179 species have been reported in this area. The dominant species in the coastal waters and embayments, such as Acartia clausi, Paracalanus parvus, Oithona similis, and Oithona nana, were different from those reported in the offshore waters around Cheju Island. Indicator species of the South Sea were also discussed in this paper. Copepod abundances were higher in the coastal waters, up to more than 390,000 individuals/$m^3$(collected with a $150{\mu}m$ mesh net), than in the offshore waters. There were two types of temporal variations in copepod abundances in the coastal waters, i.e., bimodal abundance peaks in spring and fall, md unimodal peak during summer to early fall.

• The Korean Journal of Malacology
• /
• v.26 no.3
• /
• pp.245-254
• /
• 2010

For the studies of germ cell development and maturation in the ovary, the gametogenic cycle and the number of spawning seasons per year in female Ruditapes philippinarum were investigated by quantitative statistical analysis using an Image Analyzer System. Compared with the results by qualitative and quantitative analyses, monthly variations in female gonad indice by qualitative histological analysis showed a pattern similar to that of the female gametogenic cycle calculated by quantitative statistical analysis. The number of spawning seasons occurred once per year, from June to October. In quantitative statistical analysis using an image analyzer system, monthly changes in the portions (%) of the ovary area to total tissue areas in females increased in March and reached a maximum in May, and then showed a rapid decrease from June to October when spawning occurred. And also monthly changes in portions (%) of follicle areas to the ovary area and in portions of oocyte areas to ovarian tissue areas in females began to increase in March and reached a maximum in May, and then. rapidly dropped from June to October when spawnig occurred. From these data, it is apparent that the number of spawning seasons occurred once per year, from June to October. Monthly changes in the number of the oocyte per $mm^2$ and in mean diameter of the oocyte in captured image which were calculated for each female slide showed a maximum in May and reached the minimum from December to February. Therefore, female R. philippinarum showed a unimodal gametogenic cycle during the year.

• Journal of Medicine and Life Science
• /
• v.18 no.2
• /
• pp.25-30
• /
• 2021

Many people reported suspected food poisoning after consuming food at the same snack bar on June 18, 2020. Thus, an in-depth epidemiological investigation was conducted to identify the infectious agent and establish additional food poisoning prevention measures. The study included people who reported to the local public health center after June 18 with acute gastroenteritis symptoms within 4 days of consuming food from the snack bar. The onset of symptoms and food items consumed by individuals were then investigated via phone calls and on-site visits. Afterward, the infectious agent was identified from human samples (stool or rectal swab) of four restaurant employees and 89 people and from environmental samples (materials, cooking utensils, and water). The analysis revealed that the incubation period ranged from 2 hours to 92 hours, with a median and mode of 16 hours and 12 hours, respectively. Moreover, the epidemic curve had a unimodal shape because of common exposure, which reached its peak on June 18. After monitoring for 8 days, which is more than twice the maximum incubation period of 92 hours, the end of the epidemic was declared on June 28 as no additional cases were reported. Analysis of human and environmental samples revealed Salmonella bareilly of the pulsed-field gel electrophoresis pulsotype SAPX01.017 as the causative agent. Therefore, it was concluded that the food poisoning outbreak was caused by S. bareilly.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.1
• /
• pp.27-32
• /
• 2019

Research that implements the classification process through Deep Learning algorithm, one of the outstanding human abilities, includes a unimodal model, a multi-modal model, and a multi-modal method using music videos. In this study, the results were better by suggesting a system to analyze each song's spectrum into short samples and vote for the results. Among Deep Learning algorithms, CNN showed superior performance in the category of music genre compared to RNN, and improved performance when CNN and RNN were applied together. The system of voting for each CNN result by Deep Learning a short sample of music showed better results than the previous model and the model with Softmax layer added to the model performed best. The need for the explosive growth of digital media and the automatic classification of music genres in numerous streaming services is increasing. Future research will need to reduce the proportion of undifferentiated songs and develop algorithms for the last category classification of undivided songs.