• Korean Journal of Remote Sensing
• /
• v.30 no.2
• /
• pp.241-249
• /
• 2014

The continuous observations for atmospheric aerosol were carried out during 3 years (2009-2011) by using a multi-wavelength Raman lidar at the Gwangju Institute of Science and Technology (GIST), Korea ($35.11^{\circ}N$, $126.54^{\circ}E$). The particle depolarization ratios were retrieved from the observations in order to distinguish the Asian dust layer. The vertical information of Asian dust layers were used as input parameter for the Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) model for analysis of its backward trajectories. The source regions and transport pathways of the Asian dust layer were identified. The most frequent source region of Asian dust in Korea was Gobi desert during observation period in this study. The statistical analysis on the particle depolarization ratio of Asian dust was conducted according to their transport route in order to retrieve the variation of optical properties of Asian dust during long-range transport. The transport routes were classified into the Asian dust which was transported to observation site directly from the source regions, and the Asian dust which was passed over pollution regions of China. The particle depolarization ratios of Asian dust which were transported via industrial regions of China was ranged 0.07-0.1, whereas, the particle depolarization ratio of Asian dust which was transported directly from the source regions to observation site were comparably higher and ranged 0.11-0.15. It is considered that the pure Asian dust particle from source regions were mixed with pollution particles, which is likely to spherical particle, during transportation so that the values of particle depolarization of Asian dust mixed with pollution was decreased.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.28 no.7
• /
• pp.1216-1221
• /
• 2022

Ships tend to be larger to increase the efficiency of cargo transportation. Larger ships lead to increased travel time for ship workers, increased work intensity, and reduced work efficiency. Problems such as increased work intensity are reducing the influx of young people into labor, along with the phenomenon of avoidance of high intensity labor by the younger generation. In addition, the rapid aging of the population and decrease in the young labor force aggravate the labor shortage problem in the maritime industry. To overcome this, the maritime industry has recently introduced technologies such as an intelligent production design platform and a smart production operation management system, and a smart autonomous logistics system in one of these technologies. The smart autonomous logistics system is a technology that delivers various goods using intelligent mobile robots, and enables the robot to drive itself by using sensors such as lidar and camera. Therefore, in this paper, it was checked whether the mobile robot could autonomously drive to the stop sign by detecting the passage way of the ship deck. The autonomous driving was performed by detecting the passage way of the ship deck through the camera mounted on the mobile robot based on the data learned through Nvidia's End-to-end learning. The mobile robot was stopped by checking the stop sign using SSD MobileNetV2. The experiment was repeated five times in which the mobile robot autonomously drives to the stop sign without deviation from the ship deck passage way at a distance of about 70m. As a result of the experiment, it was confirmed that the mobile robot was driven without deviation from passage way. If the smart autonomous logistics system to which this result is applied is used in the marine industry, it is thought that the stability, reduction of labor force, and work efficiency will be improved when workers work.

• The Korea Journal of Herbology
• /
• v.32 no.5
• /
• pp.57-63
• /
• 2017

Objectives : The aim of this study is to search Korean Medicine with the superior inhibitory effect of pancreatic lipase including a possibility as a well-matched sauce material with pork. Methods : We chose 5 samples which have the potent inhibitory effect of pancreatic lipase enzyme among 9 samples. 5 samples were Glycyrrhizae rhizoma, Cinnamomi cortex, Ganoderma lucidum, Syzygium aromaticum, and Schisandra chinensis. Animals were divided into eight groups (n=7). The experimental groups except for normal group were fed 60% high-fat diet. 5 samples were orally administered at a dose of 200 mg/kg body weight and orlistat were orally administrated at a dose of 60 mg/kg body weight for 7 days. Biochemical anaylses of 5 samples were executed based on lipid parameters such as triglyceride (TG) and total cholesterol (TC). Results : Korean Medicines with $IC_{50}$ of below $1mg/m{\ell}$ were Glycyrrhizae rhizoma, Cinnamomi cortex, Ganoderma lucidum, Syzygium aromaticum, and Schisandra chinensis. Body weight change of all drug-treated groups except Glycyrrhizae rhizoma reduced significantly during the experimental period. Orlistat treatment reduced significantly both TG and TC in serum. The significant emission of TG in fece showed in Cinnamomi cortex treatment, whereas Ganoderma lucidum and Syzygium aromaticum showed a tendency to increase without a significance. Besides, Ganoderma lucidum and Schisandra chinensis emitted significantly TC to fece. Conclusions : In conclusion, Cinnamomi cortex may exert anti-obesity effect by directly inhibiting pancreatic lipase, which would prevent the absorption of lipid from the small intestine. Accordingly, Cinnamomi cortex may suggest a high possibility as a well-matched sauce material with pork.

• Korean Journal of Remote Sensing
• /
• v.28 no.1
• /
• pp.11-19
• /
• 2012

Air-borne pollen, biogenically created aerosol particle, influences Earth's radiative balance, visibility impairment, and human health. The importance of pollens has resulted in numerous experimental studies aimed at characterizing their dispersion and transport, as well as health effects. There is, however, limited scientific information concerning the optical properties of airborne pollen particles contributing to total ambient aerosols. In this study, for the first time, optical characteristics of pollen such as aerosol backscattering coefficient, aerosol extinction coefficient, and depolarization ratio at 532 nm and their effect to the atmospheric aerosol were studied by lidar remotes sensing technique. Dual-Lidar observations were carried out at the Gwangju Institute of Science & Technology (GIST) located in Gwagnju, Korea ($35.15^{\circ}E$, $126.53^{\circ}N$) for a spring pollen event from 5 to 7 May 2009. The pollen concentration was measured at the rooftop of Gwangju Bohoon hospital where the building is located 1.0 km apart from lidar site by using Burkard trap sampler. During intensive observation period, high pollen concentration was detected as 1360, 2696, and $1952m^{-3}$ in 5, 6, and 7 May, and increased lidar return signal below 1.5km altitude. Pollen optical depth retrieved from depolarization ratio was 0.036, 0.021, and 0.019 in 5, 6, and 7 May, respectively. Pollen particles mainly detected in daytime resulting increased aerosol optical depth and decrease of Angstrom exponent.

• Korean Journal of Remote Sensing
• /
• v.37 no.6_2
• /
• pp.1793-1801
• /
• 2021

From 2015 to June 2020, the backscattering coefficients of 532 and 1064 nm measured using LIDAR and the depolarization ratio at 532 nm were used to separate the backscattering coefficient at 532 nm as three types as PM₁₀, PM_2.5-10, PM_2.5 according to particle size. The mass extinction efficiency (MEE) of three types was calculated using the mass concentration measured on the ground. The overall mean values of the calculated MEE were 5.1 ± 2.5, 1.7 ± 3.7, and 9.3 ± 6.3 m²/g in PM₁₀, PM_2.5-10, and PM_2.5, respectively. When the mass concentration of PM₁₀ and PM_2.5 was low, higher than average MEE was calculated, and it was confirmed that the MEE decreased as the mass concentration increased. When the MEE was calculated for each type according to the mixing degree of Asian dust, PM_2.5-10 was twice at pollution aerosol as high as 2.1 ± 2.8 m²/g, compare to pollution-dominated mixture, dust-dominated mixture, and pure dust of 1.1 ± 1.8, 1.4 ± 3.3, 1.1 ± 1.5 m²/g, respectively. However, PM_2.5 MEE showed similar values irrespective of type: 9.4 ± 6.5, 9.0 ± 5.8, 10.3 ± 7.5, and 9.1 ± 9.0 m²/g. The MEE of PM₁₀ was 5.6 ± 2.9, 4.4 ± 2.0, 3.6 ± 2.9, and 2.8 ± 2.4 m²/g in pollution aerosol (PA), pollution-dominated mixture (PDM), dust-dominated mixture (DDM), and pure dust (PD), respectively, and increased as the dust ratio value decreased. Even if the same type according to the same mass concentration or Asian dust mixture was shown, as the PM_2.5/PM₁₀ ratio decreased, the MEE of PM_2.5-10 decreased and the MEE of PM_2.5 showed a tendency to increase.

• Journal of Intelligence and Information Systems
• /
• v.23 no.2
• /
• pp.1-17
• /
• 2017

The deep learning framework is software designed to help develop deep learning models. Some of its important functions include "automatic differentiation" and "utilization of GPU". The list of popular deep learning framework includes Caffe (BVLC) and Theano (University of Montreal). And recently, Microsoft's deep learning framework, Microsoft Cognitive Toolkit, was released as open-source license, following Google's Tensorflow a year earlier. The early deep learning frameworks have been developed mainly for research at universities. Beginning with the inception of Tensorflow, however, it seems that companies such as Microsoft and Facebook have started to join the competition of framework development. Given the trend, Google and other companies are expected to continue investing in the deep learning framework to bring forward the initiative in the artificial intelligence business. From this point of view, we think it is a good time to compare some of deep learning frameworks. So we compare three deep learning frameworks which can be used as a Python library. Those are Google's Tensorflow, Microsoft's CNTK, and Theano which is sort of a predecessor of the preceding two. The most common and important function of deep learning frameworks is the ability to perform automatic differentiation. Basically all the mathematical expressions of deep learning models can be represented as computational graphs, which consist of nodes and edges. Partial derivatives on each edge of a computational graph can then be obtained. With the partial derivatives, we can let software compute differentiation of any node with respect to any variable by utilizing chain rule of Calculus. First of all, the convenience of coding is in the order of CNTK, Tensorflow, and Theano. The criterion is simply based on the lengths of the codes and the learning curve and the ease of coding are not the main concern. According to the criteria, Theano was the most difficult to implement with, and CNTK and Tensorflow were somewhat easier. With Tensorflow, we need to define weight variables and biases explicitly. The reason that CNTK and Tensorflow are easier to implement with is that those frameworks provide us with more abstraction than Theano. We, however, need to mention that low-level coding is not always bad. It gives us flexibility of coding. With the low-level coding such as in Theano, we can implement and test any new deep learning models or any new search methods that we can think of. The assessment of the execution speed of each framework is that there is not meaningful difference. According to the experiment, execution speeds of Theano and Tensorflow are very similar, although the experiment was limited to a CNN model. In the case of CNTK, the experimental environment was not maintained as the same. The code written in CNTK has to be run in PC environment without GPU where codes execute as much as 50 times slower than with GPU. But we concluded that the difference of execution speed was within the range of variation caused by the different hardware setup. In this study, we compared three types of deep learning framework: Theano, Tensorflow, and CNTK. According to Wikipedia, there are 12 available deep learning frameworks. And 15 different attributes differentiate each framework. Some of the important attributes would include interface language (Python, C ++, Java, etc.) and the availability of libraries on various deep learning models such as CNN, RNN, DBN, and etc. And if a user implements a large scale deep learning model, it will also be important to support multiple GPU or multiple servers. Also, if you are learning the deep learning model, it would also be important if there are enough examples and references.

• Pediatric Infection and Vaccine
• /
• v.19 no.3
• /
• pp.121-130
• /
• 2012

Purpose : The purpose of this study is to investigate clinical features and causative organisms in febrile infants younger than three months, to help identification of high risk patients for serious bacterial infection (SBI). Methods : A total of 313 febrile infants younger than three months, who had visited Seoul National University Children's Hospital from January 2008 to December 2010 were included. Clinical features, laboratory findings, causative organisms, and risk factors of SBI were analyzed by retrospective chart review. Causative bacterial or viral pathogens were identified by gram stain and cultures, rapid antigen tests, or the polymerase chain reaction from clinically reliable sources. Results : Among 313 infants, etiologic organisms were identified in 127 cases (40.6%). Among 39 cases of bacterial infections, Escherichia coli (66.7%) and Streptococcus agalactiae (12.8%) were common. Enterovirus (33.7%), respiratory syncytial virus (19.8%), and rhinovirus (18.8%) were frequently detected in 88 cases of viral infection. Patients with SBI (39 cases) showed significantly higher values of the white blood cell count ($14,473{\pm}6,824/mm^3$ vs. $11,254{\pm}5,775/mm^3$, P=0.002) and the C-reactive protein ($6.32{\pm}8.51mg/L$ vs. $1.28{\pm}2.35mg/L$, P<0.001) than those without SBI (274 cases). The clinical risk factors for SBI were the male (OR 3.7, 95% CI 1.5-8.9), the presence of neurologic symptoms (OR 4.8, 95% CI 1.4-16.8), and the absence of family members with respiratory symptoms (OR 3.6, 95% CI 1.2-11.3). Conclusion : This study identified common pathogens and risk factors for SBI in febrile infants younger than three months. These findings may be useful to guide management of febrile young infants.

• Korean Journal of Social Welfare
• /
• v.57 no.2
• /
• pp.5-40
• /
• 2005

Social welfare centers and residential care facilities where provide the socially disadvantaged with proper social services, face financial difficulties. This is because not only of the lack of governmental support, but also of social welfare centers and residential care facilities' lack of skills in developing abundant resources from the private sector. In this context, this study tried to find factors affecting resource mobilization of the social welfare facilities to devise policies in resource development. Mail survey was conducted with the structured questionnaire. Employees in charge of community resource development were asked to answer the questionnaire. The study population were welfare centers and residential care facilities. A total of 293 community welfare centers and 632 residential care facilities responded to the survey. The response rate was about 62%. The dependent variables of the study were the amount of resource mobilization in the year 2001 which was measured as the number of donors, the total amount of donation, and estimated amount of gift-in-kind. Three types models were constructed per each welfare institution. Independent variables were selected based on the previous research findings: community environment factor, structural factor, and resource development factor. Multiple regression was utilized to analyze the data. The resource development factor turned out to be significant variable in various models. In the models of donors, the amount of donation, and the amount of gift-in-kind (except for the welfare center model), at least one out of six variables of the resource development factors was significant welfare center. Welfare centers which establish the resource development department or hire employees to take care of resource development, utilize computer softwares to file donors, and utilize donor management programs, have more donors and/or donations than their counterparts. In addition, residential care facilities located in urban area have more donors and donations, and among residential facilities those for the disables, those with longer history and more employees, receive more donations than their counterparts. As for the gift-in-kind model, the welfare centers located in high income area and residential care facilities for the elderly, children and mentally retarded receive less gift-in-kind than their counterparts Based on the above findings, this study suggested that to mobilize resources the welfare centers as well residential care facilities need to have community resource development department or resource development staffs, adopt computer software to systematically organize donors, and utilize donor mobilizing and maintaining programs.