• Journal of Conservation Science
• /
• v.37 no.6
• /
• pp.723-737
• /
• 2021

This study analyzed the composition and structure of materials with Buddhist paper scriptures excavated from architectural sites in 'Sum Tolgoi' of the 17^th century and carried out conservation treatment base on the result of the analysis. The scriptures were covered in dust and foreign sub stances, and were so crumpled that it was impossible to identify the form. The damage, loss, and discoloration have been identified. Buddhist scriptures written in Tibetan used indigo and ink sticks on paper as a result of UV-Vis analysis, and ink sticks as black character materials from scriptures written ancient Mongolian. SEM-EDS and Micro-XRF analyses revealed that the outlines were drawn with red lines using a mix of Minium (Pb₃O₄) and Cinnabar (HgS), or Cinnabar (HgS) alone, and the contents of the scriptures were written with silver paint. Silver chloride (AgCl) and Calcium (Ca) were identified in the silver paint component of the characters, while Calcium and Orpiment (As₂S₃) were identified in the yellow lines. Concerning the paper ground, Buddhist scriptures written in ancient Mongolian were characterized by herbal plant fiber and bast fiber, and those written in Tibetan, by bast fiber. Radiocarbon dating indicates that the paper for the scriptures was produced between the 15^th and 17^th centuries. Conservation treatment of the scriptures was carried out based on the experiment on the production of pre-coated paper and how to coat that to prevent the second damage due to the deformation and fragility of the excavated paper. The scriptures were preserved and mounted, and a neutral box was made to identify the contents of the scriptures recorded on both sides after the treatment. This conservation treatment is the result of a study that applied new conservation treatment materials and methods according to the principle of conservation treatment reversibility.

• Journal of Nutrition and Health
• /
• v.55 no.2
• /
• pp.211-226
• /
• 2022

Choline, an essential nutrient for humans, is required for the structural integrity of the cell membranes, methyl-group metabolism, synthesis of the neurotransmitter acetylcholine, synthesis of the membrane phospholipid components of the cell membranes, and the transport of lipids and cholesterol. Choline can be synthesized in the body, but it is insufficient to meet the daily requirements and hence it must be obtained through the diet. In the United States/Canada, Australia/New Zealand, Europe, China, and Taiwan, the adequate intake (AI) and tolerable upper intake level (UL) of choline have been established, while the establishment of the 2020 Dietary Reference Intakes for Koreans (KDRI) for choline was postponed due to the lack of a choline database for Korean foods and studies on the choline intake of Koreans. However, as part of the preparation work for the 2020 DRI revision and finalization, choline intake and the possibility of disease occurrence were verified through analysis of published data. The groundwork for the subsequent establishment of a choline DRI was laid through a literature search, evaluation, and review of the literature reported from 1949 up to 2019. This can be regarded as the culmination of this project. According to the results of randomized controlled trials (RCTs), cohort studies, case-control studies, and cross-sectional observational studies in humans, approximately 400-500 mg/day of choline intake was effective in preventing liver function damage (fatty liver), neural tube damage, cardiovascular disease, breast cancer, and cognitive function improvement. The same amount of choline intake, however, also correlated with the risk of prostate and colorectal cancer. At present, there is limited information available on choline intake and health outcomes, particularly for the Korean population. More human studies, including clinical trials on the requirements and the physiological benefits associated with dietary intake, are needed to establish the KDRI for choline.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.6
• /
• pp.927-933
• /
• 2022

We synthesized (C₁₀H₈N₂H)₂Cr₂O₇, The structure of the product was characterized with FT-IR(infrared) and elemental analysis. The oxidation of benzyl alcohol by (C₁₀H₈N₂H)₂Cr₂O₇ in organic solvents showed that the reactivity increased with the increase of the dielectric constant. The oxidation of alcohols was examined by (C₁₀H₈N₂H)₂Cr₂O₇ in DMF, acetone. As a resuit, (C₁₀H₈N₂H)₂Cr₂O₇ was found as efficicent oxidizing agent that converted benzyl alcohol, allyl alcohol, primary alcohol and secondary alcohols to the corresponding aldehydes or ketones(65%~95%). The selective oxidation of alcohols was also examined by (C₁₀H₈N₂H)₂Cr₂O₇ in DMF, acetone. (C₁₀H₈N₂H)₂Cr₂O₇ was selective oxidizing agent(15%~95%) of benzyl alcohol, allyl alcohol and primary alcohol in the presence of secondary ones. In the presence of DMF solvent with acidic catalyst such as H₂SO₄. (C₁₀H₈N₂H)₂Cr₂O₇ oxidized benzyl alcohol(H) and its derivatives. The Hammett reaction constant(ρ) was -0.69(308K). The observed experimental data were used to rationalize the hydride ion transfer in the rate determining step.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.40 no.2
• /
• pp.260-267
• /
• 1995

In temperate zone planting rice at different date subjects the Crop to different climatic condition. The present study aimed at comparison of the change in source-sink relationship of the Japonica(J) and that of IndicaxJaponica(I$\times$J) type rice cultivars caused by shift of heading date. Two J- and two I$\times$J-type cultivars were made to head on August 16, August 26, and September 5. Sink capacity was changed by shift of heading date in different mode between the types of cultivars. In both types major determinant of sink capacity was number of effective tillers, and the number of spikelets per panicle was the minor. In J-type earlier planting/heading was beneficial to increased panicle numbers and this was due mainly to a larger diurnal difference in temperature. I$\times$J-type cultivars favored a higher daily mean temperature to increase the sink capacity. The ability of source at heading, in terms of leaf area per panicle, chlorophyll content per spiklet, photosynthetic ability of leaves per unit area at 25$\^{\circ}C$, carbohydrate and N contents of leaves, was not so different among different heading dates in both types. However, the source activity was governed principally by temperature during grain filling. The J-type cultivars headed on Sept. 5 and I$\times$J-type cultivars headed later than August 16 could not have had sufficient source activity in grain filling due to lower temperature.

• Journal of the Korean Society of Radiology
• /
• v.17 no.4
• /
• pp.607-614
• /
• 2023

Clear overlapping of the bilateral epicondyle and proper separation of the elbow joint are crucial for obtaining accurate lateral general radiographs of the elbow. However, due to the complex anatomical structure of the elbow, achieving optimal positioning is challenging, leading to the need for repeated x-ray examinations. Therefore, the purpose of this study was to investigate the angle of the forearm in patients where accurate lateral images of the elbow joint can't be obtained after vertical incidence using a styrofoam device during elbow joint lateral x-ray imaging. Twenty patients were enrolled in our study following the established protocol. First, a vertical x-ray at an angle of 0° between the forearm and the table was taken (control group). Here, if the lateral image of the elbow joint was deemed inadequate, the forearm angle was adjusted using custom-made styrofoam supports with 5° and 10° inclinations (experimental groups). For the evaluation method, two assessors utilized a 5-point Likert scale to assess the images. The reliability of the assessments was analyzed using Cronbach's alpha coefficient. As a result, patients with inadequate overlap of the bilateral epicondyle and separation of the elbow joint in the initial examination (control group) were able to obtain the best images when setting a 10° angle between the forearm and the table. The subjective evaluation was 1.6 ± 0.8 points at 0°, 2.7 ± 0.8 points at 5°, and 4.4 ± 1.3 points at 10°, respectively. The reliability analysis for the angles of 0°, 5°, and 10° yielded Cronbach's alpha values of 0.867, 0.697, and 0.922, respectively. In conclusion, when it is not possible to obtain accurate images using the conventional position and X-ray beam direction, it is considered that by initially acquiring images with an angle of 10° between the forearm and the table, and gradually decreasing the angle while obtaining images, it would be possible to achieve the optimal image while reducing the number of repeat examinations.

• Journal of the Korea Organic Resources Recycling Association
• /
• v.31 no.1
• /
• pp.47-56
• /
• 2023

This study describes the successful synthesize strategy for the silica nanoparticles utilizing various water sources, including tap, industrial, and stream waters without using deionized water. Also, as-synthesized silica nanoparticles are employed as dispersive materials for the electro-responsive smart fluid application. Specifically, homogeneous silica nanoparticles with sizes of 500-700nm are successfully prepared in large scale at once (ca. 12.0 g) with the described experimental method and showing similar structural and chemical characteristics with silica nanoparticles synthesized using the deionized water. The size of silica nanoparticles are varied according to the ion conductivity differences of tap, industrial, stream water, and deionized water. The size of silica nanoparticles decresed with the increased ion conductivity, indicating the ion suppression of growth of silica nanoparticles. Moreover, as-synthesized silica nanoparticles from various water sources of electro-responsive characteristic are investigated by the smart fluid application. The smart fluids containing silica nanoparticles synthesized by tap, industrial, and stream water exhibited higher shear stress compared to the deionized water, owing to the more rigid fibril-like structures formed by the smaller silica nanoparticles. Conclusively, uniform silica nanoparticles from various water sources without any purification are able to successfully prepared without usage of deionized water and resulting silica nanoparticles manifested higher electro-responsive performance.

• Proceedings of the Korean Society of Tobacco Science Conference
• /
• 1997.10a
• /
• pp.134-152
• /
• 1997

Plant viruses of tobacco including tobacco mosaic virus (TMV) and potato virus Y (PVY) cause severe economic losses in leaf-tobacco production. Cultural practices do not provide sufficient control against the viruses. Use of valuable resistant cultivars is most recommendable for the control of the viruses. However, conventional breeding programs are not always proper for the development of virus-resistant plants mostly owing to the frequent lack of genetic sources and introduction of their unwanted properties. Therefore, we tried to develop virus-resistant tobacco plants by transforming commercial tobacco cultivars, NC 82 and Burley 21, with coat protein (CP) or replicase (Nlb) genes of TMV and PVY necrosis strain (PVY-VN) with or without untranslated region (UTR) and with or without mutation. Each cDNA was cloned and inserted in plant expression vectors with 1 or 2 CaMV 35S promotors, and introduced into tobacco leaf tissues by Agrobacterium tumefaciens LBA 4404. Plants were regenerated in kanamycin-containing MS media. Regenerated plants were tested for resistance to TMV and PVY In these studies, we could obtain a TMV-resistant transgenic line transformed with TMV CP and 6 genetic lines with PVY-VN cDNAs out of 8 CP and replicase genes. In this presentation, resistance rates, verification of gene introduction in resistant plants, stability of resistance through generations, characteristics of viral multiplication and translocation in resistant plants, and resistance responses relative to inoculum potential and to various PVY strains will be shown. Yield and quality of leaf tobacco of a promising resistant tobacco line will be presented.

• Clean Technology
• /
• v.29 no.4
• /
• pp.255-261
• /
• 2023

Power semiconductors are semiconductors used for power conversion, transformation, distribution, and control. Recently, the global demand for high-voltage power semiconductors is increasing across various industrial fields, and optimization research on high-voltage IGBT components is urgently needed in these industries. For high-voltage IGBT development, setting the resistance value of the wafer and optimizing key unit processes are major variables in the electrical characteristics of the finished chip. Furthermore, the securing process and optimization of the technology to support high breakdown voltage is also important. Etching is a process of transferring the pattern of the mask circuit in the photolithography process to the wafer and removing unnecessary parts at the bottom of the photoresist film. Ion implantation is a process of injecting impurities along with thermal diffusion technology into the wafer substrate during the semiconductor manufacturing process. This process helps achieve a certain conductivity. In this study, dry etching and wet etching were controlled during field ring etching, which is an important process for forming a ring structure that supports the 3.3 kV breakdown voltage of IGBT, in order to analyze four conditions and form a stable body junction depth to secure the breakdown voltage. The field ring ion implantation process was optimized based on the TEG design by dividing it into four conditions. The wet etching 1-step method was advantageous in terms of process and work efficiency, and the ring pattern ion implantation conditions showed a doping concentration of 9.0E13 and an energy of 120 keV. The p-ion implantation conditions were optimized at a doping concentration of 6.5E13 and an energy of 80 keV, and the p+ ion implantation conditions were optimized at a doping concentration of 3.0E15 and an energy of 160 keV.

• Journal of Korea Water Resources Association
• /
• v.57 no.7
• /
• pp.437-449
• /
• 2024

Rainfall-runoff prediction studies using deep learning while considering catchment attributes have been gaining attention. In this study, we selected two models: the Transformer model, which is suitable for large-scale data training through the self-attention mechanism, and the LSTM-based multi-state-vector sequence-to-sequence (LSTM-MSV-S2S) model with an encoder-decoder structure. These models were constructed to incorporate catchment attributes and predict the inflow of 10 multi-purpose dam watersheds in South Korea. The experimental design consisted of three training methods: Single-basin Training (ST), Pretraining (PT), and Pretraining-Finetuning (PT-FT). The input data for the models included 10 selected watershed attributes along with meteorological data. The inflow prediction performance was compared based on the training methods. The results showed that the Transformer model outperformed the LSTM-MSV-S2S model when using the PT and PT-FT methods, with the PT-FT method yielding the highest performance. The LSTM-MSV-S2S model showed better performance than the Transformer when using the ST method; however, it showed lower performance when using the PT and PT-FT methods. Additionally, the embedding layer activation vectors and raw catchment attributes were used to cluster watersheds and analyze whether the models learned the similarities between them. The Transformer model demonstrated improved performance among watersheds with similar activation vectors, proving that utilizing information from other pre-trained watersheds enhances the prediction performance. This study compared the suitable models and training methods for each multi-purpose dam and highlighted the necessity of constructing deep learning models using PT and PT-FT methods for domestic watersheds. Furthermore, the results confirmed that the Transformer model outperforms the LSTM-MSV-S2S model when applying PT and PT-FT methods.

• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.205-225
• /
• 2018

Convolutional Neural Network (ConvNet) is one class of the powerful Deep Neural Network that can analyze and learn hierarchies of visual features. Originally, first neural network (Neocognitron) was introduced in the 80s. At that time, the neural network was not broadly used in both industry and academic field by cause of large-scale dataset shortage and low computational power. However, after a few decades later in 2012, Krizhevsky made a breakthrough on ILSVRC-12 visual recognition competition using Convolutional Neural Network. That breakthrough revived people interest in the neural network. The success of Convolutional Neural Network is achieved with two main factors. First of them is the emergence of advanced hardware (GPUs) for sufficient parallel computation. Second is the availability of large-scale datasets such as ImageNet (ILSVRC) dataset for training. Unfortunately, many new domains are bottlenecked by these factors. For most domains, it is difficult and requires lots of effort to gather large-scale dataset to train a ConvNet. Moreover, even if we have a large-scale dataset, training ConvNet from scratch is required expensive resource and time-consuming. These two obstacles can be solved by using transfer learning. Transfer learning is a method for transferring the knowledge from a source domain to new domain. There are two major Transfer learning cases. First one is ConvNet as fixed feature extractor, and the second one is Fine-tune the ConvNet on a new dataset. In the first case, using pre-trained ConvNet (such as on ImageNet) to compute feed-forward activations of the image into the ConvNet and extract activation features from specific layers. In the second case, replacing and retraining the ConvNet classifier on the new dataset, then fine-tune the weights of the pre-trained network with the backpropagation. In this paper, we focus on using multiple ConvNet layers as a fixed feature extractor only. However, applying features with high dimensional complexity that is directly extracted from multiple ConvNet layers is still a challenging problem. We observe that features extracted from multiple ConvNet layers address the different characteristics of the image which means better representation could be obtained by finding the optimal combination of multiple ConvNet layers. Based on that observation, we propose to employ multiple ConvNet layer representations for transfer learning instead of a single ConvNet layer representation. Overall, our primary pipeline has three steps. Firstly, images from target task are given as input to ConvNet, then that image will be feed-forwarded into pre-trained AlexNet, and the activation features from three fully connected convolutional layers are extracted. Secondly, activation features of three ConvNet layers are concatenated to obtain multiple ConvNet layers representation because it will gain more information about an image. When three fully connected layer features concatenated, the occurring image representation would have 9192 (4096+4096+1000) dimension features. However, features extracted from multiple ConvNet layers are redundant and noisy since they are extracted from the same ConvNet. Thus, a third step, we will use Principal Component Analysis (PCA) to select salient features before the training phase. When salient features are obtained, the classifier can classify image more accurately, and the performance of transfer learning can be improved. To evaluate proposed method, experiments are conducted in three standard datasets (Caltech-256, VOC07, and SUN397) to compare multiple ConvNet layer representations against single ConvNet layer representation by using PCA for feature selection and dimension reduction. Our experiments demonstrated the importance of feature selection for multiple ConvNet layer representation. Moreover, our proposed approach achieved 75.6% accuracy compared to 73.9% accuracy achieved by FC7 layer on the Caltech-256 dataset, 73.1% accuracy compared to 69.2% accuracy achieved by FC8 layer on the VOC07 dataset, 52.2% accuracy compared to 48.7% accuracy achieved by FC7 layer on the SUN397 dataset. We also showed that our proposed approach achieved superior performance, 2.8%, 2.1% and 3.1% accuracy improvement on Caltech-256, VOC07, and SUN397 dataset respectively compare to existing work.