• Journal of Internet Computing and Services
• /
• v.20 no.4
• /
• pp.91-102
• /
• 2019

In autonomous driving systems, the ability to classify pedestrians in images captured by cameras is very important for pedestrian safety. In the past, after extracting features of pedestrians with HOG(Histogram of Oriented Gradients) or SIFT(Scale-Invariant Feature Transform), people classified them using SVM(Support Vector Machine). However, extracting pedestrian characteristics in such a handcrafted manner has many limitations. Therefore, this paper proposes a method to classify pedestrians reliably and effectively using CNN's(Convolutional Neural Network) deep features and transfer learning. We have experimented with both the fixed feature extractor and the fine-tuning methods, which are two representative transfer learning techniques. Particularly, in the fine-tuning method, we have added a new scheme, called M-Fine(Modified Fine-tuning), which divideslayers into transferred parts and non-transferred parts in three different sizes, and adjusts weights only for layers belonging to non-transferred parts. Experiments on INRIA Person data set with five CNN models(VGGNet, DenseNet, Inception V3, Xception, and MobileNet) showed that CNN's deep features perform better than handcrafted features such as HOG and SIFT, and that the accuracy of Xception (threshold = 0.5) isthe highest at 99.61%. MobileNet, which achieved similar performance to Xception and learned 80% fewer parameters, was the best in terms of efficiency. Among the three transfer learning schemes tested above, the performance of the fine-tuning method was the best. The performance of the M-Fine method was comparable to or slightly lower than that of the fine-tuningmethod, but higher than that of the fixed feature extractor method.

• Korean Chemical Engineering Research
• /
• v.57 no.5
• /
• pp.620-627
• /
• 2019

In this study, ignition delay characteristics of various bio aviation fuels (Bio-ADD, Bio-6308, Bio-7720) produced by HEFA process using different raw materials were compared and analyzed. In order to confirm the feasibility of applying bio aviation fuel to actual system, ignition delay characteristics of petroleum-based aviation fuel (Jet A-1) and blended aviation fuel (50:50, v:v) also analyzed. Ignition delay time of each aviation fuel was measured by using CRU, surface tension measurement and GC/MS and GC/FID analysis were performed to interpret the results. As a result, ignition delay time of Jet A-1 was the longest at all temperature because it contains aromatic compounds about 22.8%. The aromatic compounds can produce benzyl radical which is thermally stable and has low reactivity with oxygen during decomposition process. In the case of bio aviation fuels, ignition delay times were measured similarly because the ratio of n-paraffin/iso-paraffin constituting each aviation fuel is similar (about 0.12) and the composition ratio of cycloparaffin also has no difference. In addition, ignition delay times of blended aviation fuels (50:50, v:v) were measured close to the mean value those of each fuel so it was confirmed that it can be applied without any changing or improving of existing system.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.23 no.4
• /
• pp.391-404
• /
• 2021

This study developed a long-term prediction model for the potential yield of garlic based on a genetic algorithm (GA) by utilizing global climate reanalysis data. The GA is used for digging the inherent signals from global climate reanalysis data which are both directly and indirectly connected with the garlic yield potential. Our results indicate that both deterministic and probabilistic forecasts reasonably capture the inter-annual variability of crop yields with temporal correlation coefficients significant at 99% confidence level and superior categorical forecast skill with a hit rate of 93.3% for 2 × 2 and 73.3% for 3 × 3 contingency tables. Furthermore, the GA method, which considers linear and non-linear relationships between predictors and predictands, shows superiority of forecast skill in terms of both stability and skill scores compared with linear method. Since our result can predict the potential yield before the start of farming, it is expected to help establish a long-term plan to stabilize the demand and price of agricultural products and prepare countermeasures for possible problems in advance.

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.32 no.5
• /
• pp.175-182
• /
• 2022

Rare-earth elements were doped with Mg to enhance the temperature stability of dielectric properties of BaTiO₃ for its application to MLCC (Multi-Layer Ceramic Capacitor). The additives strongly affect both grain growth and densification behaviors during sintering, and hence dielectric properties. The additive effects therefore should be examined in each system with different additives. This study investigated the crystal structure, grain growth and densification behaviors and related variations in dielectric constant with respect to sintering temperature. Dielectric constant appears to be varied with grain size in a temperature range between 1200 and 1300℃, suggesting the importance of grain size control. The temperature dependence of grain size variation was well explained by an established theory correlating the grain growth behavior with grain boundary structure. This accordance provides a basis for sintering technique to control grain growth thus to improve dielectric constant in rare-earth doped BaTiO₃.

• Journal of the Korean Wood Science and Technology
• /
• v.31 no.4
• /
• pp.24-30
• /
• 2003

In order to develop structural 'bamboo zephyr board', properties of boards produced by various methods making zephyr were estimated. All of boards were tested with KS(Korean Standards) F 3014 Particle board, and results were summarized as follow. The zephyr boards produced by final 1.5 mm roller space had the best properties(MOR, MOE and IB strength) among boards produced by different final roller space. Also, they had the best properties(MOR and MOE) in bending test after 1 hour soaking treatment after 2 hour-boiling in water, which were similar to properties of boards in dry-condition. Thickness swelling(TS) of all boards was less than 12% required to the Korean Standards A 3014. Zephyr boards produced by final 1.5 mm roller space also had the lowest values among all of boards. Boards produced by non-treatment (in green condition) had the better strength than those of boards produced by different pre-treatment methods(boiling in water and in 0.3% NaOH) in bending test.

• Korean Journal of Heritage: History & Science
• /
• v.55 no.3
• /
• pp.60-72
• /
• 2022

A number of wooden built heritage remain in Korea, and most have been damaged by various biological factors including termite. Owing to the irreversible damage caused by termites, wooden built heritage are losing their authenticity and structural stability. In this study, Area-Wide Integrated Termite Management(AW-ITM) was proposed to prevent termite damage. First, to understand the locational characteristics of these sites, the distance from adjacent forests and surrounding forest areas was analyzed for 182 national designated wooden built heritage(national treasures, treasures) using the Geographic Information System(GIS). By analyzing existing pest control projects(2003-2020) and the components of the ITM, the characteristics of termite control for cultural heritages were determined. Based on these results, the cultural heritage sites and their surrounding spaces were divided into three areas, and the types of cultural properties were divided into six types according to the location and number of buildings. Along with this, termite control measures were proposed for each area and type. The concept of AW-ITM has been partially applied to the "Comprehensive Control of Termites in wooden built heritages Sites" by the Cultural Heritage Administration. Caution must be taken with regard to the establishment of a cultural heritage management policy; AW-ITM should be applied on a trial basis with the results then being carefully analyzed and reflected in the establishment of policies pertaining to the conservation management of cultural heritage.

• Journal of the Korean Electrochemical Society
• /
• v.25 no.4
• /
• pp.154-161
• /
• 2022

The carbon-neutrality induced by the global warming is important for the modern society. Hydrogen has been received the attention as a new energy source to replace the fossil fuels. Polymer electrolyte membrane fuel cells, which convert the chemical reaction energy of hydrogen into electric power directly, are a type of eco-friendly power for future vehicles. Due to the sluggish oxygen reduction reaction and costly Pt catalyst in the cathode, the research related to the replacement of Pt-based catalysts has been vitally carried out. In this case, however, the performance is significantly different from each other and a variety of factors have existed. In this review paper, we rearrange and summarize relevant papers published within 5 years approximately. The selection of precursors, synthesis method, and co-catalyst are represented as a core factor, while the necessity of research for the further enhancement of activity may be raised. It can be anticipated to contribute to the replacement of precious metal catalysts in the various fields of study. The final objective of the future research is depicted in detail.

• Journal of Industrial Convergence
• /
• v.21 no.1
• /
• pp.1-8
• /
• 2023

Study surveyed perception of dental hygienist occupation was surveyed by examining the qualifications, occupational evaluation Criteria, professional intuition of some college students for the dental hygienist occupation analyzed by the SPSS WIN18.0 program. Among the necessary qualities of dental hygienist, 'professional knowledge of dental hygiene', 'dental hygiene skills' were the highest. Average of Job evaluation Criteria was 4.06, which was higher than the average of 3.80 Job evaluation factors of dental hygienists. As for the Job evaluation criteria, 'stable job', 'job with good work atmosphere and interpersonal relationship' was the highest, 'job with contact with people', 'independent job' was the most. Job evaluation Criteria for dental hygienists, 'jobs with contact with people', 'jobs that can help others' were high, and 'high wages', 'high probability of promotion' were low. As a result of regression analysis of general characteristics and dental hygiene professionalism, it was analyzed that the professionalism of dental hygienists increased when the major was non-health. Overall, In order to improve the qualifications of dental hygienists and raise awareness of vocational evaluation, efforts should be made systematically to identify the strengths and weaknesses of vocational evaluation and dental hygienists and to seek improvements.

• Smart Media Journal
• /
• v.11 no.10
• /
• pp.65-75
• /
• 2022

Deep learning is used as a creative tool that could overcome the limitations of existing analysis models and generate various types of results such as text, image, and music. In this paper, we propose a method necessary to preprocess audio data using the Niko's MIDI Pack sound source file as a data set and to generate music using Bi-LSTM. Based on the generated root note, the hidden layers are composed of multi-layers to create a new note suitable for the musical composition, and an attention mechanism is applied to the output gate of the decoder to apply the weight of the factors that affect the data input from the encoder. Setting variables such as loss function and optimization method are applied as parameters for improving the LSTM model. The proposed model is a multi-channel Bi-LSTM with attention that applies notes pitch generated from separating treble clef and bass clef, length of notes, rests, length of rests, and chords to improve the efficiency and prediction of MIDI deep learning process. The results of the learning generate a sound that matches the development of music scale distinct from noise, and we are aiming to contribute to generating a harmonistic stable music.

• Journal of International Area Studies (JIAS)
• /
• v.15 no.1
• /
• pp.31-60
• /
• 2011

Medvedev administration has adopted new foreign policy line because of Obama administration's 'Reset' policy and Global economic crisis. The objectives of Medvedev's new foreign policy are the creation of a favorable external environment and the efficient use of external resources for Russia's economic modernization. To achieve these goals, Medvedev's government fleshed out such specific action plans as the avoidance of conflicts with other powers, the prevention of conflicts around Russia's borders, the activation of capital investment, and the introduction of advanced technology from the outside. This foreign policy line takes shape in the building of a foundation for strategic cooperation with the United States, the preparation for 'Modernization Alliance' with Europe, the management of cooperation and conflict with China and Japan, and the introduction of a dual strategy of strategic stability and economic integration in relations with post-soviet states. In Russia's new foreign policy line the strengthening of relations with the United States and Europe acquires highest priority. However, this does not mean a return to a 'pro-Western liberal line' in the early 1990s. The ultimate goal of Russia's 'modernization' program still lies in the rebuilding of a powerful Russia in accordance with the multipolar world order that was Putin administration's foreign policy line. In this context, foreign policy change under Medvedev administration could be defined as a 'program change at international level' that signifies a change in the means to achieve goals without changing them.