• Journal of Multimedia Information System
• /
• v.6 no.4
• /
• pp.225-234
• /
• 2019

Deep learning has become increasingly popular in both academic and industrial areas nowadays. Various domains including pattern recognition, Computer vision have witnessed the great power of deep neural networks. However, current studies on deep learning mainly focus on quality data sets with balanced class labels, while training on bad and imbalanced data set have been providing great challenges for classification tasks. We propose in this paper a method of data analysis-based data reduction techniques for selecting good and diversity data samples from a large dataset for a deep learning model. Furthermore, data sampling techniques could be applied to decrease the large size of raw data by retrieving its useful knowledge as representatives. Therefore, instead of dealing with large size of raw data, we can use some data reduction techniques to sample data without losing important information. We group PCB characters in classes and train deep learning on the ResNet56 v2 and SENet model in order to improve the classification performance of optical character recognition (OCR) character classifier.

• 한국지구물리탐사학회:학술대회논문집
• /
• 2003.11a
• /
• pp.335-339
• /
• 2003

The estimation of seawater intrusion into deep aquifers has been becoming an important subject in terms of site characterization for geological disposal of radioactive waste. Conventional direct-current resistivity methods have been used for ground water explorations and recently have been applied to environmental problems. However, electromagnetic methods are more practical and useful for such a deep investigation. We consider audio-frequency magnetotelluric (AMT) and surface-to-borehole electromagnetic (EM) tomography methods as promising tools for the investigation of deep aquifer. These methods were tested in the Hasunuma area, Chiba Prefecture, Japan. Although the study area is in an urban area, high-quality AMT data were acquired, which was mainly accomplished by night-time data recording and remote-reference data processing. One-dimensional inversion results of the AMT data revealed two extremely conductive zones, which is consistent with the electrical conductivity profile of pore water in core samples. It can be interpreted as the seawater intrusions into both zones. However, the chemical analysis of the groundwater sampled in the deep zone suggests that this groundwater must be fossil seawater that had been confined during sedimentation processes. In addition, the permeability coefficient of the deep layer is very low. Thus the deep conductive zone corresponds to the fossil seawater regarded as being difficult to flow.

• Transactions of Materials Processing
• /
• v.24 no.1
• /
• pp.5-12
• /
• 2015

In the current study, fundamental deep drawing characteristics of Ti-6Al-4V alloy sheets were investigated to establish the effect of processing conditions on large size square deep drawn cups. To accomplish this study, FE-simulations (Abaqus) were performed to determine optimum blank size, friction coefficient, the gap between punch and die, etc. The simulated processing parameters were verified experimentally. Based on the FE-simulation results, deep drawing was performed with various blank holding loads and sample sizes. In order to improve the formability of Ti-6Al-4V sheet, various lubricant methods were evaluated. Tensile tests and thickness measurements were conducted on the formed sheets. Processing parameters including blank holding force, lubricants, and optimum blank size, were selected to achieve improved drawing quality. With the optimum processing condition, a $200mm{\times}200mm$ cup was deep drawn successfully.

• Journal of the Korean Society for information Management
• /
• v.34 no.4
• /
• pp.7-32
• /
• 2017

Recently, deep learning has been rapidly spreading as an innovative machine learning technique in various domains. This study explored the research trends of deep learning via modified ego centered topic citation analysis. To do that, a few seed documents were selected from among the retrieved documents with the keyword 'deep learning' from Web of Science, and the related documents were obtained through citation relations. Those papers citing seed documents were set as ego documents reflecting current research in the field of deep learning. Preliminary studies cited frequently in the ego documents were set as the citation identity documents that represents the specific themes in the field of deep learning. For ego documents which are the result of current research activities, some quantitative analysis methods including co-authorship network analysis were performed to identify major countries and research institutes. For the citation identity documents, co-citation analysis was conducted, and key literatures and key research themes were identified by investigating the citation image keywords, which are major keywords those citing the citation identity document clusters. Finally, we proposed and measured the citation growth index which reflects the growth trend of the citation influence on a specific topic, and showed the changes in the leading research themes in the field of deep learning.

• The Korean Journal of Food And Nutrition
• /
• v.22 no.2
• /
• pp.261-271
• /
• 2009

Deep sea water exists at depths of over 200m under the sea. As no sunlight reaches it, photosynthesis does not take place within it, and it contains no organic matter. In addition, its temperature is maintained at a stable low level throughout the year, so it does not get mixed with the sea water on the surface. It contains a large amount of nutritious salts, whose cleanness is maintained. It is a marine resource that has matured for a long period of time. Research into deep sea water, which started in the 1970s, has been made around the whole world, including the USA and Japan. In Korea, research has been active in this area since 2000. As there has been a good amount of research into industrial applications for deep sea water, since 1993, patents for the relevant technologies have been applied. This paper intends to provide a resource to researchers of deep sea water, by summarizing of all domestic deep sea water-related patents applied with Korean Intellectual Property Office from 1993 to 2008. This research was conducted using a computer and KIPRIS Database owned by the Korea Institute of Patent Information. 'Deep sea water' was used as the search keyword. A total of 222 Korean patents relating to deep sea water have been registered on the basis of IPC. Of these, 126 patents relate to the manufacturing and the treatment of foods, foodstuffs, or non-alcoholic beverages(A23L), while 50 patents relate to the production for medical, dental, or cosmetic purposes(A61K). 38 patents relate to water purification, treatment of wastewater, sewage and sludge (C02F), while 8 patents relate to fishery and farming(A01K). In summary, it was found that studies for the practical use of deep sea water have been conducted in relation to the manufacturing and the treatment of foods, foodstuffs, beverages, and cosmetics.

• Journal of Ocean Engineering and Technology
• /
• v.26 no.1
• /
• pp.78-83
• /
• 2012

The abundant nutrients contained in deep seawater are delivered by natural upwellings from the deep sea to the surface sea. However, the natural upwelling phenomenon is limited to specific areas of the sea; in other areas, the thermocline separates the surface sea from the lower layer. Thus, the surface layer is often deficient in nutritive salts, causing the deterioration of its primary productivity and ultimately leading to an imbalance in the marine ecosystem. Without a consistent supply of nitrogenous nutritive salts, they are absorbed by phytoplankton, resulting in a considerable problem in primary productivity. To solve this issue, a floating type of artificial upwelling system is suggested to artificially pump up, distribute, and diffuse deep seawater containing rich nutritive salts. The key technologies for developing such a floating artificial upwelling system are a floating offshore structure with a large diameter riser, self-supplying energy system, density current generating system, method for estimating the emission and absorption of CO2, and way to evaluate the primary production variation. Strengthening the primary production of the sea by supplying deep seawater to the sea surface will result in a sea environment with abundant fishery resources.

• Electrical & Electronic Materials
• /
• v.8 no.4
• /
• pp.426-433
• /
• 1995

Effect of boron in GaAs have been investigated by photo induced current transient spectroscopy(PICTS). The starting material was undoped liquid encapsulated Czochralski(LEC) semi insulating GaAs and boron ion implantation at 150keV energy was conducted with dose of 10$\^$12/ and 10$\^$13/ions/cm$\^$2/. In ion implanted samples, the peaks related arsenic vacancy(V$\_$As/) were decreased but complex lattice defect was increased with annealing temperature. U band was observed at ion implanted(10$\^$13/ ions/cm$\^$2/) and thermally treated(550.deg. C) sample. More negative peak was detected after annealing at temperature between 600 and 700.deg. C. The measurement of dark current showed that the formation of B$\_$GA/-V$\_$As/, complex defect and complex lattice defect by ion implantation were a reasonable explanation for the decrease in dark current.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.11 no.5
• /
• pp.521-530
• /
• 2018

This paper proposes "The Agriculture Decision-making System(ADS) based on Deep Learning for improving crop productivity" that collects weather information based on location supporting precision agriculture, predicts current crop condition by using the collected information and real time crop data, and notifies a farmer of the result. The system works as follows. The ICM(Information Collection Module) collects weather information based on location supporting precision agriculture. The DRCM(Deep learning based Risk Calculation Module) predicts whether the C, H, N and moisture content of soil are appropriate to grow specific crops according to current weather. The RNM(Risk Notification Module) notifies a farmer of the prediction result based on the DRCM. The proposed system improves the stability because it reduces the accuracy reduction rate as the amount of data increases and is apply the unsupervised learning to the analysis stage compared to the existing system. As a result, the simulation result shows that the ADS improved the success rate of data analysis by about 6%. And the ADS predicts the current crop growth condition accurately, prevents in advance the crop diseases in various environments, and provides the optimized condition for growing crops.

• East Asian Economic Review
• /
• v.16 no.4
• /
• pp.363-394
• /
• 2012

WTO Government Procurement Agreement (GPA) was designed to liberalize and expand trade in government procurement. Revised GPA was implemented in 1996 and the latest revision was completed (but not yet implemented) in 2012, but as a plurilateral agreement. Since the end of the UR, there has been attempts by various WTO members to liberalize trade in the government procurement market - through an expansion of Parties who are signatories to GPA, and through a negotiated agreement on transparency in government procurement. The attempt to expand the Parties who are signatories to the GPA - attempt to increase the width of the coverage of the agreement - has been somewhat successful, but I argue that the goal should be to further liberate the government procurement markets of the current Party members - to reduce thresholds and other barriers which limit market access even to other GPA members, in other words, to increase the depth of coverage. Taking cue from Korea's FTA, I propose a two-level liberalization of the government procurement market under the GPA: A "light" level which would be the same as the current level of liberalization; and a "deep" level with lower thresholds and less exemptions. I argue that, as seen in Korea, with FTAs, many GPA Parties already have multiple levels of liberalization (i.e, spaghetti-bowl effect of FTAs), but by limiting the levels of liberalization to two, we can seek the best of deep liberalization but reduce the spaghetti-bowl effect.

• Journal of the Korea Convergence Society
• /
• v.10 no.6
• /
• pp.235-242
• /
• 2019

Recently, the use of artificial intelligence technology including deep learning has become active in various fields. In particular, several algorithms showing superior performance in object recognition and detection based on deep learning technology have been presented. In this paper, we propose the proper direction for the implementation of mobile healthcare application that user's convenience is effectively reflected. By effectively analyzing the current state of use satisfaction research for the existing fitness applications and the current status of mobile healthcare applications, we attempt to secure survival and superiority in the fitness application market, and, at the same time, to maintain and expand the existing user base.