• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.3
• /
• pp.1086-1103
• /
• 2020

The automatic identification and classification of image-based weld defects is a difficult task due to the complex texture of the X-ray images of the weld defect. Several depth learning methods for automatically identifying welds were proposed and tested. In this work, four different depth convolutional neural networks were evaluated and compared on the 1631 image set. The concavity, undercut, bar defects, circular defects, unfused defects and incomplete penetration in the weld image 6 different types of defects are classified. Another contribution of this paper is to train a CNN model "RayNet" for the dataset from scratch. In the experiment part, the parameters of convolution operation are compared and analyzed, in which the experimental part performs a comparative analysis of various parameters in the convolution operation, compares the size of the input image, gives the classification results for each defect, and finally shows the partial feature map during feature extraction with the classification accuracy reaching 96.5%, which is 6.6% higher than the classification accuracy of other existing fine-tuned models, and even improves the classification accuracy compared with the traditional image processing methods, and also proves that the model trained from scratch also has a good performance on small-scale data sets. Our proposed method can assist the evaluators in classifying pipeline welding defects.

• Asian-Australasian Journal of Animal Sciences
• /
• v.15 no.5
• /
• pp.675-681
• /
• 2002

Among Italian buffalo farmers, it is widely held that administering diets with high energy and protein concentrations is an effective way to increase milk production. In order to assess the validity of this opinion, we verified milk yield and physico-chemical characteristics from buffaloes that, from the $5^{th}$ month of lactation, were fed two total mixed rations (TMRs) which, given the same intake, should have led to satisfaction of protein requirements though with a slight energy deficit (diet A) or excessive amounts of energy and protein (diet B). Estimate of the energy and protein value of the diets and that of the corresponding requirements was carried out both by using two software programs derived from the Cornell Net Carbohydrate and Protein System (1992), and with the method set up by INRA researchers (1988). The results obtained show that the two diets administered did not result in significant changes to the quantity of milk produced. However, with Diet B the protein concentration in the milk was significantly (p<0.01) higher, although this was partly offset by the higher concentration (p<0.05) of non-protein nitrogen (NNP). The Group B buffaloes also showed significantly higher blood urea levels (p<0.01), with concentrations exceeding those considered physiological for lactating buffaloes. Finally, while administering Diet A the Body Condition Score (BCS) was close to 6.5 (Wagner et al., 1988), whereas in buffaloes which used Diet B it sometimes increased by over 0.5 points. As regards which of the two methods compared is more suitable for expressing dietary energy and protein value and corresponding requirements, we feel that due to the high variability in the Italian Mediterranean buffalo's milk production aptitude, it would be premature to express a judgement on methods which rest on a common scientific base and do not differ substantially.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.13 no.5
• /
• pp.55-61
• /
• 2018

The main purpose of this study is to test the effects of education and training invest of firm performance and to find the strategy of human resources development to industry include company and government, each stakeholder etc. For this purpose, this study empirical tested the causality, effects between education and training and firm performance. Using the HCCP panel data set, panel analysis model is performed with the dependent variables. The results of the study are summarized as follows. First, education and training has a significant impact on the firm sales per employee head. But there is no evidence for the effect of education and training on the net profit per employee head. These study results indicate that firms have to increase the concern and invest for education and training.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.6B
• /
• pp.483-488
• /
• 2012

With the fast growing data traffic, the performance of the convergent overlay network environment under the cognitive networking environment is crucial for the implementation of the efficient network structure. In order to achieve high capacity and reliable link quality in wireless communication of the overlay convergent networks with the cognitive networking based on the advanced capability of the mobile terminal, a Distributed Wireless Communication System (DWCS) can provide the capability of ambient-aware dual cell system's operation. This paper has considered virtual cell: the Dual Virtual Cell (DVC), and also proposes DVC employment strategy based on DWCS network. One is the Active Virtual Cell which exists for user's actual data traffic and the other is the Candidate Virtual Cell which contains a set of candidate antennas to protect user's link quality from performance degradation or interruption. The considered system constructs DVC by using cognitive ability of finding useful virtual signals. Also, for multi-user high-rate data transmission, the DWCS system exploits Space-Time Trellis Codes. The effects of changing environments on the system performance has been investigated thereafter.

• Smart Media Journal
• /
• v.9 no.3
• /
• pp.122-129
• /
• 2020

The possibility that an asymptotic coronavirus-19 infected person around the world is not aware of his infection and can spread it to people around him is still a very important issue in that the public is not free from anxiety and fear over the spread of the epidemic. In this paper, the K-means clustering algorithm and deep learning-based crowd aggregation were proposed to determine the possibility of contact with confirmed cases of infectious diseases. As a result of 300 iterations of all input learning images, the PSNR value was 21.51, and the final MAE value for the entire data set was 67.984. This means the average absolute error between observations and the average absolute error of fewer than 4,000 people in each CCTV scene, including the calculation of the distance and infection rate from the confirmed patient and the surrounding persons, the net group of potential patient movements, and the prediction of the infection rate.

• Journal of Biomedical Engineering Research
• /
• v.41 no.5
• /
• pp.185-194
• /
• 2020

Arrhythmia is basically diagnosed with the electrocardiogram (ECG) signal, however, ECG is difficult to measure and it requires expert help in analyzing the signal. On the other hand, the radial pulse can be measured with easy and uncomplicated way in daily life, and could be suitable bio-signal for the recent untact paradigm and extensible signal for diagnosis of Korean medicine based on pulse pattern. In this study, we developed an irregular radial pulse detection algorithm based on a learning model and considered its applicability as arrhythmia screening. A total of 1432 pulse waves including irregular pulse data were used in the experiment. Three data sets were prepared with minimal preprocessing to avoid the heuristic feature extraction. As classification algorithms, elastic net logistic regression, random forest, and extreme gradient boosting were applied to each data set and the irregular pulse detection performances were estimated using area under the receiver operating characteristic curve based on a 10-fold cross-validation. The extreme gradient boosting method showed the superior performance than others and found that the classification accuracy reached 99.7%. The results confirmed that the proposed algorithm could be used for arrhythmia screening. To make a fusion technology integrating western and Korean medicine, arrhythmia subtype classification from the perspective of Korean medicine will be needed for future research.

• Journal of Korean Society for Quality Management
• /
• v.44 no.4
• /
• pp.771-784
• /
• 2016

Purpose: The purpose of this study is to analyze the efficiency and productivity of the quality by integrating the product quality and service quality of global automobile brands from the customer's perspective. Methods: In this study, the data from JD Power and GoodCarBadCar.net were used to analyze the efficiency and productivity of a total of 24 automobile brands (10 luxury brands and 14 mainstream brands) between 2009 and 2013. For this, DEA (Data Envelopment Analysis) and MPI (Malmquist Productivity Index) were used. Results: The mean efficiency of the quality of global automobile brands were 0.725 for luxury brands and 0.587 for mainstream brands, which suggests generally higher efficiency for luxury brands. The productivity of the quality of global automobile brands increased by 16.1% for luxury brands while it decreased by 3.1% for mainstream brands. Conclusion: The study provides a theoretical implication in that it emphasized the efficiency of the quality viewed from the customer's perspective, and investigated the quality of the product and that of service in an integrative manner. In addition, this study provides also a practical implication in that it suggests how to set the sales goal by the brand and how to manage according to the characteristics of the brand to the managers of automobile manufacturers.

• Health Policy and Management
• /
• v.26 no.3
• /
• pp.172-184
• /
• 2016

Background: The objective of this study was to examine the effect of occurrence and reoccurrence of catastrophic health expenditure (CHE) on transition to poverty and persistence of poverty in South Korea. Methods: The data of the year 2008-2011 from the Korea Health Panel were used. CHE was defined as the share of total health expenditure in a household out of a household's total income at various threshold levels (more than 5%, 10%, 15%, and 20%). The effect of catastrophic expenditure on transition to poverty and persistence of poverty was analyzed through multivariate logistic regression. Results: The shares of households facing CHE at various threshold levels have increased gradually with 37.7%, 21%, 13.1%, and 9.5% in 2011. Households facing CHE were more likely to experience transition to poverty at thresholds level of more than 5% and 20% in 2010 set. Households facing CHE seemed to experience persistence of poverty, but it was not statistically significant. About 40% of households facing CHE in 2009 encountered another shock of CHE in 2010. Households without CHE seemed to experience more transition to poverty and persistence of poverty, but it was not statistically significant. For household with multiple CHE, those with medical aid were more likely to experience transition to poverty with statistical significance, but the statistical significance disappeared in case of persistence of poverty. Conclusion: The Korean health system needs to be improved to serve as a social security net for addressing transition to poverty and persistence of poverty due to facing CHE.

• Water for future
• /
• v.22 no.1
• /
• pp.81-90
• /
• 1989

A numerical simulation technique of three-dimensional finite difference model is developed to study the groundwater flow system in Dejima, an upland area which faces Kasumigaura Lake. For general perspectives of the groundwater flow system, a steady state three-dimensinal model is simulated. For the sedimentary mud formations which are found in the representative formation, three situations where the horizontal permeability is equal to 10 times and 100 times of the vertical one. The finite difference grid used in the simulation has 60$\times$50$\times$30=90,000 nodes. A converged solution with a tolerance of 0.001 meter of hydraulic head is set. Having determined the flow net by using a steady state three-dimensional model, the results for the three cases of hydraulic conductivity are compared with the results of the tracer methods (Bae and Kayane, 1987). With the aid of four representative vertical cross-sections, groundwater flow systems in the study area are assumed. The results of these are reasonally good comparable with the results of the basin yield and thd recharge-discharge distribution 8.

• Journal of Hydrogen and New Energy
• /
• v.22 no.1
• /
• pp.118-127
• /
• 2011

An IGCC was evaluated as one of the next generation technologies that would be able to substitute for coal-fired power plants. According to "The 4th Basic Plan of Long-term Electricity Supply & Demand" which is developed by the Electricity Business Acts, the first IGCC will be operated at 2015. Like other new and renewable energy such as solar PV, Fuel cell, The IGCC is considered as non-competitive generation technology because it is not maturity technology. Before the commercial operation of an IGCC in our electricity market, its economic feasibility under the Korean electricity market, which is cost-based trading system, is studied to find out institutional support system. The results of feasibility summarized that under the current electricity trading system, if the IGCC is considered like a conventional plant such as nuclear or coal-fired power plants, it will not be expected that its investment will be recouped within life-time. The reason is that the availability of an IGCC will plummet since 2016 when several nuclear and coal-fired power plants will be constructed additionally. To ensure the reasonable return on investment (NPV>0 IRR>Discount rate), the availability of IGCC should be higher than 77%. To do so, the current electricity trading system is amended that the IGCC generator must be considered as renewable generators to set up Price Setting Schedule and it should be considered as pick load generators, not Genco's coal fired-generators, in the Settlement Payment.