• Korean Journal of Radiology
• /
• v.24 no.7
• /
• pp.698-714
• /
• 2023

In recent years, artificial intelligence, especially object detection-based deep learning in computer vision, has made significant advancements, driven by the development of computing power and the widespread use of graphic processor units. Object detection-based deep learning techniques have been applied in various fields, including the medical imaging domain, where remarkable achievements have been reported in disease detection. However, the application of deep learning does not always guarantee satisfactory performance, and researchers have been employing trial-and-error to identify the factors contributing to performance degradation and enhance their models. Moreover, due to the black-box problem, the intermediate processes of a deep learning network cannot be comprehended by humans; as a result, identifying problems in a deep learning model that exhibits poor performance can be challenging. This article highlights potential issues that may cause performance degradation at each deep learning step in the medical imaging domain and discusses factors that must be considered to improve the performance of deep learning models. Researchers who wish to begin deep learning research can reduce the required amount of trial-and-error by understanding the issues discussed in this study.

• ALGAE
• /
• v.39 no.2
• /
• pp.59-74
• /
• 2024

Terrestrial cyanobacteria are extremely diverse. In urban areas, they can be found as black stains on the surface of building walls, stone monuments, or man-made structures. Many of the terrestrial cyanobacteria are still understudied. To expand knowledge of terrestrial cyanobacterial diversity, a polyphasic characterization was performed to identify 12 strains isolated from campus of University of the Ryukyus, Okinawa, Japan. Multigene phylogenetic analyses based on 16S rRNA gene and 16S-23S rRNA internal transcribed spacer (ITS) region showed that the isolated strains formed two independent subclades within Nodosilinea, and were distantly related to all described Nodosilinea species. The 16S-23S rRNA ITS secondary structures showed variations for D1-D1' and Box B domain, while V3 domain was almost identical among entire species of Nodosilinea, including the studied strains. In addition, a unique morphological character, i.e. forming nodule or spiral shape, was also observed in certain studied strains. According to polyphasic characterization, Nodosilinea coculeatus sp. nov. and Nodosilinea terrestrialis sp. nov., were proposed as two new species of terrestrial cyanobacteria from Okinawa.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.60 no.1
• /
• pp.1-8
• /
• 2024

This study examined the power consumption of angling boats during entry, departure, and fishing operations using a black box-type storage device. Through this analysis, it determined the energy consumption and carbon emissions of small fishing boats used for catching the largehead hairtail. The energy consumption and carbon emissions were calculated using formulas provided by the Korea Energy Agency, which incorporated updated emission coefficients from 2022. The findings revealed that the average power consumption of small fishing boats for the largehead hairtail was 546.3 kWh, with a total energy consumption of 0.1164 TOE and carbon emissions of 24.057 CO₂. The average energy consumption was calculated at 0.0006 TOE per kilogram, and the carbon emissions were determined to be 0.135 CO₂/kg.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.27 no.2
• /
• pp.256-264
• /
• 2024

In the last 10 years, AI has made rapid progress, and image classification, in particular, are showing excellent performance based on deep learning. Nevertheless, due to the nature of deep learning represented by a black box, it is difficult to actually use it in critical decision-making situations such as national defense, autonomous driving, medical care, and finance due to the lack of explainability of judgement results. In order to overcome these limitations, in this study, a model description algorithm capable of local interpretation was applied to the inception network-derived AI to analyze what grounds they made when classifying national defense data. Specifically, we conduct a comparative analysis of explainability based on confidence values by performing LIME analysis from the Inception v2_resnet model and verify the similarity between human interpretations and LIME explanations. Furthermore, by comparing the LIME explanation results through the Top1 output results for Inception v3, Inception v2_resnet, and Xception models, we confirm the feasibility of comparing the efficiency and availability of deep learning networks using XAI.

• Korean Journal of Heritage: History & Science
• /
• v.52 no.4
• /
• pp.38-55
• /
• 2019

At Otani University in Kyoto, Japan, there is a rubbed copy of Tripitaka Koreana, presumably printed in 1381. According to the postscript of the copy, written by Saek Lee himself, the rubbed copy was made at Haeinsa temple in 1381 and was kept at Sinluksa temple in Yeuju. The copy was delivered as a gift to Japan in 1414 and now is kept at the Library of Otani University. Although an approximate summary of the content of the copy was reported in the early 2000s after a basic survey, details of the copy, including the concrete format and packaging paper, are not known yet. In this paper a detailed survey of the copy is conducted on the 109 pages. The copy is divided into two parts: the wrapping and the inner pages. The wrapping paper is divided into yellow and brown colors depending on the material of the paper. The yellow colorwrapping paper was possibly made in 1381 at the time of the rubbed printing, and the brown wrapping paper was repaired after being moved to Japan. Using funds collected in February 1380, the copy of Gyeong(經), Yul(律), and Ron(論) chapters was printed in April 1381. Binding of the copy was completed in September, and the wrapping paper with the title in gold was made in October 1380. The box for keeping Buddhist scriptures was manufactured in November 1380. The copy was moved to Sinluksa temple in April 1382 and delivered to Japan in 1414. At Otani University, the copy is stored in separate rectangular boxes 32.1×25.3cm in size with a height of 23.6cm. The rectangular plate on the four sides is red in external color but black colorinside. The box for keeping Buddhist scriptures was probably made in 1381, but a partial repair was made later. Because of the difficulty of executing a detailed survey of the box for Buddhist scriptures, it is hard to find out its nation and period of production. We look forward to studying the copy as well as the box for Buddhist scriptures in future.

• Journal of Internet Computing and Services
• /
• v.9 no.1
• /
• pp.115-127
• /
• 2008

The component for satisfying several domain requirements must be developed to support variety. But, when the application is developed using the component, it happens other requirements. So, it is difficult to design component to satisfy severaldomain requirements. Also, it is impossible to support the special business logic. As this problem, the component must provide to the white-box component, it is not the black-box component. So, in this paper, we propose the variability design technique and the customization technique using the design technique that can support the various requirements. This technique is not focus on designing the component to analyze various domains. The variability of the component is designed to the initial variability in the component development phase and we use the customization technique for applying the variability to developing application. The variability can be re-designed during developing the application to use the component applying the variability. The variability of the component is evolved and the generation of the component is increased via the iteration Generally, the range of the component variability is classified the function modification within the component and the component internal structure modification as requirements in the component outside. As the range of the variability, we propose the variability design technique of the behavior and the message flow. This paper proposes a message flow design technique for modifying function call.

• Korean Journal of Heritage: History & Science
• /
• v.42 no.3
• /
• pp.176-191
• /
• 2009

This report is a result of investigating the materials and techniques of the lacquer ware excavated from Hwangnamdaechong Tomb known as an old tomb in the Gyeongju maximum district. First, the fragment of lacquer ware was classified by the shape of it. Next, the thin section of lacquer film (all sides about 2mm) was made based on the result. Then they were observed under the microscope, and classified the kind of the foundation mixture and the applied structure. It is a result as follows. 1 The bone dust was mixed by the foundation of the made of cloth-body lacquer ware such as eared lacquer cup, lacquer bowl, and the lacquer rectangular containers excavated from the south tomb. 2 Eared lacquer cup and lacquer bowl, and rectangular box excavated from south and north tomb had the possibility they were the set of articles partially of tableware and the cosmetic box. 3 There is a difference in the quality of burial lacquer ware between in south tomb and north tomb. A lot of highlevel lacquer ware such as eared lacquer cup, lacquer bowl and the rectangular lacquer containers were discovered from the south tomb. 4 Many of containers where animal and bird had been drawn on the black lacquered ground excavated from north tomb were wooden bath.

• Journal of Korean Society of Transportation
• /
• v.32 no.4
• /
• pp.369-379
• /
• 2014

The Statistical regression model has been used to construct crash prediction models, despite its limitations in assuming data distribution and functional form. In response to the limitations associated with the statistical regression models, a few studies based on non-parametric methods such as neural networks have been proposed to develop crash prediction models. However, these models have a major limitation in that they work as black boxes, and therefore cannot be directly used to identify the relationships between crash frequency and crash factors. A genetic programming model can find a solution to a problem without any specified assumptions and remove the black box effect. Hence, this paper investigates the application of the genetic programming technique to develope the crash prediction model. The data collected from the Gyeongbu expressway during the past three years (2010-2012), were separated into straight and curve sections. The random forest technique was applied to select the important variables that affect crash occurrence. The genetic programming model was developed based on the variables that were selected by the random forest. To test the goodness of fit of the genetic programming model, the RMSE of each model was compared to that of the negative binomial regression model. The test results indicate that the goodness of fit of the genetic programming models is superior to that of the negative binomial models.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.9 no.3
• /
• pp.119-128
• /
• 2014

One of the theoretically and empirically grounded black boxes in HRD and firm performance link is employee' attitudes such as organizational commitment and job satisfaction. However, most studies were conducted with the regression analysis at the organizational level. This study used HLM(hierarchical linear modeling) analysis, which made it possible to estimate more accurate relationship between variables that were measured from two different levels. In addition, this study attempted to open an the black box(learning culture) in the relationship between HRD and employee attitudes. The result showed that the HRD have a positive effect on the organizational commitment and the job satisfaction. Also the HRD showed full mediation effect of organization commitment and the job satisfaction on the Learning culture. And the result showed that the HRD in 2007 have a positive effect on employee' attitudes in 2009. These findings concluded that systematic HRD like employee's education and training must be built and also the positive culture for employee's learning like support of management's learning organization must be improved in order to promote the organizational performance(organizational commitment, job satisfaction) in company.

• Animal Bioscience
• /
• v.34 no.8
• /
• pp.1392-1402
• /
• 2021

Objective: The growth rate of pigs is related to differentiation and proliferation of muscle cells, which are regulated by growth factors and expression of growth-related genes. Thus, the objective of this study was to establish optimal culture conditions for Jeju black pig (JBP) muscle cells and determine the relationship of various factors involved in muscle growth with the proliferation of JBP muscle cells. Methods: Muscles were taken from the femur skeletal muscle of JBP embryos. After isolation of the muscle cells, cells were cultured in a 6-well plate under four different culture conditions to optimize culture conditions for JBP muscle cells. To analyze proliferation rate of JBP muscle cells, these muscle cells were seeded into 6-well plates at a density of 1.5×10⁵ cells per well and cultured for 3 days. Western blot and quantitative real-time polymerase chain reaction were applied to verify the myogenic differentiation 1 (MyoD) expression and growth-related gene expression in JBP muscle cells, respectively. Results: We established a muscle cell line from JBP embryos and optimized its culture conditions. These muscle cells were positive for MyoD, but not for paired box 7. The proliferation rate of these muscle cells was significantly higher in a culture medium containing bFGF and epidermal growth factor + basic fibroblast growth factor (EGF+bFGF) than that without a growth factor or containing EGF alone. Treatment with EGF and bFGF significantly induced the expression of MyoD protein, an important transcription factor in muscle cells. Moreover, we checked the changes of expression of growth-related genes in JBP muscle cells by presence or absence of growth factors. Expression level of collagen type XXI alpha 1 gene was changed only when EGF and bFGF were added together to culture media for JBP muscle cells. Conclusion: Concurrent use of EGF and bFGF increased the expression of MyoD protein, thus regulating the proliferation of JBP muscle cells and the expression of growth-related genes.