• Journal of Korean Tunnelling and Underground Space Association
• /
• v.21 no.3
• /
• pp.419-432
• /
• 2019

Most of deep learning model training was proceeded by supervised learning, which is to train labeling data composed by inputs and corresponding outputs. Labeling data was directly generated manually, so labeling accuracy of data is relatively high. However, it requires heavy efforts in securing data because of cost and time. Additionally, the main goal of supervised learning is to improve detection performance for 'True Positive' data but not to reduce occurrence of 'False Positive' data. In this paper, the occurrence of unpredictable 'False Positive' appears by trained modes with labeling data and 'True Positive' data in monitoring of deep learning-based CCTV accident detection system, which is under operation at a tunnel monitoring center. Those types of 'False Positive' to 'fire' or 'person' objects were frequently taking place for lights of working vehicle, reflecting sunlight at tunnel entrance, long black feature which occurs to the part of lane or car, etc. To solve this problem, a deep learning model was developed by simultaneously training the 'False Positive' data generated in the field and the labeling data. As a result, in comparison with the model that was trained only by the existing labeling data, the re-inference performance with respect to the labeling data was improved. In addition, re-inference of the 'False Positive' data shows that the number of 'False Positive' for the persons were more reduced in case of training model including many 'False Positive' data. By training of the 'False Positive' data, the capability of field application of the deep learning model was improved automatically.

• Korean Journal of Construction Engineering and Management
• /
• v.22 no.2
• /
• pp.3-11
• /
• 2021

Myanmar is an undeveloped country with high development value among Asian countries. Therefore, various countries including the U.S. are considering entering the market. In this respect, demand for real estate development project is forecast to grow on increased inflow of foreigners and Myanmar's economic growth. However, Myanmar is a high-risk country in terms of overseas companies, including national risk. In this study, we conducted an in-depth interview with experts (law, finance, technology, and local experts) after analyzing data on Myanmar to extract risk-causing factors. Through this, 106 risk factors were extracted, and the final risk classification system was established by conducting three-time groupings using the affinity diagramming. And the relative importance of each factor was presented using the analytic hierarchy process (AHP) technique. As a result, the country-related risk, the fund-related risk, and the pre-sale-related risk were highly important. The research results are expected to provide risk management standards to companies entering the Myanmar real estate development type project.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.28 no.1
• /
• pp.141-152
• /
• 2022

As interest increases related to the development of eco-friendly energy, the offshore wind turbine market is growing at an increasing rate every year. In line with this, the demand for an installation vessel with large scaled capacity is also increasing rapidly. The wind turbine installation vessel (WTIV) is a fixed penetration of the spudcan in the sea-bed to install the wind turbine. At this time, a review of the spudcan is an important issue regarding structural safety in the entire structure system. In the study, we analyzed the current procedure suggested by classification of societies and new procedures reflect the new loading scenarios based on reasonable operating conditions; which is also verified through FE-analysis. The current procedure shows that the maximum stress is less than the allowable criteria because it does not consider the effect of the sea-bed slope, the leg bending moment, and the spudcan shape. However, results of some load conditions as defined by the new procedure confirm that it is necessary to reinforce the structure to required levels under actual pre-load conditions. Therefore, the new procedure considers additional actual operating conditions and the possible problems were verified through detailed FE-analysis.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.4
• /
• pp.163-168
• /
• 2022

As cloud computing and the Internet of things are getting popular, the number of devices in the Internet of things computing environments is increasing. In addition, there exist various Internet-based applications, such as home automation and healthcare. In turn, existing studies explored the quality of service, such as downtime and reliability of tasks for Internet of things applications. To enhance the quality of service of Internet of things applications, cloud-fog computing (combining cloud computing and edge computing) can be used for offloading burdens from the central cloud server to edge servers. However, when devices inherit the mobility property, continuity and the quality of service of Internet of things applications can be reduced. In this paper, we propose a resource management scheme based on live migrations for mobility support in edge-based fog computing environments. The proposed resource management algorithm is based on the mobility direction and pace to predict the expected position, and migrates tasks to the target edge server. The performance results show that our proposed resource management algorithm improves the reliability of tasks and reduces downtime of services.

• Journal of the Society of Disaster Information
• /
• v.18 no.2
• /
• pp.374-385
• /
• 2022

Purpose: Even if a damage is applied to the dust of the construction site containing the first-class carcinogen, it is dismissed or 5~30% of the amount of noise damage compensation is paid., Because of such loopholes, some construction companies are neglecting the dust management of the construction site, and the damage of the workers and the residents in the construction site continues. Method: The purpose of this study is to examine the problems of the calculation criteria of damage compensation amount of construction site dust, the measurement of dust concentration, the analysis of measurement data (the data of electric signboard measuring device by the mining scattering method), the prediction and evaluation methods such as modeling, and to suggest improvement measures. Result: It is found that it is impossible to calculate the amount of damages from dust damage in the construction site by calculating the current dust damage compensation amount and dust concentration modeling and measurement. Conclusion: It will receive an application for compensation for damage within the site where damage is expected (about 100m in the straight line and the boundary line of the site), and present a method of calculating the amount of compensation that differentially evaluates dust damage to the degree of dust management and compliance with dust-related legal standards.

• Journal of Korea Entertainment Industry Association
• /
• v.15 no.1
• /
• pp.53-62
• /
• 2021

After the first to present projected moving pictures to audiences, the film industry has been reshaping along with technological advancements. Through the full-scale introduction of visual effects-oriented post-production and digital technologies in the film-making process, the film industry has not only undergone significant changes in the production, but is also embracing the cutting edge technologies broadly and expanding the scope of industry. Not long after the change to digital cinema, the concept of artificial intelligence, first known at the Dartmouth summer research project in 1956, before the digitalization of film, is expected to bring about a big transformation in the film industry once again. Large volume of clear digital data from digital film-making makes easy to apply recent artificial intelligence technologies represented by machine learning and deep learning. The use of artificial intelligence techniques is prominent around major visual effects studios due to automate many laborious, time-consuming tasks currently performed by artists. This study aims to predict how artificial intelligence technology will change the film industry in the future through analysis of visual effects production cases using artificial intelligence technology as a production tool and to discuss the industrial potential of artificial intelligence as visual effects technology.

• Journal of Korea Entertainment Industry Association
• /
• v.15 no.8
• /
• pp.403-409
• /
• 2021

The aim of this study is to investigate the coping behavior according to the personality type of mothers of children with disabilities. The study included 102 mothers of children with disabilities and the study was conducted from june to september in 2020 based on the questionnaire about the personality type and coping behavior. Personality types were evaluated using Enneagram, and coping behavior was evaluated using the Korean Coping Behavior Scale for Parents of Children and Adolescents with Disabilities As a result of the study, the belly type was the most common personality type of mothers with disabilities. As for coping behavior, active problem-solving for children with disabilities showed the highest score. As a result of analyzing the coping behavior according to the personality type of mothers of children with disabilities, there were significant differences in the coping behavior, strengthening marital cooperation, and pursuing social-emotional support. As a result of post-verification, there were significant differences between heart type and belly type in coping behavior, strengthening marital cooperation, and pursuing social-emotional support, and heart type scored higher than belly type. When conducting interventions for families of disabled children, it will be possible to provide more effective services by predicting coping behaviors according to the personality type of mothers of disabled children and presenting individualized programs.

• Trans-
• /
• v.12
• /
• pp.1-19
• /
• 2022

This study predicts and raises the changes that AI will bring to dance art when machine-based choreography began, and finds questions we can ask as human artists. Research suggests that one of the crises of dance in the era of machine creative arts is that artificial intelligence does not stay in the tool of human choreography but becomes the subject of choreography. It is based on the political discourse of choreography that artificial intelligence has the power to control and restrict human dancers. This comes from a sense of crisis that the AI takes over the area of choreography and the human choreographer remains an incompetent coordinator, and as a result, the dancer's dancing body can be reduced to a mechanical body controlled by AI. In order for these concerns not to become a reality, this study proposes three measures. First, choreographer and dancer should develop digital literacy to live in the age of AI art. Secondly, choreographer should acquire the ability to accurately distinguish the roles of human choreographer, dancer, and AI in creative work. Thirdly, various levels of discourse on AI dance should be formed by actively conducting mutual media research of dance and technology. Through these efforts, the human dancer will exist as a subject of art, not a passive agent in the new dance ecosystem brought by the innovation of artificial intelligence technology and will be able to face an era coexistence with artificial intelligence creativily and productively.

• Korean Association of Arts Management
• /
• no.53
• /
• pp.207-247
• /
• 2020

is a project for the senior generation with humanistic knowledge to become a mentor and communicate with them to present the wisdom and direction of life to the new generations of mentees based on various life experiences. has been expanding since 2015, starting with the pilot operation in 2014. In general, projects such as these are assessed to establish effectiveness indicators to verify effectiveness and to establish project management and development strategies. However, most of the evaluations have been conducted quantitatively and qualitatively based on the short-term duration of the project. Therefore, in the case of continuous projects such as , especially in the field of culture and arts where long-term effectiveness verification is required, the short-term evaluation is difficult to predict and judge the actual meaningful effects. In this regard, tried to examine the qualitative change of key participants in this project through the 2017 and 2018 image tracking survey. For this purpose, we adopted qualitative research methodology through interview video shooting, field shooting, and value coding as a research method suitable for the research subject. To analyze the results, first, the interview images were transcribed, keywords were extracted, value encoding works were matched with human psychological values, and the theoretical method was used to identify changes and to derive the meaning. In fact, despite the fact that the study conducted in this study was a follow-up survey, it remained a limitation that it analyzed the changed pattern in a rather short time of 2 years. However, this study systemized the specific methodology that researchers should conduct for follow-up and provided the flow of research at the present time when there is hardly a model for follow-up in the field of culture and arts education business in Korea as well as abroad. Significance can be derived from this point. In addition, it can be said that it has great significance in preparing the detailed system and case of comparative analysis methodology through value coding.

• The Journal of the Korea Contents Association
• /
• v.21 no.11
• /
• pp.135-144
• /
• 2021

Recently, data-driven decision-making technology has become a key technology leading the data industry, and machine learning technology for this requires high-quality training datasets. However, real-world data contains missing values for various reasons, which degrades the performance of prediction models learned from the poor training data. Therefore, in order to build a high-performance model from real-world datasets, many studies on automatically imputing missing values in initial training data have been actively conducted. Many of conventional machine learning-based imputation techniques for handling missing data involve very time-consuming and cumbersome work because they are applied only to numeric type of columns or create individual predictive models for each columns. Therefore, this paper proposes a new data imputation technique called 'Denoising Self-Attention Network (DSAN)', which can be applied to mixed-type dataset containing both numerical and categorical columns. DSAN can learn robust feature expression vectors by combining self-attention and denoising techniques, and can automatically interpolate multiple missing variables in parallel through multi-task learning. To verify the validity of the proposed technique, data imputation experiments has been performed after arbitrarily generating missing values for several mixed-type training data. Then we show the validity of the proposed technique by comparing the performance of the binary classification models trained on imputed data together with the errors between the original and imputed values.