• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.407-409
• /
• 2021

Under the influence of the 4th industrial revolution, various technologies such as artificial intelligence and automation are being grafted into industrial sites, and accordingly, the importance of data processing is increasing. Digital images may generate noise due to various reasons, and may affect various systems such as image recognition and classification and object tracking. To compensate for these shortcomings, we propose an image restoration algorithm based on pattern information of non-noise pixels. According to the distribution of non-noise pixels inside the filtering mask, the proposed algorithm switched the filtering process by dividing the interpolation method into a pattern that can be applied, a pattern based on region division, and a randomly arranged pixel pattern. preserves and restores the image. The proposed algorithm showed superior performance compared to the existing impulse noise removal algorithm.

• Journal of Fashion Business
• /
• v.27 no.5
• /
• pp.60-77
• /
• 2023

The scale of the industry utilizing the Metaverse platform is gradually growing around the world. Fashion brands are also starting to utilize the Metaverse platform as a new market to replace the next e-commerce platform by targeting new consumers, MZ generation, and even Alpha generation. In this study, a real K-fashion designer's outfit was made into a 3D outfit using CLO 3D program to express it in a new market, the Metaverse 3D platform. It was then compared with a real outfit. An avatar prototype was completed using Max program to simulate the 3D digital fashion outfit and produce an avatar through an optimization process. The 3D outfits showed the same level of results as the actual outfits in terms of fabric surface, material texture, drapability, overall outfit, details, and trimmings. In addition, we proposed a 2D work on total styling suggestion and modeling to secure data sets for future AI-based styling services. In conclusion, this study revealed that actual outfits and 3D outfits had the same results. It is significant that it can be a sample work to build a styling data set through styling suggestion and content production as a significant amount of styling DB construction will be required before AI styling automation services.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.11
• /
• pp.353-369
• /
• 2024

This paper presents strategies for enhancing productivity and strengthening global competitiveness as the domestic shipbuilding industry transitions into the era of Artificial Intelligence Transformation (AX), moving beyond digital transformation. Historically a labor-intensive industry, shipbuilding has evolved into smart shipyards powered by automation and digitalization, with increasing emphasis on green regulations and the importance of green fuels. The urgent adoption of alternative fuels, such as ammonia and liquid hydrogen, is critical in this context. However, the industry faces new challenges amid intensifying global competition and rapid technological changes. This study analyzes both domestic and international cases of AI transformation and the adoption of eco-friendly fuels in shipbuilding companies, proposing ways to manage risks through the establishment of AI governance to ensure sustainable growth. In particular, by utilizing the backcasting method, the study sets short-term, mid-term, and long-term goals while deriving phased strategies to provide significant insights and implications for policy formulation and corporate strategies aimed at the AI transformation of the domestic shipbuilding industry while complying with environmental regulations.

• Korean Journal of Remote Sensing
• /
• v.40 no.5_1
• /
• pp.465-478
• /
• 2024

Land use constantly changes due to climate change and human activities. Above all else, monitoring forests is crucial for global carbon management, particularly in the context of climate change. Land use changes in forested areas occur due to various factors such as development projects or natural disasters; forest fires are one of the primary drivers of large-scale forest loss. Therefore, it is essential to detect forest changes including forest fires accurately and to develop an automated system for periodic monitoring. In response, this study proposes a machine learning-based method for automating forest change detection using multi-temporal medium-resolution satellite imagery. As a case study area, the Dogye-eup, Samcheok was selected which experienced rapid forest change following significant forest fires in 2017. To construct spatial datasets for the factors influencing forest changes, key spectral bands were extracted after preprocessing the satellite images, and indices such as the normalized difference vegetation index (NDVI) and normalized difference water index (NDWI) were computed. Additionally, slope values were derived from digital elevation model (DEM) data to further enhance the dataset. Using the training set based on NDVI derived from a forest map and single-season imagery, a forest probability map was generated through a machine-learning model based on artificial neural network (ANN). The final estimate of forest reduction was determined by analyzing seasonal imagery differentials and their summation. This automated approach to extracting training data from satellite imagery and pre-existing datasets offers significant potential to enhance the automation of forest monitoring.

• Journal of Digital Convergence
• /
• v.15 no.11
• /
• pp.121-131
• /
• 2017

The Fuzzy-analytic hierarchy process (AHP) was adopted as the methodology for this study because it allows for the use qualitative judgments by experts. Based on results of the analysis of the success factors for frozen storage/cold storage warehouses, the facility factor was identified as the most important to consider. This factor had a weight of 0.307, followed by systems and operations, accessibility, and standardization/automation with weights of 0.263, 0.255, and 0.175, respectively. The conclusions and implications of the study are as follows. First, the efficiency of constant temperature and humidity systems and the heat insulation property of buildings need to be enhanced. Second, the efficiency of the operations should be enhanced through the standardization of equipment rather than by standardizing product loading. Finally, since logistics and transportation costs are higher for frozen storage/cold storage warehouses than for general distribution, accessibility needs to be considered as the first priority.

• Korean Journal of Construction Engineering and Management
• /
• v.11 no.2
• /
• pp.106-115
• /
• 2010

Recently, as the size of construction projects are getting bigger and higher, more effective managing methods are required in management areas such as duration reduction, cost reduction and quality management. In the current construction industry, conjunction with IT(Information Technology) is being noticed as a solution to support these needs. Various IT solutions such as Bar Code, Personal digital assistant(PDA), global positioning system(GPS), radio frequency identification(RFID) are being developed. In this research, among the various IT solutions, the Real Time Locating System(RTLS) which is acknowledged as a technology with high applicational potential is analyzed. Based on this analysis, a locating system to apply in construction sites is developed and validated. The locating system is developed to prevent construction disasters through real-time management of workers and equipment, which enables effective application in the area of construction safety management. Moreover, applications of the locating system in many different areas like construction material realtime monitoring, construction automation, construction quality management, maintenance management are expected.

• Journal of Digital Contents Society
• /
• v.9 no.4
• /
• pp.661-667
• /
• 2008

The dynamics of business cycles has been changed by the macroscopic economic forces because of the introduction of new technical know-how each year. These the dynamics of business has a significant influence on the investment of enterprise in the information communication field. Today, the most important goal of the IT investment is simply not to lower the production cost any more, but to improve the usefulness for the customers and partners in order to obtain the optimized mass products. Therefore, the enterprises have been concentrating their all abilities on the automation, integration, and optimization of business process using BPM. In addition, they are concentrating their efforts on the business expansion by approaching the technical aspect using RFID application system. However, in order to accomplish a successful enterprise ability, the technical view, business process view, and organization view must be considered together. We suggested the method considering organization view, via the technical element, i.e., RFID system for approaching the business process. Furthermore, we tried the optimization of assignment using Context Analysis methodology and proposed the method to reduce the element with respect to the time, human, and expense by applying the Case Study method that minimizes the iteration times through the transmitted processing procedure and type. The proposed method gave us the expectation that it will bring out the innovative improvement with respect to the time, expense, quality, and customer's satisfaction in the process from the analysis of business process to the analysis and design of system.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.3
• /
• pp.187-201
• /
• 2021

Recently, the demand for a 3D urban spatial information infrastructure for storing, operating, and analyzing a large number of digital data produced in cities is increasing. CityGML is a 3D spatial information data standard of OGC (Open Geospatial Consortium), which has strengths in the exchange and attribute expression of city data. Cases of constructing 3D urban spatial data in CityGML format has emerged on several cities such as Singapore and New York. However, the current ecosystem for the creation and editing of CityGML data is limited in constructing CityGML data on a large scale because of lack of completeness compared to commercial programs used to construct 3D data such as sketchup or 3d max. Therefore, in this study, a method of constructing CityGML data is proposed using commercial 3D mesh data and 2D polygons that are rapidly and automatically produced through aerial LiDAR (Light Detection and Ranging) or RGB (Red Green Blue) cameras. During the data construction process, the original 3D mesh data was geometrically transformed so that each object could be expressed in various CityGML LoD (Levels of Detail), and attribute information extracted from the 2D spatial information data was used as a supplement to increase the utilization as spatial information. The 3D city features produced in this study are CityGML building, bridge, cityFurniture, road, and tunnel. Data conversion for each feature and property construction method were presented, and visualization and validation were conducted.

• Journal of the Korean Applied Science and Technology
• /
• v.38 no.1
• /
• pp.241-262
• /
• 2021

This paper reviewed the necessity of a information and communication technology (ICT)-based smart livestock system as a development strategy for the animal life industry in the future. It also predicted the trends of livestock and animal food until 2050, 30 years later. Worldwide, livestock raising and consumption of animal food are rapidly changing in response to population growth, aging, reduction of agriculture population, urbanization, and income growth. Climate change can change the environment and livestock's productivity and reproductive efficiencies. Livestock production can lead to increased greenhouse gas emissions, land degradation, water pollution, animal welfare, and human health problems. To solve these issues, there is a need for a preemptive future response strategy to respond to climate change, improve productivity, animal welfare, and nutritional quality of animal foods, and prevent animal diseases using ICT-based smart livestock system fused with the 4th industrial revolution in various aspects of the animal life industry. The animal life industry of the future needs to integrate automation to improve sustainability and production efficiency. In the digital age, intelligent precision animal feeding with IoT (internet of things) and big data, ICT-based smart livestock system can collect, process, and analyze data from various sources in the animal life industry. It is composed of a digital system that can precisely remote control environmental parameters inside and outside the animal husbandry. The ICT-based smart livestock system can also be used for monitoring animal behavior and welfare, and feeding management of livestock using sensing technology for remote control through the Internet and mobile phones. It can be helpful in the collection, storage, retrieval, and dissemination of a wide range of information that farmers need. It can provide new information services to farmers.

• Journal of the Korean Geotechnical Society
• /
• v.37 no.11
• /
• pp.119-129
• /
• 2021

The process of constructing the DB in the current geotechnical information DB system needs a lot of human and time resource consumption. In addition, it causes accuracy problems frequently because the current input method is a person viewing the PDF and directly inputting the results. Therefore, this study proposes building an automatic digital DB using AI (artificial intelligence) of boring logs. In order to automatically construct DB for various boring log formats without exception, the boring log forms were classified using the deep learning model ResNet 34 for a total of 6 boring log forms. As a result, the overall accuracy was 99.7, and the ROC_AUC score was 1.0, which separated the boring log forms with very high performance. After that, the text in the PDF is automatically read using the robotic processing automation technique fine-tuned for each form. Furthermore, the general information, strata information, and standard penetration test information were extracted, separated, and saved in the same format provided by the geotechnical information DB system. Finally, the information in the boring log was automatically converted into a DB at a speed of 140 pages per second.