• Convergence Security Journal
• /
• v.22 no.3
• /
• pp.25-32
• /
• 2022

Recently, the image processing industry has been activated as deep learning-based technology is introduced in the image recognition and detection field. With the development of deep learning technology, learning model vulnerabilities for adversarial attacks continue to be reported. However, studies on countermeasures against poisoning attacks that inject malicious data during learning are insufficient. The conventional countermeasure against poisoning attacks has a limitation in that it is necessary to perform a separate detection and removal operation by examining the training data each time. Therefore, in this paper, we propose a technique for reducing the attack success rate by applying modifications to the training data and inference data without a separate detection and removal process for the poison data. The One-shot kill poison attack, a clean label poison attack proposed in previous studies, was used as an attack model. The attack performance was confirmed by dividing it into a general attacker and an intelligent attacker according to the attacker's attack strategy. According to the experimental results, when the proposed defense mechanism is applied, the attack success rate can be reduced by up to 65% compared to the conventional method.

• The Journal of Bigdata
• /
• v.7 no.2
• /
• pp.153-172
• /
• 2022

Online shopping has transformed and rapidly grown the entire market at the forefront of wholesale and retail services as an effective solution to issues such as digital transformation and social distancing policy (COVID-19 pandemic). Small business owners, who form the majority at the center of the online shopping industry, are constantly collecting policy changes and market trend information to overcome these problems and use them for marketing and other sales activities in order to overcome these problems and continue to grow. Objective and refined information that is more closely related to the business is also needed. Therefore, in this paper, through the collection and analysis of big data information, which is the core technology of digital transformation, key variables are set in product classification, sales trends, consumer preferences, and review information of online shopping malls, and a method of using them for competitor comparison analysis and business sustainability evaluation has been prepared and we would like to propose it as a service. If small and medium-sized businesses can benchmark competitors or excellent businesses based on big data and identify market trends and consumer tendencies, they will clearly recognize their level and position in business and voluntarily strive to secure higher competitiveness. In addition, if the sustainable growth of the online shopping mall operator can be confirmed as an indicator, more efficient policy establishment and risk management can be expected because it has an improved measurement method.

• The Journal of Bigdata
• /
• v.7 no.2
• /
• pp.31-41
• /
• 2022

Steel is one of the most fundamental components to mechanical industry. However, the quality of products are greatly impacted by the surface defects in the steel. Thus, researchers pay attention to the need for surface defects detector and the deep learning methods are the current trend of object detector. There are still limitations and rooms for improvements, for example, related works focus on developing the models but don't take into account real-time application with practical implication on industrial settings. In this paper, a real-time application of steel surface defects detection based on YOLOv4 is proposed. Firstly, as the aim of this work to deploying model on real-time application, we studied related works on this field, particularly focusing on one-stage detector and YOLO algorithm, which is one of the most famous algorithm for real-time object detectors. Secondly, using pre-trained Yolov4-Darknet platform models and transfer learning, we trained and test on the hot rolled steel defects open-source dataset NEU-DET. In our study, we applied our application with 4 types of typical defects of a steel surface, namely patches, pitted surface, inclusion and scratches. Thirdly, we evaluated YOLOv4 trained model real-time performance to deploying our system with accuracy of 87.1 % mAP@0.5 and over 60 fps with GPU processing.

• The Journal of Bigdata
• /
• v.8 no.1
• /
• pp.49-59
• /
• 2023

There are various items in the safety and health standards of the manufacturing industry, but they can be divided into work-related diseases and musculoskeletal diseases according to the standards for sickness and accident victims. Musculoskeletal diseases occur frequently in manufacturing and can lead to a decrease in labor productivity and a weakening of competitiveness in manufacturing. In this paper, to detect the musculoskeletal harmful factors of manufacturing workers, we defined the musculoskeletal load work factor analysis, harmful load working postures, and key points matching, and constructed data for Artificial Intelligence(AI) learning. To check the effectiveness of the suggested dataset, AI algorithms such as YOLO, Lite-HRNet, and EfficientNet were used to train and verify. Our experimental results the human detection accuracy is 99%, the key points matching accuracy of the detected person is @AP0.5 88%, and the accuracy of working postures evaluation by integrating the inferred matching positions is LEGS 72.2%, NECT 85.7%, TRUNK 81.9%, UPPERARM 79.8%, and LOWERARM 92.7%, and considered the necessity for research that can prevent deep learning-based musculoskeletal diseases.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.1D
• /
• pp.157-167
• /
• 2006

During the first and second NGIS projects by the Korean government, The first one (1995~2000) was limited on constructing geographic information and the second (2001~2005) was focused on circulation and practical use of geoinformation from the result of the first project. In the latter half of 2nd NGIS project, However, the geographic information from the NGIS projects have not been renewed even though there were significant geographical changes. The accurate renewal of geoinformation is a matter of great importance to the next generation industry (e.g. LBS, Ubiquitous, Telematics). In this respect, it is time to update the geographic information in the latter half of the second NGIS project. Therefore, It is not only important to build an accurate geoinformation but also rapid and correct renewal of the geoinformation. NGII (National Geographic Information Institute) has been studying for improvement of digital map that was constructed by the result of the 1st NGIS project. Through the construction of clean digital map, NGII constructed Framework Data to three kinds of formats (NGI, NDA, NRL). Framework Data was contained to other database, and provided the reference system of location or contents for combining geoinformation. Framework Data is consist of Data Set, Data Model and UFID (Unique Feature Identifier). It will be achieved as national infrastructure data. This paper attempts to explore a method of the update to practical framework data with realtime geoinformation on feature's creation, modification and destruction managed by 'Feature management agency' using UFID's process. Furthermore, it suggests a method which can provide important data in order to plan the Framework update with the land change ratio.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.276-279
• /
• 2022

The 4th industrial revolution refers to the transition to a knowledge society in which the production speed of knowledge is fast and the proportion of the knowledge industry is greatly increased. The reorganization of the industrial structure and the change of occupations and jobs due to new technologies are bringing about changes in education, and the development of digital technology has made education that is borderless, individual and dynamic, becoming the new standard of education. With these changes, interest in nano-degrees on new technologies or micro-degrees focused on core courses rather than regular course degrees is increasing. As a representative example, Udacity in the United States has opened and operated online nanodegree courses related to jobs, and collaborates with major companies to develop and educate core training courses necessary for companies, thereby efficiently supporting companies in securing talent. With the revitalization of online vocational and job training, an environment has been prepared in which individuals can set goals for vocational competency development and continue their portfolio-based sustainable learning. However, for effective vocational education, automated and personalized educational content design should be preceded. To this end, in this paper, we propose a personalized career and course map recommendation system in the era of online learning

• International Area Studies Review
• /
• v.15 no.1
• /
• pp.447-466
• /
• 2011

This paper examined the supply chain management Integration factors and environmental performance in Korean industry. Focusing on SCM Integration, GreenSCM practice, and environmental performance factors, purpose of this study is finding linkage between SCM Integration factors with new environmental practice factors and environmental performance. Based on the analysis of eighty-eight cases, the following results were found. First, We found External environmental collaboration factors and Internal environmental monitoring factors within Green Supply Chain Practices. Second, SCM Integration have a positively significant influence on environmental performance. Third, Internal environmental monitoring factors have a positively significant influence on Environmental performance but External environmental collaboration factors doesn't. This study suggests that only with high level of Integration firms may have good result on entire supply chain environmental performance. Finally, our empirical evidence shows that company should be prepared for new environmental trade regulation with Green Supply chain management integration.

• 지역과문화
• /
• v.6 no.4
• /
• pp.47-72
• /
• 2019

The purpose of this paper is to explore what roles can and should regional culture play in establishing a regional innovation system, which has been recently described as an essential factor for balanced regional development and specialized regional development. In doing so, we first investigated key concepts and discourses regarding regional innovation system and examined the success factors of a regional innovation system. We also explored how the Korean policy makers have developed the nation's regional innovation policy and regional cultural policy in order to understand the policy context and its limitations. After examining the theoretical background, we reviewed the representative cases of regional innovation in Japan, Yufuin and Yubari, to grasp the way in which 'culture' plays its roles in the formation of the regional innovation system. Since culture has a role as a unique sector, such as art and content industry, as well as a community's style of life, this study tried to explore the aspects and ways of how culture can contribute to the establishment of regional innovation systems considering these distinctive levels. In drawing the implications of the case study, we found that special attention is required for the 'process of formation' of regional innovation systems. We also confirmed that the success of the core activities at each stage heavily relies on the culture of the region; in order to fully understand the relations, it is necessary to re-establish the existing concept of culture and cultural policy with a more holistic perspective.

• Asia Marketing Journal
• /
• v.9 no.2
• /
• pp.97-129
• /
• 2007

The critical link between consumer-based internet ordering and the delivery of the product to the consumer is a key success factor in online shopping. Market areas of online shopping company implies the range of space where products ordered from online shopping can be physically delivered to customers distributed over space with reasonable shipping cost and lead time through the physical distribution network. The average rate of growth in online shopping is 36% per year in Korea for the last 5 years. But there are no maps available that describe sales/delivery density of online shopping, few researches are focused on the short-term/long-term adaptation to demand increase by online shopping. In this paper (1) Maps of trade area are described indicating the sales/delivery density around the nation. (2) Empirical researches suggested that short-term adaptation to demand increase resulted in price reduction and service in enhancement of service quality in local transportation. But the long-term adaptation on the parts of parcel delivery industry are to be investigated in future researches.

• Journal of the Microelectronics and Packaging Society
• /
• v.30 no.3
• /
• pp.56-63
• /
• 2023

In this paper, we studied the effect of electron beam irradiation on sol-gel indium-gallium-zinc oxide (IGZO) thin films under air and nitrogen atmosphere and carried out the electrical characterization of the s ol-gel IGZO thin film transistors (TFTs). To investigate the optical properties, crystalline structure and chemical state of the sol-gel IGZO thin films after electron beam irradiation, UV-Visible spectroscopy, X-ray diffraction (XRD), and X-ray photoelectron spectroscopy (XPS) were carried out. The sol-gel IGZO thin films exhibited over 80% transmittance in the visible range. The XRD analysis confirmed the amorphous nature of the sol-gel IGZO films regardless of electron beam irradiation. When electron beam irradiation was conducted in a nitrogen (N₂) atmosphere, we observed an increased proportion of peaks related to M-O bonding contributed to the improved quality of the thin films. Sol-gel IGZO TFTs subjected to electron beam exposure in a nitrogen atmosphere exhibited enhanced electrical characteristics in terms of on/off ratio and electron mobility. In addition, the electrical parameters of the transistor (on/off ratio, threshold voltage, electron mobility, subthreshold swing) remained relatively stable over time, indicating that the electron beam exposure process in a nitrogen atmosphere could enhance the reliability of IGZO-based thin-film transistors in the fabrication of sol-gel processed TFTs.