• 한국실내디자인학회논문집
• /
• 제26권1호
• /
• pp.63-74
• /
• 2017

This study is for the characteristic of expression of exhibition presentation in 'The Korean peninsula' which is the Golden Lion Winner at the Venice Biennale in 2014. The Korean peninsula provides two opposite political and economic systems and adaptations of modernism through various multi media images. The curator, Minsuk Cho, presented cinema montage image of Collision for analyzing dynamic exhibition organization and provides the foundation of his theory from Bergson's image and duration to Deleuze's movement-image and time-image which is mentioned from Deleuze's book "cinema 1" and "cinema 2". Furthermore, the Korean peninsula has showed perception-image and affection-image from Eisenstein & Vertov's cinemathology and systemized exhibition presentation. The montage of Collision has maximized the movement from the variety and complexity in a collision and it made difference between information images which are space, time, emotion, intelligence, the tumult between subjectivity and objectivity, fragments from reorganizing itself, and distribution and art images. The limit of montage of Collision's dialectic is not only visual but also space from organic organization but the Korean peninsula overcomes its limit and shows leap of tactile perception and time-reflection from the montage of Collision's dialectic. Therefore, the exhibition of the Korean peninsula presents the conviction of adaptations of modernism.

• 한국항행학회논문지
• /
• 제22권6호
• /
• pp.670-674
• /
• 2018

최근 댓글 / 텍스트, 이미지, 비디오, 블로그 및 사용자 경험을 포함한 소셜네트워크서비스(SNS) 데이터에는 다양한 고객의 추천 시스템을 구축하고 비즈니스 분석가에게 통찰력 있는 데이터 / 결과를 제공하는데 사용할 수 있는 많은 정보가 포함되어 있다. 멀티미디어 데이터, 특히 이미지 및 비디오와 같은 시각적 데이터는 SNS 데이터 중에서도 특정(문화권) 지역을 반영할 수 있는 가장 풍부한 데이터이며, 문화적 가치 및 관심사는 전반적으로 데이터의 많은 부분을 차지하고 있다. 이러한 방대한 데이터로부터 원하는 데이터를 지능적으로 추출하고, 엄청난 양의 데이터를 마이닝 하려면 보다 효율적이고 지능적인 데이터 분석 방법이 필요하다. 따라서 본 논문의 목적은 이러한 데이터를 모델링하고, 색인하고, 검색하는 방법에 대해 제안하고자 한다.

• 패션비즈니스
• /
• 제22권6호
• /
• pp.39-52
• /
• 2018

The objective of this study was to enhance understanding and appreciation of Pre-Orientalism in costumes and textiles by revealing examples of Oriental influences in Europe from the 16th century to the mid-18th century through in-depth study. The research method used were the presentation and analysis of previous literature research and visual data. The result were as follows; Pre-Orientalism had been influenced by Morocco, Thailand, and Persia as well as Turkey, India, and China. In this study, Pre-Orientalism refers to oriental influence and oriental taste in Western Europe through cultural exchanges from the 16th century to the mid-18th century. The oriental costume was the most popular subspecies of fancy, luxury dress and was a way to show off wealth and intelligence. Textiles were used for decoration and luxury. The Embassy and the court in Versailles and Vienna led to a frenzy of oriental fashion. It appeared that European in the royal family and aristocracy of Europe had been accommodated without an accurate understanding of the Orient. Although in this study, the characteristics, factors, and impacts of Pre-Orientalism have not been clarified, further study can be done. Recognizing a broad perspective on oriental influence in Europe before Orientalism, we can have a balanced view of future Orientalism and global fashion.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권3호
• /
• pp.1689-1701
• /
• 2019

In recent years, personal videos have been shared online due to the popular uses of portable devices, such as smartphones and action cameras. A recent report predicted that 80% of the Internet traffic will be video content by the year 2021. Several studies have been conducted on the detection of main video events to manage a large scale of videos. These studies show fairly good performance in certain genres. However, the methods used in previous studies have difficulty in detecting events of personal video. This is because the characteristics and genres of personal videos vary widely. In a research, we found that adding a dataset with the right perspective in the study improved performance. It has also been shown that performance improves depending on how you extract keyframes from the video. we selected frame segments that can represent video considering the characteristics of this personal video. In each frame segment, object, location, food and audio features were extracted, and representative vectors were generated through a CNN-based recurrent model and a fusion module. The proposed method showed mAP 78.4% performance through experiments using LSVC data.

• 한국표면공학회지
• /
• 제55권6호
• /
• pp.425-432
• /
• 2022

Antifouling paints that inhibit the attachment and contamination of marine organisms mainly use TBT compounds, but because of their toxic components, they cause ecosystem disturbance and environmental destruction problems, so It is necessary to research eco-friendly antifouling paints that are easy to maintain and effective antifouling technologies. In this study, physical surface treatment of silane coating and chemical antifouling technology were applied to the metal surface to secure the stability of the surface of the marine structure and inhibit the attachment and growth of marine organisms. Adhesion of marine organisms was evaluated according to the coating conditions through surface evaluation of the charged material for 15 months in the waters of the west coast of Korea. In accordance with ASTM D6990-05, antifouling properties fouling rates (FR) and physical degradation rates(PDR) were evaluated through visual inspection of the evaluation specimens. As a result of evaluating the antifouling performance of the coated surface, it was confirmed that the antifouling performance was maintained at the 50% level even after 15 months in the sample subjected to physical processing and silane coating.

• 국제초고층학회논문집
• /
• 제9권4호
• /
• pp.351-360
• /
• 2020

Monitoring civil structures periodically is necessary for ensuring the fitness of the structures. Cracks on inner and outer surfaces of the building plays a vital role in indicating the health of the building. Conventionally, human visual inspection techniques were carried up to human reachable altitudes. Monitoring of high rise infrastructures cannot be done using this primitive method. Also, there is a necessity for more accurate prediction of cracks on building surfaces for ensuring the health and safety of the building. The proposed research focused on developing an efficient crack classification model using Transfer Learning enabled EfficientNet (TL-EN) architecture. Though many other pre-trained models were available for crack classification, they rely on more number of training parameters for better accuracy. The TL-EN model attained an accuracy of 0.99 with less number of parameters on large dataset. A bench marked METU dataset with 40000 images were used to test and validate the proposed model. The surfaces of high rise buildings were investigated using vision enabled Unmanned Arial Vehicles (UAV). These UAV is fabricated with TL-EN model schema for capturing and analyzing the real time streaming video of building surfaces.

• KEPCO Journal on Electric Power and Energy
• /
• 제7권2호
• /
• pp.323-328
• /
• 2021

Diagnostic models are required. Data augmentation is one of the best ways to improve deep learning performance. Traditional augmentation techniques that modify image brightness or spatial information are difficult to achieve great results. To overcome this, a generative adversarial network (GAN) technology that generates virtual data to increase deep learning performance has emerged. GAN can create realistic-looking fake images by competitive learning two networks, a generator that creates fakes and a discriminator that determines whether images are real or fake made by the generator. GAN is being used in computer vision, IT solutions, and medical imaging fields. It is essential to secure additional learning data to advance deep learning-based fault diagnosis solutions in the power industry where facilities are strictly maintained more than other industries. In this paper, we propose a method for generating power facility images using GAN and a strategy for improving performance when only used a small amount of data. Finally, we analyze the performance of the augmented image to see if it could be utilized for the deep learning-based diagnosis system or not.

• Journal of Medicine and Life Science
• /
• 제19권1호
• /
• pp.1-7
• /
• 2022

Learning disabilities (LD), also known as learning disorders, refers to cases in which an individual experiences lower academic ability as compared to the normal range of intelligence, visual or hearing impairment, or an inability to peform learning. Children and adolescents with learning disabilities often have emotional or behavioral problems or co-existing conditions, including depression, anxiety disorders, difficulties with peer relationships, family conflicts, and low self-esteem. In most cases, attention deficit and hyperactivity disorder coexists. As learning disabilities have the characteristics of a difficult heterogeneous disease group that cannot be attributed to a single root cause, they are diagnosed based on an interdisciplinary approach through medicine and education, such as mental health medicine, education, psychology, special education, and neurology. In addition, for the accurate diagnosis and treatment of learning disabilities, the diagnosis, prescription, treatment, and educational intervention should be conducted in cooperation with doctors, teachers, and psychologists. The treatment of learning disabilities requires a multimodal approach, including medical and educational intervention. It is suggested that educational interventions such as the Individualized Education Plan (IEP) and the Response to Invention (RTI) should be implemented.

• 대한원격탐사학회지
• /
• 제39권4호
• /
• pp.459-466
• /
• 2023

In the last decade, artificial intelligence's dramatic advancement with the development of various deep learning techniques has significantly contributed to remote sensing fields and satellite image applications. Among many prominent areas, super-resolution research has seen substantial growth with the release of several benchmark datasets and the rise of generative adversarial network-based studies. However, most previously published remote sensing benchmark datasets represent spatial resolution within approximately 10 meters, imposing limitations when directly applying for super-resolution of small objects with cm unit spatial resolution. Furthermore, if the dataset lacks a global spatial distribution and is specialized in particular land covers, the consequent lack of feature diversity can directly impact the quantitative performance and prevent the formation of robust foundation models. To overcome these issues, this paper proposes a method to generate benchmark datasets by simulating the modulation transfer functions of the sensor. The proposed approach leverages the simulation method with a solid theoretical foundation, notably recognized in image fusion. Additionally, the generated benchmark dataset is applied to state-of-the-art super-resolution base models for quantitative and visual analysis and discusses the shortcomings of the existing datasets. Through these efforts, we anticipate that the proposed benchmark dataset will facilitate various super-resolution research shortly in Korea.

• 한국건설관리학회논문집
• /
• 제24권4호
• /
• pp.60-73
• /
• 2023

The need for adopting automation technologies to improve inefficiencies in interior finishing modeling work is increasing during the Building Information Modeling (BIM) design stage. As a result, the use of visual programming languages (VPL) for practical applications is growing. However, undefined or incorrect space designations in BIM models can hinder the development of automated finishing modeling processes, resulting in erroneous corrections and rework. To address this challenge, this study first developed a rule-based automated interior finishing detailing module for floors, walls, and ceilings. In addition, an automated space integrity checking module with 86.69% ACC using the Multi-Layer Perceptron (MLP) model was developed. These modules were integrated into a design automation module for interior finishing, which was then verified for practical utility. The results showed that the automation module reduced the time required for modeling and integrity checking by 97.6% compared to manual work, confirming its utility in assisting BIM model development for interior finishing works.