Although the proportion of coal-fired power generation is decreasing, efficient operating technology is needed to continuously invest in facilities and reduce maintenance costs until it is abolished. Boilers, one of the main facilities of power plants, operate for a long time in harsh environments of high temperature and high pressure. In addition, damage due to deterioration is likely to occur depending on the fuel and tube material used. It is very important to judge soundness because damage caused by deterioration adversely affects facility operation. Previously, replication method was used to analyze the progress of deterioration. In the replication method, pre-treatment such as chemical treatment is performed on the boiler tube in the field, the area is reproduced by attaching a film, and the replicated film is determined by an expert in the laboratory with an expensive microscope. However, this method involves substantial costs and time requirements, as well as the possibility of human errors. To address these issues, we developed a mobile health assessment system in this research. Since it is detachable and takes images in real time, this system enables swift evaluations across a broad range and facilitates the assessment of preprocessing quality. In addition, it was intended to reduce existing human mistakes by developing a degradation classification algorithm using the merger cluster method.

With the recent advent of Metaverse, the character industry that reflects the characteristics of users' faces is drawing attention. there is a hassle that users have to select face components such as eyes, nose, and mouth one by one. In this paper, we propose a diffusion-based model that automatically generates characters from content human photographs. Our model generates user artistic characters by reflecting content information such as face angle, direction, and shape of a content human photo. In particular, our model automatically analyzes detailed information such as glasses and whiskers from content photo images and reflects them in artistic characters generated. Our network generates the final character through a three-step: diffusion process, UNet, and denoising processes. We use image encoders and CLIP encoders for the connection between style and input data. In the diffusion process, a collection of noise vectors is gradually added to a style vector to enable lossless learning of the detailed styles. All input values except for the style images are vectorized with CLIP encoders and then learned with noise style vectors in the UNet. Subsequently, noise is removed from the vectors through the UNet to obtain the artistic character image. We demonstrate our performance by comparing the results of other models with our results. Our method reflects content information without loss and generates natural high-definition characters.

There are many types of foam molding methods. The most commonly used methods are the pressure foaming method, in which foam resin is mixed with a foaming agent at high temperature and high pressure, and the normal pressure foaming method, which foams at high temperature without pressure. The polymer resins used for foaming have different viscosities. For foaming under normal pressure, they need to be designed and analyzed for optimal foaming conditions, to obtain resins with low melt-viscosity or a narrow optimal viscosity range. This study investigated how changes in viscosity, molding temperature, and cross-link foaming conditions affected the characteristics of the molded foam, prepared by blending rubber polymer with biodegradable resin. The morphologies of cross sections and the cell structures of the normal pressure foam were investigated by SEM analysis. Properties were also studied according to cross-link/foaming conditions and torque. Also, the correlation between foaming characteristics was studied by analyzing tensile strength and elongation, which are mechanical properties of foaming composites.

In the hospitality and tourism sectors, service robots have become increasingly adopted by companies to facilitate frontline service to reduce human labor, improve efficiency and provide better customer experience. In addition, the COVID-19 pandemic has intensified consumers’ demand for service robots to reduce human contact. To better understand the impact of service robots in the service sector, researchers have been examining user behavior of different technologies. While prior empirical studies have mainly focused on utilitarian-based acceptance models, other perspectives, such as the interpersonal relations, have been largely overlooked. With the ubiquitous of social media, people become connected but alone, leading to negative effects on interpersonal relations.

Instagram marketing is a popular tool for communication in the tourism industry. However, research into effective execution tactics has been limited to the content, recipient characteristics, and sender types (e.g., celebrities) (i.e., Ferwerda & Tkalcic, 2018; Jin et al., 2019; Singh, 2020). The layouts of Instagram ads themselves have received little scholarly attention. Instagram ads typically convey multiple pieces of information in a single advertisement unit, raising concerns about the significance of the order in which the information is presented. Specifically, we must understand how different orders of presenting information affect a firm’s performance. Therefore, this study investigates the effect of presentation order on consumer behavioral intentions toward tour programs promoted on Instagram.

Corporate social responsibility (CSR) has become an important pillar in business practice. Luxury brands are no exception to this trend and have invested deeply in different types of CSR, and this is evident with companies such as Prada and LVMH. However, empirical research has demonstrated the paradox between CSR and luxury and has identified certain undesirable effects. The reason for this inconsistency in the results to date, is the lack of a comprehensive CSR categorization system. To be more specific, we integrated two dimensions - visibility and relevance - and have proposed a novel CSR type in four types, which is external-embedded, external-peripheral, internal-embedded and internal-peripheral. We can easily classify the CSR initiatives into this category, which means the source of undesirable effects can be clearly shown for luxury brands.

부추속은 수선화과에 속하며 약 1000여종에 이른다. 좀부 추는 자생 부추속에 속하며 식용이 가능하고 크기가 작아 분 화식물로 적합하다. 이에 따라 본 연구는 좀부추를 분화 소재 로서 개발하기 위하여 식물의 측지성장을 촉진하고 꽃의 품질 을 향상시키기 위해 수행되었다. 이를 위하여 좀부추의 구근 을 생장조절제(GA3, BA, 에테폰, TDZ)에 다른 시간(1, 5시 간)동안 침지처리했을 때 식물의 생장과 꽃에 어떤 영향을 주 는지 조사하였다. GA3 처리구는 엽수와 자구의 수가 증가해 증식을 촉진시키는 효과가 있었다. 게다가 GA3는 화수의 증 가와 조기개화를 유도했다. 그러나 엽폭과 화경의 폭은 감소 하였다. 에테폰은 좀부추의 초장과 초폭에서 영향을 주지 않 았으나 1h에서 자구의 수가 증가하였다. 에테폰은 화서 발생 을 억제하여 화경의 수가 감소하였다. BA는 식물의 생장과 개 화 특성에 유의적인 영향을 보이지 않았다. TDZ처리한 식물 은 왜화되고 생육상태가 불량하였으며, 식물을 고사시켰다.

본 논문에서는 웹툰 캐릭터 영상에 대해서 심층학습에 기반한 3D 안면 재구성 기술을 제안한다. 본 연구에서 제안하는 방법은 기본 사항 모듈과 상세 사항 모듈로 구성된다. 입력 받은 웹툰 캐릭터 영상에 대해서 기본 사항 모듈의 요소인 Albedo 모듈을 적용해서 안면에 들어오는 빛의 양을 계산하여 Albedo 맵을 생성한다. 그 리고 기본 사항 모듈의 다른 구성 요소인 FLAME 모듈에서는 입력 영상에 대한 기본적인 3D 안면 형태를 생 성한다. 이와 동시에 상세사항 모듈을 적용해서 실제 사람과 다르게 이목구비가 변형된 웹툰 캐릭터 영상의 표정이나 얼굴 깊이와 같은 특징을 살리는 세부사항을 추출한다. 계산한 세부사항들을 토대로 세부사항 맵을 생성하여 앞서 FLAME 모듈에서 생성된 3D 안면 형태와 결합하여 세부사항 안면 형태를 생성한다. 그 후 Albedo 모듈에서 생성된 Albedo 맵까지 적용하면 최종적으로 웹툰 캐릭터 영상에 대한 3D 안면 재구성이 완 료된다. 본 연구에서는 웹툰 캐릭터뿐만 아니라 안면이 스타일화된 애니메이 션 캐릭터에 대해서도 결과를 생성하고, 이를 기존 연구와 비교하여 그 우수성을 입증한다.

Roridin E, a constituent of Podostroma cornu-damae, has excellent medicinal properties and exhibits powerful anticancer effects. However, being a poisonous mushroom, P. cornu-damae is difficult to find in its natural environment. Therefore, only a limited amount of roridin E is isolated from P. cornu-damae. We isolated roridin E from mycelial culture medium of P. cornu-damae and examined its effects on five breast cancer cell lines. The purity of the isolated roridin E was 96%. It exhibited anticancer effects on all five breast cancer cell lines, with an IC50 of at least 0.0006 mg/L. To the best of our knowledge, this is the first report on the investigation of effects of roridin E isolated from the mycelium of P. cornu-damae mushrooms on five breast cancer cell lines. However, safety issues should be investigated and addressed in future studies before roridin E is used in clinical settings.

본 연구는 신생아 검사 중 포수클로랄(chloral hydrate)을 투여 후 진행되는 신생아 진정 검사 대비 진정 대체 방식 중 하나인 피드 및 랩(feed and wrap) 방식의 유용성을 평가한 연구이다. 본 연구에선 진정으로 진행한 신생아의 두뇌 T2 축면 영상과 피드 및 랩 방식으로 진행한 같은 영상 각 30개의 운동 허상(motion artifact)과 백질과 회백질의 구분 정도를 두 명의 영상의학과 전문의가 정성적으로 평가하였고, 운동 허상을 측정하기 위해서 위상부호화(phase encoding) 방향의 배경 영역(background area)의 평균 신호 강도(mean signal intensity)를 구하여서 정량적 방식으로 평가하였다. 또한 총검사 시간을 정리한 뒤 정량적 방식으로 평가하였고 투약 기록의 여부와 간호일지를 토대로 피드 및 랩 방식의 총 39건의 검사 건수 대비 성공률을 측정하였다. 운동 허상의 정량적 평가와 영상 품질의 정성적 평가 모두에서 두 집단은 유의미한 차이가 없었으나, 검사 시간의 정량적 평가에선 p값이 0.001로 유의한 차이가 있었다. 피드 및 랩 방식의 총검사 건수 대비 성공률은 100%였다. 결론적으로 본 논문에선 피드 및 랩 방식과 진정 방식의 영상 품질이 유의한 차이가 없고 성공률이 높기에 유용하다고 판단하였으나, 검사 시간이 더 지연되는 한계가 있다는 사실을 확인하였다.