• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.3
• /
• pp.473-482
• /
• 2022

Wearing a mask is also necessary to limit the risk of infection in today's era of COVID-19 and wearing a helmet is inevitable for the safety of personnel who works in a dangerous working environment such as construction sites. This paper proposes an effective deep learning model, HelmetMask-Net, to classify both Helmet and Mask. The proposed HelmetMask-Net is based on CNN which consists of data processing, convolution layers, max pooling layers and fully connected layers with four output classifications, and 4 classes for Helmet, Mask, Helmet & Mask, and no Helmet & no Mask are classified. The proposed HelmatMask-Net has been chosen with 2 convolutional layers and AdaGrad optimizer by various simulations for accuracy, optimizer and the number of hyperparameters. Simulation results show the accuracy of 99% and the best performance compared to other models. The results of this paper would enhance the safety of personnel in this era of COVID-19.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.3
• /
• pp.483-490
• /
• 2022

Malaria is a disease caused by a parasite and it is prevalent in all over the world. The usual method used to recognize malaria cells is a thick and thin blood smears examination methods, but this method requires a lot of manual calculation, so the efficiency and accuracy are very low as well as the lack of pathologists in impoverished country has led to high malaria mortality rates. In this paper, a malaria cell image recognition model using transfer learning is proposed, which consists in the feature extractor, the residual structure and the fully connected layers. When the pre-training parameters of the VGG-19 model are imported to the proposed model, the parameters of some convolutional layers model are frozen and the fine-tuning method is used to fit the data for the model. Also we implement another malaria cell recognition model without residual structure to compare with the proposed model. The simulation results shows that the model using the residual structure gets better performance than the other model without residual structure and the proposed model has the best accuracy of 97.33% compared to other recent papers.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.2
• /
• pp.81-86
• /
• 2022

Given that time series are used in various fields, such as finance, IoT, and manufacturing, data analytical methods for accurate time-series forecasting can serve to increase operational efficiency. Among time-series analysis methods, multi-horizon forecasting provides a better understanding of data because it can extract meaningful statistics and other characteristics of the entire time-series. Furthermore, time-series data with exogenous information can be accurately predicted by using multi-horizon forecasting methods. However, traditional deep learning-based models for time-series do not account for the heterogeneity of inputs. We proposed an improved time-series predicting method, called the temporal fusion transformer method, which combines multi-horizon forecasting with interpretable insights into temporal dynamics. Various real-world data such as stock prices, fine dust concentrates and electricity consumption were considered in experiments. Experimental results showed that our temporal fusion transformer method has better time-series forecasting performance than existing models.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.4
• /
• pp.577-586
• /
• 2022

Detection and classification of steel surface defects are critical for product quality control in the steel industry. However, due to its low accuracy and slow speed, the traditional approach cannot be effectively used in a production line. The current, widely used algorithm (based on deep learning) has an accuracy problem, and there are still rooms for development. This paper proposes a method of steel surface defect detection combining EfficientNetV2 for image classification and YOLOv5 as an object detector. Shorter training time and high accuracy are advantages of this model. Firstly, the image input into EfficientNetV2 model classifies defect classes and predicts probability of having defects. If the probability of having a defect is less than 0.25, the algorithm directly recognizes that the sample has no defects. Otherwise, the samples are further input into YOLOv5 to accomplish the defect detection process on the metal surface. Experiments show that proposed model has good performance on the NEU dataset with an accuracy of 98.3%. Simultaneously, the average training speed is shorter than other models.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.36 no.5
• /
• pp.442-453
• /
• 2023

Tandem or multijunction solar cells (MJSCs) can convert sunlight into electricity with higher efficiency (η) than single junction solar cells (SJSCs) by dividing the solar irradiance over sub-cells having distinct bandgaps. The efficiencies of various common SJSC materials are close to the edge of their theoretical efficiency and hence there is a tremendous growing interest in utilizing the tandem/multijunction technique. Recently, III-V materials integration on a silicon substrate has been broadly investigated in the development of III-V on Si tandem solar cells. Numerous growth techniques such as heteroepitaxial growth, wafer bonding, and mechanical stacking are crucial for better understanding of high-quality III-V epitaxial layers on Si. As the choice of growth method and substrate selection can significantly impact the quality and performance of the resulting tandem cell and the terminal configuration exhibit a vital role in the overall proficiency. Parallel and Series-connected configurations have been studied, each with its advantage and disadvantages depending on the application and cell configuration. The optimization of both growth mechanisms and terminal configurations is necessary to further improve efficiency and lessen the cost of III-V on Si tandem solar cells. In this review article, we present an overview of the growth mechanisms and terminal configurations with the areas of research that are crucial for the commercialization of III-V on Si tandem solar cells.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2021.11a
• /
• pp.132-133
• /
• 2021

In recent years, energy harvesting from natural sources and waste heat has been attracting more attention from researchers in response to ever-growing energy demands, high energy prices, and climate-change-mitigation purposes. It is also an important step towards future sustainable energy usages. In thermal dynamic cycles, expanders are playing as the most important equipment for waste heat recovery and energy harvesting as well. As a kind of expander, the bladeless turbine has a promising future and more widely using owning its advantages on relatively long life, good off-design performance, easy operation cleaning and maintenance, a simple structure, no blade corrosion, and low manufacturing costs. There are numerous studies about using the Tesla Turbine as a key technology for energy harvesting in a wide range of applications and conditions. They are presented to help identify technologies that have sufficient potential for applicating to our life and marine industrial engineering. This review paper, initially, presents an overview of current studies both theoretical and experimental of Tesla Turbine usage for waste heat recovery alongside its challenges and investigation on the effect of its configuration, working fluid selection as well. To conclude, future perspectives besides possible ways of transforming waste heat energy to electricity or work, which leads to circular energy, are discussed. The ambition of this paper is to act as a first-hand reference, through the well-defined possible directions, to the young researchers and senior scientists.

• Korean Journal of Agricultural Science
• /
• v.50 no.3
• /
• pp.395-406
• /
• 2023

The large amount of energy required for successful crop production is the main challenge in greenhouse cropping systems. As a response to this challenge a comprehensive evaluation of greenhouse energy consumption was carried out in two structurally similar single-span greenhouses with different thermal curtain positions, with particular attention to energy productivity, specific energy, net energy, and energy ratio. The greenhouses are used for strawberry production. In the R-greenhouse (RGH), the thermal curtain hanged directly at the roof ridge, whereas in the Q-greenhouse (QGH), the thermal curtain was placed 5° from an imaginary vertical axis, from the middle of the roof ridge downwards to the north side of the greenhouse roof. The relevant data were recorded using standard methods. The results indicated that the energy expended in the RGH and QGH systems was 2,186.48 and 2,189.26 MJ/m², respectively. Electricity and nitrogen fertilizer contributed the highest energy input in both greenhouses and in all seasons. The output energy was 3.12 and 3.82 MJ/m², respectively, in RGH and QGH in season I and 4.40 and 4.87 MJ/m² in season II. In terms of energy expended, there was no significant difference between the two greenhouses, nor between the two seasons. These results indicate that greenhouses of the size used in this investigation are not viable in terms of energy productivity, energy-use efficiency, and subsequent economic performance. However, further studies should be conducted to scale-up the information obtained from this investigation.

• (The) Research of the performance art and culture
• /
• no.41
• /
• pp.219-255
• /
• 2020

This article was triggered by Victor H. Mair's book 'Painting and performance'. The book explained that Buddhist paintings are common in the area where Buddhism was spread, and there are also performances using them. And although it has nothing to do with Buddhism, it has been shown that this form of performance can be global. However, the problem was not 'Korea'. It was because there was no record or transmission of the corresponding performance soon. In this situation, the landing method of Samcheok stable temple was announced in 2018. On the one hand, the academic community is very pleased, but on the other hand, it is troubled. The worries are summed up as 'synchronic and diachronic universality'. Is the landing method inherited from the Samcheok stable temple a unique type of temple? Otherwise, it is a question of whether it has been passed down or is it universal at the national level. However, prior to this essential question, we do not yet know the full picture of the stable landing method. So this article was prepared to show the overall outline of the stable landing method. There is a 'picture' in common throughout the landing method, and understanding how to operate it in various ways is the first step in understanding the landing method. There are five repertoires (which are called main halls) that are considered important, and more than that. What these repertoires have in common is the narrative structure of a Buddhist character. In this narrative, the most important thing is the revised figure, and it was the earthly method to inform the contents of the revised figure in various ways. In the case of Byeonsangdo, which serves as a clue to the narrative, there was a problem that could not be seen in the evening without light, which required special design. It is the way of shadow play and Yeongdeung. In other words, there are three types of performances in the landing method. The first is the method of using reparation, and the second is the method of using shadow. The third is the way of eternity. This method is not a selection based on the contents of the repertoire, but a selection based on the performance environment. If there is light and you can see the picture, use reparation. However, in the evening, it was impossible to see it dark (when there was no electricity in the past). The use of the visual method as a tool in this method is to confirm the transition to a visual culture that is a step further from the level of culture. Moreover, unlike the epic narrative, the power of the implied image provided an opportunity for viewers to experience the mystery of Buddhism through emotional stimulation.

• Journal of the Korea Organic Resources Recycling Association
• /
• v.21 no.1
• /
• pp.39-44
• /
• 2013

This study evaluated the effect of fluid flow on the power density in a horizontal-flow microbial fuel cell (MFC). The maximum power densities in four types of flow induced by different channel types in the anode chamber were investigated. The fluid flow at each channel was analyzed using tracer tests. Results of polarization curves showed that the maximum power densities of case 1, 2, 3 and 4 were 95.7, 129.1, 190.9 and 114.2 mW/m2, respectively. Case 3 with a set of guide walls where flow had an S type-like shape showed the highest power density. Based on the Morrill Dispersion Index (MDI) value of case 4, microbial activity would be enhanced since the reactor allows even distribution of substrate but the overflow occurrence would not guarantee stable performance. Therefore, case 3 could be an effective reactor type for MFC because of high electricity generation and stable performance.

• Journal of the Korean Institute of Gas
• /
• v.21 no.5
• /
• pp.36-44
• /
• 2017

The small commercial fuel cell is a new energy system that produces electricity and heat through electrochemical reaction between air and hydrogen. In Korea, hundreds of domestic small commercial fuel cell systems have been installed and operated every years and many parts in fuel cell systems depend on overseas products. KOGAS(Korea Gas Corporation) has developed the fuel processor which is an important part of fuel cell system and has evaluated the long-term durability. And KOGAS has evaluated domestic and overseas NG blower and fuel processor connected to NG blower. The fuel processor developed by KOGAS have maintained an efficiency of 76% and constant performance over 3,000 hours. The NG blower developed in Korea showed similar characteristics as overseas NG blower in the evaluation of power consumption according to rear pressure and outside temperature. The fuel processor module, fuel processor connected to BOP showed excellent performance.