• Journal of Broadcast Engineering
• /
• v.25 no.6
• /
• pp.980-993
• /
• 2020

Deep-learning based style transfer has recently attracted great attention, because it provides high quality transfer results by appropriately reflecting the high level structural characteristics of images. This paper deals with the problem of providing more stable and more diverse style transfer results of such deep-learning based style transfer method. Based on the investigation of the experimental results from the wide range of hyper-parameter settings, this paper defines the problem of the stability and the diversity of the style transfer, and proposes a partial loss normalization method to solve the problem. The style transfer using the proposed normalization method not only gives the stability on the control of the degree of style reflection, regardless of the input image characteristics, but also presents the diversity of style transfer results, unlike the existing method, at controlling the weight of the partial style loss, and provides the stability on the difference in resolution of the input image.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.19 no.6
• /
• pp.118-133
• /
• 2020

Research interest in the Mobility-as-a-Service (MaaS) concept for enhancing users' mobility experience is increasing. In particular, dynamic pricing techniques based on reinforcement learning have emerged since adjusting prices based on the demand is expected to help mobility services, such as taxi and car-sharing services, to gain more profit. This paper provides a simulation framework that considers more practical factors, such as demand density per location, preferred prices, the distance between users and drivers, and distance to the destination that critically affect the probability of matching between the users and the mobility service providers (e.g., drivers). The aforementioned new practical features are reflected on a data structure referred to as the Matching Matrix. Using an efficient algorithm of computing the probability of matching between the users and drivers and given a set of precisely identified high-demand locations using HDBSCAN, this study developed a better reward function that can gear the reinforcement learning process towards finding more realistic dynamic pricing policies.

• The Journal of the Korea Contents Association
• /
• v.21 no.8
• /
• pp.1-9
• /
• 2021

Deep learning in computer vision has made accelerated improvement over a short period but large-scale learning data and computing power are still essential that required time-consuming trial and error tasks are involved to derive an optimal network model. In this study, we propose a similar image classification performance improvement method based on CR (Confusion Rate) that considers only the characteristics of the data itself regardless of network optimization or data reinforcement. The proposed method is a technique that improves the performance of the deep learning model by calculating the CRs for images in a dataset with similar characteristics and reflecting it in the weight of the Loss Function. Also, the CR-based recognition method is advantageous for image identification with high similarity because it enables image recognition in consideration of similarity between classes. As a result of applying the proposed method to the Resnet18 model, it showed a performance improvement of 0.22% in HanDB and 3.38% in Animal-10N. The proposed method is expected to be the basis for artificial intelligence research using noisy labeled data accompanying large-scale learning data.

• Journal of the Korean Society of Industry Convergence
• /
• v.25 no.1
• /
• pp.61-70
• /
• 2022

Fire damage time in high-rise buildings and wildland fire increasing every year. The use of high-pressure fire pumps is required to effectively extinguish fires. Reflecting the curvature effect of the fire hose occurring at the actual fire fighting site, this study provides a database of pressure drop, discharge velocity and maximum discharge height through C FD numerical analysis and it can provide using standards for fire extinguishing. Two Reynolds numbers of 200000 and 400000 were numerically analyzed at 0° -180° bending with water of 25℃ as a working fluid in hoses with a diameter of 65mm, a length of 15m, and a radius of curvature of 130mm. Realizable k-ε turbulence model was used and standard wall function was used. The pressure drop increases as the bending angle increases, and the maximum value at 90° and then decreases. The increasing rate is greater than the decrease. The velocity of the secondary flow also decreases after having the maximum value at 90°. The decreasing rate is greater than the increase. The turbulent kinetic energy increases to 120° and decreases with the maximum value. Pressure drop, velocity of the secondary flow, and turbulence kinetic energy are measured larger in the second bending region than in the first bending region.

• The Journal of the Korea Contents Association
• /
• v.22 no.1
• /
• pp.47-57
• /
• 2022

Demands for tangible contents using VR/AR technologies are much bigger as contactless services such as sports, physical activity, and fitness are expanded after COVID-19. A variety of technologies such as an offer and analysis of tangible data through a sensor technology, users' physical movement sensing through a motion recognition sensor, a real-time measurement of a physical skeleton point a multiple access to a real-time video, and AI training are being utilized as main technologies. This case study utilized motion recognition technologies as the study on tangible contents necessary for indoor-based physical education, sports, and fitness in the contactless environment and suggested cases to develop the physical measurement contents by design approach for the measurement assessment necessary for the development in tangible contents. The research established lists of the measurement assessment based on professionals' consultations within the measurement assessment function through the test to plan tangible contents and developed tangible contents by reflecting them as assessment measurement elements of tangible contents. The research can be utilized as the design approach of industrial companies which intend to develop tangible contents as well as reference cases of the research on contactless tangible contents for the sports and physical education.

• Fisheries and Aquatic Sciences
• /
• v.25 no.8
• /
• pp.450-461
• /
• 2022

A randomized, double-blind, and placebo-controlled clinical study was used to determine the cognitive functions related to working memory (WM) and antioxidant properties of fermented Laminaria japonica (FLJ) on healthy volunteers. Eighty participants were divided into a placebo group (n = 40) and FLJ group (n = 40) that received FLJ (1.5 g/day) for 6 weeks. Memory-related blood indices (brain-derived neurotrophic factor, BDNF; angiotensin-converting enzyme; human growth hormone, HGH; insulin-like growth factor-1, IGF-1) and antioxidant function-related indices (catalase, CAT; malondialdehyde, MDA; 8-oxo-2'-deoxyguanosine, 8-oxo-dG; thiobarbituric acid reactive substances, TBARS) were determined before and after the trial. In addition, standardized cognitive tests were conducted using the Cambridge Neuropsychological Test Automated Batteries. Furthermore, the Korean Wechsler Adult Intelligence Scale (K-WAIS)-IV, and the Korean version of the Montreal Cognitive Assessment (MoCA-K) were used to assess the pre and post intake changes on WM-related properties. According to the results, FLJ significantly increased the level of CAT, BDNF, HGH, and IGF-1. FLJ reduced the level of TBARS, MDA, and 8-oxo-dG in serum. Furthermore, FLJ improved physical activities related to cognitive functions such as K-WAIS-IV, MoCA-K, Paired Associates Learning, and Spatial Working Memory compared to the placebo group. Our results suggest that FLJ is a potential candidate to develop functional materials reflecting its capability to induce antioxidant mechanisms together with WM-related indices.

• Journal of Christian Education in Korea
• /
• v.65
• /
• pp.103-131
• /
• 2021

The purpose of this study is to identify the direction of future Christian ministry by examining the educational/pastoral ministries of Korean churches during the COVID-19 period. The study used the case study method, where pastors from four different churches were the primary sources of data. Although each church has its own unique ministry, special early morning prayer meetings, family worship services, family Vacation Bible Schools, Bible writing and reading programs, and on-line discipleship training programs were common. In addition, each church implemented special programs that reflected the characteristics and needs of the church, like a talk show, visible radio, surfing meeting, and book-making of Bible manuscripts. Based on these results, ministry programs reflecting the church's pastoral philosophy, strengthening community consciousness, restoring the spiritual function of the family, utilizing online teaching methods, and psychological support for the underprivileged are proposed. In addition, the recovery of Christian publicity and the help from denomination or specialized institutions to develop the programs and online materials were suggested. Through this study, the direction of educational/pastoral ministry in a time of COVID-19 and post-COVID-19 is proposed.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.9
• /
• pp.149-155
• /
• 2022

The popularity of e-commerce systems on the Internet is increasing day by day, and the recommendation system, as a core function of these systems, greatly reduces the effort to search for desired products by recommending products that customers may prefer. The collaborative filtering technique is a recommendation algorithm that has been successfully implemented in many commercial systems, but despite its popularity and usefulness in academia, the memory-based implementation has inaccuracies in its reference neighbor. To solve this problem, this study proposes a new time-aware collaborative filtering technique that integrates and utilizes the neighbors of each item and each user, weighting the recent similarity more than the past similarity with them, and reflecting it in the recommendation list decision. Through the experimental evaluation, it was confirmed that the proposed method showed superior performance in terms of prediction accuracy than other existing methods.

• Journal of The Korea Institute of Healthcare Architecture
• /
• v.28 no.3
• /
• pp.7-16
• /
• 2022

Purpose: This study aims to establish the basic directions of the modular airborne infection isolation(AII) ward. Considering a specific function and purposed use as a modular AII ward, it is a chance to derive an address of current modular technology by overview the limitation and improvement of the existing modular architecture. Methods: In addition to the literature analysis on the configuration system of mobile hospitals, research cases on the operational effectiveness of the domestic and foreign mobile construction systems are analyzed. Results: In order to meet the various and strict space guidelines of the AII ward and a chance to improve limitations of uniformed existing modulars, AII modular the negative pressurized care setting should be minimized a structural restriction for reflecting its system on a architectural plan. For this unique requirements, it could be possible to apply various space boxes called infill box which needs to secure a large-scale space. So, a rahmen structure system could be adaptable for this purpose. A dead space between beams of the rahmen structure is to be used for MEP installation. Partial separation, dismantling, and repair should be possible by separating the MEP and infill box from the structure. The infill box must keep 3.5m width under the current Road Traffic Act. Implications: It is necessary to utilize and develop an improved construction method that can reduce the problems of existing steel modular and PC modular.

• Fashion & Textile Research Journal
• /
• v.24 no.4
• /
• pp.471-478
• /
• 2022

In roadside workplaces, more attention should be paid to the safety of workers. The roadside workers underestimate the effect of the brightness of their clothes and judge that drivers will recognize them easily, and the drivers misjudge that the roadside workers are far away and that the vehicle can be stopped in sufficient time. Therefore, customized safety education reflecting this and wearing work clothes with certified visibility functions are required. In Korea, it is not compulsory for roadside workers and vehicle guide attendants to wear work clothes with a visibility function. In this study, the distance ahead perceived by drivers was measured using manikins wearing certified and non-certified reflective safety vests. The perception distance of the non-certified reflective safety vest was 1.4 times longer than that of the certified reflective safety vest, thus confirming the importance of wearing a certified reflective safety vest. To prevent roadside workers from suffering traffic accidents, we propose the enactment of a law that makes it mandatory for them to wear high-visibility safety clothing. Specifically, Article 32 of the Enforcement Regulation of the Road Traffic Act should include high-visibility safety clothing in life protection equipment, and additionally, to prevent secondary accidents, we propose the enactment of a law requiring the installation and wearing of certified reflective safety vests in vehicles.