• Journal of the Korea Convergence Society
• /
• v.12 no.1
• /
• pp.219-223
• /
• 2021

This study aims to present a non-face-to-face teaching methodology for the theory and practical lessons of design majors, especially for the arts and sports field. It was conducted to improve the existing teaching models after the non-face-to-face online lectures which began with COVID-19. Various existing smart learning methods such as online classes, interactive classes, and flip learning were reviewed, and a method to efficiently manage practical skills by supplementing the shortcomings of each study was suggested. 4 stages of teaching development - setting a teaching method, teaching progress, evaluation, and follow-up management-were designated and applied to the class of design majors. The result showed that it is effective in terms of teaching method and progress; however, the limitations of non-face-to-face classes were found in the stages of evaluation and follow-up management. Therefore, it is expected that further research on evaluation and follow-up management, such as specific practical instruction methods is required to improve completion.

• Journal of Christian Education in Korea
• /
• v.64
• /
• pp.347-368
• /
• 2020

The purpose of this study was explored the effect of Flipped PBL on Communication and Problem Solving skill of University students. For study, students in the college of education at A University and S University located in Gyeonggi-do were set up 60 students. They majored in Christian education and early childhood education, and the level of education was similar in the same region and department. In the experiment, the comparison group and the experimental group were separated, and data were collected through pre-test before class and post-test after class, followed by analysis covariance(ANCOVA). Between the comparative group performing flipped learning and the experimental group performing flipped PBL, the problem solving ability and communication ability using the flipped PBL averaged 4.28 and the communication ability averaged 4.06. As a result, it was higher than each of the comparison groups, confirming that it had a significant effect. This will be able to present a flipped PBL class design as an alternative method to overcome the limitations of the existing flip learning in a changing classroom situation and to improve the communication and problem solving ability of the learner.

• Journal of radiological science and technology
• /
• v.43 no.6
• /
• pp.511-518
• /
• 2020

In the pandemic situation due to corona 19, we examined the satisfactory level of students' between non-face-to-face and online classes in the first year of radiology. In the case of online non-face classes related to radiation majors, it is considered better to conduct real-time lectures than recorded lectures, which allows students to understand difficult radiation majors throughout real-time questions and answers, and at the same time allows them to sympathize with difficult subjects such as intimacy and bonds between students. As a result of the analysis of the answers to questionnaires, the coefficient of answer correlation between first and second grade students is P>0.05, and there is no answer correlation between grades, and it can be seen that the higher the grade, the higher the demand for face-to-face classes was.

• Journal of Digital Convergence
• /
• v.18 no.11
• /
• pp.355-363
• /
• 2020

This study is conducted to examine the changes in class perception and career perception according to the teaching method for dental hygiene students in some areas of Gangwon-do. Using the same questionnaire, the first questionnaire was conducted after contact classes in December 2019, and the second questionnaire was conducted after non-contact classes in July 2020. Data were analyzed using SPSS Windows 24.0, and the results are as follows. There were differences in class perception and career perception according to grades, and career interest, career curiosity, and career confidence were different according to school life satisfaction and professional tooth cleaning Class satisfaction. In the case of non-contact classes, the proportion of career decisions through the internet increased, and career confidence also increased. These results can be provided as basic data necessary for the use of methods such as flip-learning classes in consideration of the convergence aspect in the practical subject class management and for the production of standardized practical videos for practical tests.

• Journal of the Institute of Convergence Signal Processing
• /
• v.21 no.4
• /
• pp.183-194
• /
• 2020

Recently, the number of hip dysplasia (DDH) that occurs during infant and child growth has been increasing. DDH should be detected and treated as early as possible because it hinders infant growth and causes many other side effects In this study, two modelling techniques were used for multiple training techniques. Based on the results after the first transformation, the training was designed to be possible even with a small amount of data. The vertical flip, rotation, width and height shift functions were used to improve the efficiency of the model. Adam optimization was applied for parameter learning with the learning parameter initially set at 2.0 x 10e-4. Training was stopped when the validation loss was at the minimum. respectively A novel image overlay system using 3D laser scanner and a non-rigid registration method is implemented and its accuracy is evaluated. By using the proposed system, we successfully related the preoperative images with an open organ in the operating room

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.666-669
• /
• 2024

연합학습은 원본 데이터를 공유하지 않고 모델을 학습할 수 있는 각광받는 프라이버시를 위한 학습방법론이다. 이를 위해 참여자의 데이터를 수집하는 대신, 데이터를 인공지능 모델 학습의 요소들(가중치, 기울기 등)로 변환한 뒤, 이를 공유한다. 이러한 강점에 더해 기존 연합학습을 개선하는 방법론들이 추가적으로 연구되고 있다. 기존 연합학습은 모델 가중치를 평균내는 것으로 참여자 간에 동일한 모델 구조를 강요하기 때문에, 참여자 별로 자신의 환경에 알맞은 모델 구조를 사용하기 어렵다. 이를 해결하기 위해 지식 증류 기반의 연합학습 방법(Knowledge Distillation-based Federated Learning)으로 서로 다른 모델 구조를 가질 수 있도록(Model Heterogenousity) 하는 방법이 제시되고 있다. 연합학습은 여러 참여자가 연합하기 때문에 일부 악의적인 참여자로 인한 모델 포이즈닝 공격에 취약하다. 수많은 연구들이 기존 가중치를 기반으로한 연합학습에서의 위협을 연구하였지만, 지식 증류 기반의 연합학습에서는 이러한 위협에 대한 조사가 부족하다. 본 연구에서는 최초로 지식 증류 기반의 연합학습에서의 모델 성능 하락 공격에 대한 위협을 실체화하고자 한다. 이를 위해 우리는 GMA(Gaussian-based Model Poisoning Attack)과 SMA(Sign-Flip based Model Poisoning Attack)을 제안한다. 결과적으로 우리가 제안한 공격 방법은 실험에서 최신 학습 기법에 대해 평균적으로 모델 정확도를 83.43%에서 무작위 추론에 가깝게 떨어뜨리는 것으로 공격 성능을 입증하였다. 우리는 지식 증류 기반의 연합학습의 강건성을 평가하기 위해, 새로운 공격 방법을 제안하였고, 이를통해 현재 지식 증류 기반의 연합학습이 악의적인 공격자에 의한 모델 성능 하락 공격에 취약한 것을 보였다. 우리는 방대한 실험을 통해 제안하는 방법의 성능을 입증하고, 결과적으로 강건성을 높이기 위한 많은 방어 연구가 필요함을 시사한다.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.26 no.2
• /
• pp.28-36
• /
• 2022

As port structures are exposed to various extreme external loads such as wind (typhoons), sea waves, or collision with ships; it is important to evaluate the structural safety periodically. To monitor the port structure, especially the rubber fender, a fender segmentation system using a vision sensor and deep learning method has been proposed in this study. For fender segmentation, a new deep learning network that improves the encoder-decoder framework with the receptive field block convolution module inspired by the eccentric function of the human visual system into the DenseNet format has been proposed. In order to train the network, various fender images such as BP, V, cell, cylindrical, and tire-types have been collected, and the images are augmented by applying four augmentation methods such as elastic distortion, horizontal flip, color jitter, and affine transforms. The proposed algorithm has been trained and verified with the collected various types of fender images, and the performance results showed that the system precisely segmented in real time with high IoU rate (84%) and F1 score (90%) in comparison with the conventional segmentation model, VGG16 with U-net. The trained network has been applied to the real images taken at one port in Republic of Korea, and found that the fenders are segmented with high accuracy even with a small dataset.

• The Korean Journal of Applied Statistics
• /
• v.36 no.2
• /
• pp.101-114
• /
• 2023

The data augmentation technique is effectively used to solve the problem of overfitting the model by allowing the training dataset to be viewed from various perspectives. In addition to image augmentation techniques such as rotation, cropping, horizontal flip, and vertical flip, occlusion-based data augmentation methods such as Cutmix and Cutout have been proposed. For models based on speech data, it is possible to use an occlusion-based data-based augmentation technique after converting a 1D speech signal into a 2D spectrogram. In particular, SpecAugment is an occlusion-based augmentation technique for speech spectrograms. In this study, we intend to compare and study data augmentation techniques that can be used in the problem of false-voice detection. Using data from the ASVspoof2017 and ASVspoof2019 competitions held to detect fake audio, a dataset applied with Cutout, Cutmix, and SpecAugment, an occlusion-based data augmentation method, was trained through an LCNN model. All three augmentation techniques, Cutout, Cutmix, and SpecAugment, generally improved the performance of the model. In ASVspoof2017, Cutmix, in ASVspoof2019 LA, Mixup, and in ASVspoof2019 PA, SpecAugment showed the best performance. In addition, increasing the number of masks for SpecAugment helps to improve performance. In conclusion, it is understood that the appropriate augmentation technique differs depending on the situation and data.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.4
• /
• pp.428-436
• /
• 2020

The purpose of this study is to analyze the needs for developing a curriculum for strengthening the long-term care service expertise and job competency. Specifically, the researchers analyzed previous studies on national long-term care services and national policy data, and conducted focus group interviews with 14 experts from related agencies. Activity theory was applied as a framework for analysis and a questionnaire about the importance and difficulty of subjects from 25 long-term service employees was administered for validating the results of the qualitative data analysis. The upper part of the subject-goal-tool of the activity system was considered the main area of action, and the following rule-community-division was divided into contextual parts for action, and the implications for demand analysis and future operation of the online curriculum are summarized. In total, six courses were required for development. These courses could be applied to as a learner-centered flip learning for long-term care service workers and various educational methods of collective education and supplementary education have been proposed. Based on the study results, implications in the educational field for effective management of courses were suggested at the end of the study.

• Journal of Internet of Things and Convergence
• /
• v.6 no.1
• /
• pp.31-38
• /
• 2020

This study was intended to verify the parameter effects of empowerment in the effect of social support of the employees of youth training facilities on the quality of service. The purpose of this study was to provide basic data for improving the service quality of youth training facility workers. As a result of this study, first, social support and empowerment had a positive effect on service quality. Second, the partial mediating effect of empowerment can be confirmed. Based on the results of this study, the following suggestions were made. First, for the social support of the employees, the middle managers and facility managers of youth training facilities should use the Internet environment such as the Internet and smart phones. Second, in order to improve empowerment, support for information related to work, material support related to compensation, and evaluation support related to business processing should be provided. Third, to improve the empowerment of workers, it is necessary to augmented reality(AR), virtual reality(VR), and flip learning program using the Internet of Things environment.