• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.550-551
• /
• 2021

Positioning technology is performing important functions in augmented reality, smart factory, and autonomous driving. Among the positioning techniques, the positioning method using beacons has been considered a challenging task due to the deviation of the RSSI value. In this study, the position of a moving object is predicted by training a neural network that takes the RSSI value of the receiver as an input and the distance as the target value. To do this, the measured distance versus RSSI was collected.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.124-125
• /
• 2023

Defect detection during liquid crystal display (LCD) manufacturing has always been a critical challenge. This study aims to address this issue by proposing a data augmentation method based on generative adversarial networks (GAN) to improve defect identification accuracy in LCD production. By leveraging synthetically generated image data from GAN, we effectively augment the original dataset to make it more representative and diverse. This data augmentation strategy enhances the model's generalization capability and robustness on real-world data. Compared to traditional data augmentation techniques, the synthetic data from GAN are more realistic, diverse and broadly distributed. Experimental results demonstrate that training models with GAN-generated data combined with the original dataset significantly improves the detection accuracy of critical defects in LCD manufacturing, compared to using the original dataset alone. This study provides an effective data augmentation approach for intelligent quality control in LCD production.

• ETRI Journal
• /
• v.46 no.3
• /
• pp.461-472
• /
• 2024

We propose a network traffic prediction model based on linear and nonlinear model combination. Network traffic is modeled by an autoregressive moving average model, and the error between the measured and predicted network traffic values is obtained. Then, an echo state network is used to fit the prediction error with nonlinear components. In addition, an improved slime mold algorithm is proposed for reservoir parameter optimization of the echo state network, further improving the regression performance. The predictions of the linear (autoregressive moving average) and nonlinear (echo state network) models are added to obtain the final prediction. Compared with other prediction models, test results on two network traffic datasets from mobile and fixed networks show that the proposed prediction model has a smaller error and difference measures. In addition, the coefficient of determination and index of agreement is close to 1, indicating a better data fitting performance. Although the proposed prediction model has a slight increase in time complexity for training and prediction compared with some models, it shows practical applicability.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.823-828
• /
• 2008

This study set out investigate what kind of changes would be caused to the craving of alcohol dependent patients by stimuli through virtual reality in the preparation stage for drinking refusal training. With regard to stimulation, it included alcohol exposure, a positive situation, and a negative situation on the basis of drinking encouraging situations(social pressure) known as drinking stimulating situations for alcoholics. The purpose of the study is to provide fundamental materials for the development of new training programs to refuse alcohol and for the examination of the possibility of utilizing virtual reality technology as a new treatment and training tool for alcoholics. To this end, a virtual reality program was conducted involving 12 alcoholics admitted to Severance Mental Health Hospital in Gyeonggi province from December 2006 to September 2007. The data was dealt with various statistic analyses such as frequency analysis, Wilcoxon Matched-pairs Signed-Ranks Test using SPSS/WIN 11.5 The analysis results indicate that avatars encouraging drinking caused more craving than just the background, that the scenes with alcohol exposed caused more craving than the scenes with no alcohol exposed, and that there were no significant changes to craving according to a positive or negative situation. The results confirmed that training sessions using virtual reality presented a situation and environment of drinking pressure that's similar to the actual social pressure and that the virtual reality approach had enormous potential as an effective treatment tool when combined with the existing treatment techniques.

• 대한공업교육학회지
• /
• v.35 no.2
• /
• pp.82-111
• /
• 2010

The goal of this paper was to understand the skill-up experiences of ex-participants of the customized training program in Technical High Schools for Small and Medium Business during first 2 years in Company through qualitative interviews with 3 purposefully selected ex-participants. Their core skill-up experience in this period was assumed as the shift from' dependent worker' to 'independent worker' on the base of literature review. The results of this study were following : 1. The small and medium companies offered a few formal training for newcomers, production-site orientation through short job rotation, linking them with skilled workers and job manuals or job standards. 2. Authentic skill-up experiences were combined with a structured reprimand, peer learning, deep learning through reflection on one's own experiences. 3. There were a few handicap conditions that disturbed their skill-up activities: the skilled worker don't open their skill toward new corner; the ex-participants in company had no time to learn anything meaningful to up-grade their competency.

• Journal of Christian Education in Korea
• /
• v.62
• /
• pp.135-160
• /
• 2020

This study closely analyzed the curriculum of the Christian Education Counseling Department and the general Education Counseling Department, and found the current status and problems of the curriculum of the Christian Education Counseling Department and the general Education Counsel Department. This study presented a balanced curriculum of the Christian Education Counseling Department with above analysis. For this purpose, the analysis focused on the educational operation process of Christian education counseling departments and general education counseling departments, such as educational goals, subjects, and counseling practical training. The Christian Education Counseling Department and the general Education Counseling Department are often combined with departments such as Christian Education, Youth, Children and Youth, and Lifelong Education, with the characteristics of convergence majors, so the basic subjects of the department were analyzed to have a higher percentage of subjects than counseling subjects. The results of the analysis showed that both departments lacked a considerable number of subjects related to counseling practical training. In the counseling course, the subjects of personal analysis, education analysis, counseling ethics, and counseling case super-vision for the professional development of counselors are still lacking, according to the analysis. In order to train counselors, it was analyzed that the system of systematic clinical practice system, various counseling analysis for counselor education, and the expansion of super vision subjects were urgently needed. In a modern society where the demand for counseling and the need for counseling experts are increasing as society becomes more complex, it is hoped that Korean universities will be able to actively contribute and cooperate in developing models of counseling education and training counseling experts through them, focusing on standardized indicators for fostering counselors.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.40 no.1
• /
• pp.19-29
• /
• 2003

This paper presents a novel hybrid classifier for face recognition with artificially generated virtual training samples. We utilize both the nearest neighbor approach in feature angle space and a connectionist model to obtain a synergy effect by combining the results of two heterogeneous classifiers. First, a classifier called the nearest feature angle (NFA), based on angular information, finds the most similar feature to the query from a given training set. Second, a classifier has been developed based on the recall of stored frontal projection of the query feature. It uses a frontal recall network (FRN) that finds the most similar frontal one among the stored frontal feature set. For FRN, we used an ensemble neural network consisting of multiple multiplayer perceptrons (MLPs), each of which is trained independently to enhance generalization capability. Further, both classifiers used the virtual training set generated adaptively, according to the spatial distribution of each person's training samples. Finally, the results of the two classifiers are combined to comprise the best matching class, and a corresponding similarit measure is used to make the final decision. The proposed classifier achieved an average classification rate of 96.33% against a large group of different test sets of images, and its average error rate is 61.5% that of the nearest feature line (NFL) method, and achieves a more robust classification performance.

• Journal of the Korea Convergence Society
• /
• v.12 no.6
• /
• pp.27-32
• /
• 2021

The purpose of this study was to compare the immediate effect of EMS (electrical muscle stimulation) on rectus femoris thickness during resisted knee extension exercise in healthy adults. This experiment was conducted on 20 healthy adults as pretest-posttest nonequivalent one group design. The subject's 1RM of both knee extension was measured indirectly using an elastic band, and the knee extension resistance exercise using an elastic band was applied to high intensity (80% of 1RM) on the right leg and low intensity (50% of 1RM with EMS) on the left leg, which were applied with 5 sets. Muscle measurements were performed on the rectus femoris (1/2 site, 1/4 site) using ultrasonography before and after exercise. There was a statistically significant difference on the thickness of the rectus femoris in low intensity exercise of the elastic band applied with EMS between pre-test and post-test (p<.05). The results of this study showed that elastic band low intensity exercise combined with EMS had an immediate effective in increasing the thickness of rectus femoris. Based on this result, it is also necessary to verify the effectiveness of intervention methods incorporating low-intensity resistance exercises applying EMS to elderly who cannot exercise high-intensity resistance training in the future, and to develop exercise programs for various body parts.

• Journal of Korea Entertainment Industry Association
• /
• v.13 no.8
• /
• pp.583-593
• /
• 2019

The purpose of this study was to investigate the relation between and subjective instability pain behavior (SIPB) and physical instability test(PIT) according to the presence of characteristic past histories(CPH) in people with chronic low back pain(CLBP). Forty CLBP subjects participated in this study. The presence of four characteristics past histories(long term history, traumatic experience, sports activities, neurologic sign) were examined. According to presence number(PN) of CPH, subjects were divided into 5 groups[group 1(PN:0): n=8, group 2(PN:1): n=8, group 3(PN:2): n=8, group 4(PN:3): n=8, group 5(PN:4): n=8]. After 16 items were examined for the SIPBs, then Seven PITs were conducted, and the results were scored. The SIPBs and PITs were compared according to the presence numbers of CPH, and the relation between them was analyzed. There was a significant difference(p<.05) in both SIPB scores and PIT scores in the comparison of groups according to the presence number of CPH. There was high positive correlation between the presence numbers of CPH and SIPB score(r=.819, p=.000) and PIT score(r=.606, p=.000). Also, there was a correlation between SIPB score and PIT score(r=.571, p=.000). Based on the findings in the present study, the presence of three or more CPH in people with CLBP may be a useful variable in the diagnosis of lumbar instability. The combined findings of the three variables such as CPH, SIPB, and PIT can improve the accuracy of lumbar instability diagnosis.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_4
• /
• pp.1111-1123
• /
• 2023

In recent times, semantic image segmentation methods using deep learning models have been widely used for monitoring changes in surface attributes using remote sensing imagery. To enhance the performance of various UNet-based deep learning models, including the prominent UNet model, it is imperative to have a sufficiently large training dataset. However, enlarging the training dataset not only escalates the hardware requirements for processing but also significantly increases the time required for training. To address these issues, transfer learning is used as an effective approach, enabling performance improvement of models even in the absence of massive training datasets. In this paper we present three transfer learning models, UNet-ResNet50, UNet-VGG19, and CBAM-DRUNet-VGG19, which are combined with the representative pretrained models of VGG19 model and ResNet50 model. We applied these models to building extraction tasks and analyzed the accuracy improvements resulting from the application of transfer learning. Considering the substantial impact of learning rate on the performance of deep learning models, we also analyzed performance variations of each model based on different learning rate settings. We employed three datasets, namely Kompsat-3A dataset, WHU dataset, and INRIA dataset for evaluating the performance of building extraction results. The average accuracy improvements for the three dataset types, in comparison to the UNet model, were 5.1% for the UNet-ResNet50 model, while both UNet-VGG19 and CBAM-DRUNet-VGG19 models achieved a 7.2% improvement.