• Journal of the Korea Institute of Military Science and Technology
• /
• v.22 no.1
• /
• pp.49-59
• /
• 2019

Generative adversarial networks(GANs) have received great attention in the machine learning field for their capacity to model high-dimensional and complex data distribution implicitly and generate new data samples from the model distribution. This paper investigates the model training methodology, architecture, and various applications of generative adversarial networks. Experimental evaluation is also conducted for generating synthetic image dataset for defense using two types of GANs. The first one is for military image generation utilizing the deep convolutional generative adversarial networks(DCGAN). The other is for visible-to-infrared image translation utilizing the cycle-consistent generative adversarial networks(CycleGAN). Each model can yield a great diversity of high-fidelity synthetic images compared to training ones. This result opens up the possibility of using inexpensive synthetic images for training neural networks while avoiding the enormous expense of collecting large amounts of hand-annotated real dataset.

• Journal of architectural history
• /
• v.31 no.1
• /
• pp.41-50
• /
• 2022

Traditional wooden buildings deform over time and are vulnerable to fire or earthquakes. Therefore, traditional wooden buildings require continuous management and repair, and securing architectural drawings is essential for repair and restoration. Unlike modernized CAD drawings, traditional wooden building drawings scan and store hand-drawn drawings, and in this process, many noise is included due to damage to the drawing itself. These drawings are digitized, but their utilization is poor due to noise. Difficulties in systematic management of traditional wooden buildings are increasing. Noise removal by existing algorithms has limited drawings that can be applied according to noise characteristics and the performance is not uniform. This study presents deep artificial neural network based noised reduction for architectural drawings. Front/side elevation drawings, floor plans, detail drawings of Korean wooden treasure buildings were considered. First, the noise properties of the architectural drawings were learned with both a cycle generative model and heuristic image fusion methods. Consequently, a noise reduction network was trained through supervised learning using training sets prepared using the noise models. The proposed method provided effective removal of noise without deteriorating fine lines in the architectural drawings and it showed good performance for various noise types.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.28 no.1
• /
• pp.79-95
• /
• 2003

This paper investigates how the weights of performance measures vary depending on corporate life cycle, such as birth, growth, maturity, revival, and decline. Balanced Scorecard performance measures are employed for this investigation. Balanced Scorecard has been widely used for measuring a corporate Performance by incorporating financial and non-financial measures simultaneously. Because these performance measures are related to the compensation and promotion of employees, research of weights of performance measures would be instrumental. Questionnaires from 218 companies are analyzed. Depending on the corporate life cycle, our survey results demonstrate that the weights of the business Performance measures differ In the four Perspectives - financial. customer, Internal Process, and learning/growth. Our results can be used for enhancing the Duality of performance measurement systems.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.6
• /
• pp.2799-2823
• /
• 2019

With the help of learning analytics, MOOCs have wider potential to succeed in learning through promoting self-regulated learning (SRL). The current study aims to apply and validate visualization design guidelines for a MOOC dashboard to enhance such SRL capabilities based on learning analytics. To achieve the research objective, a MOOC dashboard prototype, LM-Dashboard, was designed and developed, reflecting the visualization design guidelines to promote SRL. Then, both expert and learner participants evaluated LM-Dashboard through iterations to validate the visualization design guidelines and perceived SRL effectiveness. The results of expert and learner evaluations indicated that most of the visualization design guidelines on LM-Dashboard were valid and some perceived SRL aspects such as monitoring a student's learning progress and assessing their achievements with time management were beneficial. However, some features on LM-Dashboard should be improved to enhance SRL aspects related to achieving their learning goals with persistence. The findings suggest that it is necessary to offer appropriate feedback or tips as well as to visualize learner behaviors and activities in an intuitive and efficient way for the successful cycle of SRL. Consequently, this study contributes to establishing a basis for the visual design of a MOOC dashboard for optimizing each learner's SRL.

• Korean Journal of Radiology
• /
• v.22 no.6
• /
• pp.983-993
• /
• 2021

Objective: To investigate the image quality of ultralow-dose CT (ULDCT) of the chest reconstructed using a cycle-consistent generative adversarial network (CycleGAN)-based deep learning method in the evaluation of pulmonary tuberculosis. Materials and Methods: Between June 2019 and November 2019, 103 patients (mean age, 40.8 ± 13.6 years; 61 men and 42 women) with pulmonary tuberculosis were prospectively enrolled to undergo standard-dose CT (120 kVp with automated exposure control), followed immediately by ULDCT (80 kVp and 10 mAs). The images of the two successive scans were used to train the CycleGAN framework for image-to-image translation. The denoising efficacy of the CycleGAN algorithm was compared with that of hybrid and model-based iterative reconstruction. Repeated-measures analysis of variance and Wilcoxon signed-rank test were performed to compare the objective measurements and the subjective image quality scores, respectively. Results: With the optimized CycleGAN denoising model, using the ULDCT images as input, the peak signal-to-noise ratio and structural similarity index improved by 2.0 dB and 0.21, respectively. The CycleGAN-generated denoised ULDCT images typically provided satisfactory image quality for optimal visibility of anatomic structures and pathological findings, with a lower level of image noise (mean ± standard deviation [SD], 19.5 ± 3.0 Hounsfield unit [HU]) than that of the hybrid (66.3 ± 10.5 HU, p < 0.001) and a similar noise level to model-based iterative reconstruction (19.6 ± 2.6 HU, p > 0.908). The CycleGAN-generated images showed the highest contrast-to-noise ratios for the pulmonary lesions, followed by the model-based and hybrid iterative reconstruction. The mean effective radiation dose of ULDCT was 0.12 mSv with a mean 93.9% reduction compared to standard-dose CT. Conclusion: The optimized CycleGAN technique may allow the synthesis of diagnostically acceptable images from ULDCT of the chest for the evaluation of pulmonary tuberculosis.

• Journal of Technology Innovation
• /
• v.10 no.2
• /
• pp.131-148
• /
• 2002

This paper explores how international political factors influence the role of conventional external factors in the course of technological learning. The research goes on to investigate whether the role of the techno-economic factors has changed due to the involvement of international political factors in the technological learning mechanism. To this end, this paper examines how US political intervention affected Korean technological learning in the back-end of the nuclear fuel cycle. The export policy, prior consent policy and international political influence of the US are employed as international political factors. The empirical findings show that international political factors are very likely to restrain the impact of the techno-economic factors on technological learning process. Accordingly, this paper hypothesizes that the role of techno-economic factors in the technological learning mechanism is weaker when international political intervention is involved.

• Journal of Science Education
• /
• v.32 no.1
• /
• pp.19-32
• /
• 2008

The purpose of this study was to identify the influence of using the concept cartoons about middle school students' science attitude in the lessons on water cycle unit. For the purpose, they were developed to the learning program with concept cartoon and the instrument for the science attitude which has four categories; attitudes about science, attitudes about science subject, learning motives for science subject, and scientific attitudes. The research method was designed to quasi-experimental design. The concept cartoon was provided to the experimental group during nine lessons. Before and after the lessons in all two groups, the pre-post tests with the instrument were performed. The results from twice t-tests were shown that the domain of learning motives for science subject was only improved. From these, it were indicated that the concept cartoon was not effective all areas in science attitude, therefore the use in science lessons need to be restricted within narrow purpose.

• Journal of Digital Contents Society
• /
• v.19 no.7
• /
• pp.1423-1431
• /
• 2018

IoT(Internet of Things) is applied to technologies such as agriculture and dairy farming, making it possible to cultivate crops easily and easily in cities.In particular, IoT technology that intelligently judge and control the growth environment of cultivated crops in the agricultural field is being developed. In this paper, we propose a method of predicting the growth environment of plants by learning the moisture supply cycle of plants using the intelligent object internet. The proposed system finds the moisture level of the soil moisture by mapping learning and finds the rules that require moisture supply based on the measured moisture level. Based on these rules, we predicted the moisture supply cycle and output it using media, so that it is convenient for users to use. In addition, in order to reduce the error of the value measured by the sensor, the information of each plant is exchanged with each other, so that the accuracy of the prediction is improved while compensating the value when there is an error. In order to evaluate the performance of the growth environment prediction system, the experiment was conducted in summer and winter and it was verified that the accuracy was high.

• Journal of the Korean School Mathematics Society
• /
• v.24 no.1
• /
• pp.59-81
• /
• 2021

This study aimed to investigate how the learners' mathematics learning processes change with repeatedly reading mathematical text. As a way to teach and learn mathematics, we also wanted to examine the effect of repeated reading and to explore the implications for a more efficient teaching and learning strategy. To help us with this study, we mainly used eye tracking and heart rate (HR) measurement. There were four cycles in a cycle of repeated reading, and the number of repeated readings for all cycles was fixed to three times. Eight prospective mathematics teachers in the Department of Mathematics Education of a National University in South Korea participated. Data were analyzed in five aspects: (1) the total reading time per round, the total reading time per slide; (2) the change trends of total reading time per round and slide; (3) the order of slides read; (4) the change trends of HR per round. We found that most participants read in a similar pattern in the first reading, but the second and third reading patterns appeared more diverse for each learner. Also, the first reading required the most time regardless of the repeat cycle, and the time it took to repeatedly read afterward varied depending on the individual. Based on the findings of this study, the most primary conclusion is that self-directed mathematics learning by using repeated reading is effective regardless of cycle. In addition, we suggested four strategies to improve the efficiency of this teaching and learning method.

• Journal of the Korean earth science society
• /
• v.42 no.1
• /
• pp.88-101
• /
• 2021

The purpose of this study was to qualitatively examine the meaning of students' drawings in outdoor classes and modeling-based learning cycles. Ten students were observed in a gifted education center in Seoul. Under the theme of the Hantan River, three outdoor classes and three modeling activities were conducted. Data were collected to document all student activities during field trips and classroom modeling activities using simultaneous video and audio recording and observation notes made by the researcher and students. Please note it is unclear what this citation refers to. If it is the previous sentence it should be placed within that sentence's punctuation. Hatisaru (2020) Ddrawing typess were classified by modifying the representations in a learning context in geological field trips. We used deductive content analysis to describe the drawing characteristics, including students writing. The results suggest that students have symbolic images that consist of geologic concepts, visual images that describe topographical features, and affective images that express students' emotion domains. The characteristics were classified into explanation, generality, elaboration, evidence, coherence, and state-of-mind. The characteristics and drawing types are consecutive in the modeling-based learning cycle and reflect the students' positive attitude and cognitive scientific domain. Drawing is a useful tool for reflecting students' thoughts and opinions in both outdoor class and classroom modeling activities. This study provides implications for emphasizing the importance of drawing activities.