• Smart Media Journal
• /
• v.8 no.2
• /
• pp.46-57
• /
• 2019

In recent research, deep learning-based methods have achieved state-of-the-art performance in various computer vision tasks. However, these methods are commonly supervised, and require huge amounts of annotated data to train. Acquisition of data demands an additional costly effort, particularly for the tasks where it becomes challenging to obtain large amounts of data considering the time constraints and the requirement of professional human diligence. In this paper, we present a data level synthetic sampling solution to learn from small and imbalanced data sets using Generative Adversarial Networks (GANs). The reason for using GANs are the challenges posed in various fields to manage with the small datasets and fluctuating amounts of samples per class. As a result, we present an approach that can improve learning with respect to data distributions, reducing the partiality introduced by class imbalance and hence shifting the classification decision boundary towards more accurate results. Our novel method is demonstrated on a small dataset of 2789 tomato plant disease images, highly corrupted with class imbalance in 9 disease categories. Moreover, we evaluate our results in terms of different metrics and compare the quality of these results for distinct classes.

• Journal of Digital Convergence
• /
• v.10 no.5
• /
• pp.27-35
• /
• 2012

The importance of smart learning (SL) has been emphasized in schools. Additionally, the significance of smart work (SW) in improving business performance has gained much attention among industries. From theoretical, technological, and environmental perspectives, SL and SW are somewhat similar. Therefore, a case study was performed to find a way to link SL and SW, and a linking model was proposed for this purpose. Because laptops are considered a pivotal element in the technological aspect of SL, graduate students' use of personal laptops in classes (Bring Your Own Laptop, BYOL) was investigated. The results showed that the students reacted positively to the idea of using personal laptops in class and that they expected to learn IT skills more effectively this way. They listed being able to study even after class and the easy accessibility of relevant data as the strengths of BYOL. However, they cited the heaviness of the laptops and occasional loss of focus during classes as the weaknesses of BYOL. Thus, this study showed the possibility of that students who experience SL can perform better in an SW situation. Therefore, if a policy is enacted that allows students to efficiently use laptops, a greater number of educational achievements will be attained on smart campuses and, subsequently, a greater number of smart workers will be produced.

• Journal of Digital Convergence
• /
• v.13 no.9
• /
• pp.243-250
• /
• 2015

Mobile learning content implementing is IT tourism convergence study that IT convergence IT and tourism. Learning to increase the effectiveness of mobile learning content for each learning module, It proposed u-smart tourist information 2.0 systems. Mobile learning content, implementation is u-smart tourist information 2.0 can use the system. Convergence/integration of design patterns and XML is so interesting to students. This is the maximum benefit which is taught classes for each learning module divided into learning the Design Pattern NCS. As a result, the learner. In particular, attendance has come out better the effect of learning and improved. Another advantage is tourism, information content information quality mobile learning content for and construct a tourist information content that you can do in real time. Also, mobile learning content, implementation in the next NCS expected to use a lot of help in learning. This study is the result of increased learning the analysis of the lessons learned. Implement mobile learning content gives fun and interesting to the learner to ten design process using the u-Smart Tourist Information class 2.0.

• The Korean Journal of Community Living Science
• /
• v.22 no.4
• /
• pp.655-668
• /
• 2011

Although consumer ethical behavior related with illegal copies of digital software has been considered to be an important issue, not many studies have attempted to examine the issue. Firstly, this study attemped to explain the moral judgment and intention to make illegal copies of smart phone applications for college students. Secondly, psychological factors such as moral intensity and perceived risk related to making illegal copies were tested to be significantly different in individual characteristics such as experience of ethical education and past experience of making illegal copies of software, sex, age and household income. Thirdly, the effect of related factors such as psychological factors and individual characteristics was estimated to significantly influence moral judgment and intention to make illegal copies. Two step method(using LIMDEP program) was applied to estimate the model as a structural equation model. According to the results of this study, magnitude of consequences, financial risk and performance risk were found to be significantly different in income groups(less than middle class vs more than middle class). Prosecution risk was found to be significantly different in gender groups(female vs male). In addition, social consensus, financial risk, performance risk and prosecution risk were found to be significantly different in ethical education groups(experience vs no experience). Furthermore, moral judgment for making illegal copies of smart phone applications was found to be significantly influenced by income, ethical education, magnitude of consequences, temporal immediacy and social consensus. And intention to make illegal copies of smart phone applications was found to be significantly influenced by moral judgment, age, financial risk, performance risk and prosecution risk.

• Korean Journal of Agricultural Science
• /
• v.47 no.4
• /
• pp.1135-1145
• /
• 2020

The purpose of this study was to compare and analyze the emissions of SOx and NH3 for a 78 kW class agricultural tractor during agricultural operations. A real-time monitoring system was constructed for measuring the load data. The field test was conducted during plow and rotary tillage. The working conditions were selected with the transmission gears in M3 Low and M2 High for the plow tillage and L3 High and L3 Low for the rotary tillage. The engine torque and fuel consumption were measured using controller area network (CAN) communication, and the emissions of SOx and NH3 were calculated based on the fuel consumption. As a result of the field tests, the engine torque was higher for the plow tillage than for the rotary tillage. As the gear stage was increased, the engine torque became higher. The emissions of SOx and NH3 were higher for the plow tillage than for the rotary tillage because the fuel consumption increased. Moreover, the emissions of SOx and NH3 tended to be more distributed for the rotary tillage than for the plow tillage. To develop an emission factor for agricultural machinery, it is important to measure reliable emission data during agricultural operations. In a future study, we will collect various emission data using a portable emission measurement system during agricultural operations.

• Korean Journal of Agricultural Science
• /
• v.48 no.4
• /
• pp.1039-1050
• /
• 2021

The purpose of this study is to measure load data for a 55 kW class agricultural tractor during a harvesting operation and to analyze the required power according to the working conditions. A field test was conducted at three different tractor speeds (1.2, 1.3, and 1.4 km·h^-1). A load measurement system was developed for the front axles, rear axles, and for power take-off (PTO). The torque and rotational speeds of the axles and PTO were measured during the field test and were calculated as the required power. The results showed that the total required power was in the range of 4.86 - 5.48 kW during the harvesting operation according to the tractor speed, and it was confirmed that this represents a ratio of 8.8 - 10.0% of the engine rated power. Also, it was confirmed that the required power of the axle and PTO increased as the tractor speed increased. In future studies, we plan to supplement the measurement system for a tractor to include a hydraulic system and perform a field test for harvesting various underground crops.

• Journal of Digital Contents Society
• /
• v.13 no.1
• /
• pp.67-74
• /
• 2012

Students with learning disabilities need special education programs. In the traditional class, those students may not be satisfied with their studies. Thus, it is important to provide individualized class for those students. Classes using smart devices may give one of the solutions for individualized class. Unlike the typical mathematical problems, written problems require students to use various cognitive strategies, mathematical reasoning, inference ability, and so on. In this sense, written problems are good tools to develop the logical minds for students with learning disabilities. In this paper, a WOE-based smart learning system is proposed to help those students develop learning abilities. The proposed system has the following characteristics. First, students can learn naturally problem-solving abilities by following the work-out examples given from experts. Second, the proposed system can invoke motivation and interests of students using attractive icons and guidance rules provided with smart phone. Third, the proposed system can provide self-directed study for those students. The proposed system is applied for some students with learning disabilities. The following results are obtained. First, the individualized study can be possible since the system can provide continuous feedbacks and level-differentiated classes. Second, students can increase written problem solving abilities with natural understanding of study contents from smart phone. Finally, satisfaction, study motivation, and self-concept of students are increased through their successful experience during study processes.

• Science of Emotion and Sensibility
• /
• v.25 no.1
• /
• pp.129-142
• /
• 2022

This study evaluates a program developed for a high school-university linked class. The program combines maker education and smart clothing technology, which has great potential. First, in the preparation stage, the dyeing design course incorporates the contents of previous studies on smart clothing and maker education. Second, in the development stage, a program was developed to make emotional eco-bags by applying the transfer dyeing technique using transfer paper for inkjet printers and smart clothing technology using conductivity thread and LED bulbs. Third, in the implementation stage, the class was offered to 17 high school students who want to major in clothing and textiles. Lastly, the class was evaluated. The program had a 4.95 satisfaction level as measures on a five-point scale. Furthermore, this paper suggested an advanced program with Lilypad Arduino. In conclusion, through this study, it was confirmed that the emotional eco-bag development program applied with conductive yarn and transfer dye could more easily produce smart clothing technology, thereby expanding the thinking of high school students regarding the clothing major.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.4
• /
• pp.141-152
• /
• 2013

The current web implementation largely falls into process-oriented and object-oriented methods. Both ways implement the programs to be used for simple design of external outfit and auxiliary function of hybrid webs and it might lead to cost problems arising from waste of time and man-power to develop with it. Therefore, managed structural system programming which can be applied under .NET framework is suggested in a bid to solve these problems. The structural system design method under .NET environment by applying CS concept prefers to adopting the optimized design configuration suited to proper interfaces of UI configuration, which offers the environment to maintain the inheritance relationship by taking form of the balanced class allotment by domain, and continue to keep the connectivity between UX and UI which leads to application of interface of UX objects as form of objects through connection with objects in class design process. This development method enables programs to be implemented focused on representation methods of UI and UX, and helps relieve much of the burden on development costs.

• Journal of Digital Convergence
• /
• v.20 no.2
• /
• pp.57-62
• /
• 2022

This study was conducted with the purpose of listening to the explanations of artworks in the non-face-to-face class and confirming the learner's feelings as a result of the class. The proposed system listens to the explanation of the artwork, inputs the learner's emotions with a dedicated key, and expresses the result in music. To this end, the direction of the non-face-to-face art appreciation class model using the emotion key was set, and based on this, a system for non-face-to-face art appreciation was constructed. The learner will use the 'smart device using the emotion key' proposed in this study to listen to the explanation of the artwork and to input the emotion for the question presented. Through the proposed system, learners can express their emotional state in online art classes, and instructors receive the results of class participation and use them in various ways for educational analysis.