• The Journal of Korea Robotics Society
• /
• v.18 no.3
• /
• pp.293-298
• /
• 2023

The military is facing a continuous decrease in personnel, and in order to cope with potential accidents and challenges in operations, efforts are being made to reduce the direct involvement of personnel by utilizing the latest technologies. Recently, the use of various sensors related to Manned-Unmanned Teaming and artificial intelligence technologies has gained attention, emphasizing the need for flexible utilization methods. In this paper, we propose four dataset construction methods that can be used for effective training of robots that can be deployed in military operations, utilizing not only RGB image data but also data acquired from IR image sensors. Since there is no publicly available dataset that combines RGB and IR image data, we directly acquired the dataset within buildings. The input values were constructed by combining RGB and IR image sensor data, taking into account the field of view, resolution, and channel values of both sensors. We compared the proposed method with conventional RGB image data classification training using the same learning model. By employing the proposed image data fusion method, we observed improved stability in training loss and approximately 3% higher accuracy.

• Geophysics and Geophysical Exploration
• /
• v.24 no.3
• /
• pp.89-97
• /
• 2021

Fault detection in seismic data is well suited to the application of machine learning algorithms. Accordingly, various machine learning techniques are being developed. In recent studies, machine learning models, which utilize synthetic data, are the particular focus when training with deep learning. The use of synthetic training data has many advantages; Securing massive data for training becomes easy and generating exact fault labels is possible with the help of synthetic training data. To interpret real data with the model trained by synthetic data, the synthetic data used for training should be geologically realistic. In this study, we introduce a method to generate realistic synthetic seismic data. Initially, reflectivity models are generated to include realistic fault structures, and then, a one-way wave equation is applied to efficiently generate seismic stack sections. Next, a migration algorithm is used to remove diffraction artifacts and random noise is added to mimic actual field data. A convolutional neural network model based on the U-Net structure is used to verify the generated synthetic data set. From the results of the experiment, we confirm that realistic synthetic data effectively creates a deep learning model that can be applied to field data.

• Journal of Institute of Control, Robotics and Systems
• /
• v.4 no.4
• /
• pp.551-560
• /
• 1998

This paper presents a neural network model representing complex hydro-thermo-dynamic characteristics of a steam generator in nuclear power plants. The key modeling processes include training data gathering process, analysis of system dynamics and determining of the neural network structure, training process, and the final process for validation of the trained model. In this paper, we suggest a training data gathering method from an unstable steam generator so that the data sufficiently represent the dynamic characteristics of the plant over a wide operating range. In addition, we define the inputs and outputs of neural network model by analyzing the system dimension, relative degree, and inputs/outputs of the plant. Several types of neural networks are applied to the modeling and training process. The trained networks are verified by using a class of test data, and their performances are discussed.

• Journal of Korean Academy of Nursing
• /
• v.29 no.6
• /
• pp.1244-1253
• /
• 1999

The purpose of this study was to explore and describe the experience of Dan-Jeon breathing training and of Qi as a essential substance in forming human body. The sample consists of 7 participants who are Dan-Jeon Breathing training in a Training center, Pusan, Korea. They were asked open-ended questions in order for them to talk about their experiences. With permission of the subjects, the interviews were recorded and transcribed. The summarized results of this research are following. 1. The purpose of Dan-Jeon Breathing The interview data was organized by themes into 4 categories : hope for health recovery, a concern about Dan-Jeon Breathing, seeking meaning of life, change of lifestyle 2. The experience of Qi during Dan-Jeon Breathing training The interview data was organized by themes into 3 categories : an autonomic movement of body, spiritual experience, conviction of existence of Qi. 3. The change after Dan-Jeon Breathing training. The interview data was organized by themes into 7 categories : physical health promotion, emotional relaxation, promoting brain function, positive attitude about life, love to others, investigation for self, improvement on Qi feeling..

• The Journal of Korea Robotics Society
• /
• v.14 no.2
• /
• pp.122-130
• /
• 2019

Collecting a rich but meaningful training data plays a key role in machine learning and deep learning researches for a self-driving vehicle. This paper introduces a detailed overview of existing open-source simulators which could be used for training self-driving vehicles. After reviewing the simulators, we propose a new effective approach to make a synthetic autonomous vehicle simulation platform suitable for learning and training artificial intelligence algorithms. Specially, we develop a synthetic simulator with various realistic situations and weather conditions which make the autonomous shuttle to learn more realistic situations and handle some unexpected events. The virtual environment is the mimics of the activity of a genuine shuttle vehicle on a physical world. Instead of doing the whole experiment of training in the real physical world, scenarios in 3D virtual worlds are made to calculate the parameters and training the model. From the simulator, the user can obtain data for the various situation and utilize it for the training purpose. Flexible options are available to choose sensors, monitor the output and implement any autonomous driving algorithm. Finally, we verify the effectiveness of the developed simulator by implementing an end-to-end CNN algorithm for training a self-driving shuttle.

• Journal of Practical Engineering Education
• /
• v.15 no.2
• /
• pp.381-388
• /
• 2023

Through artificial neural network education using spreadsheets, non-major undergraduate students can understand the operation principle of artificial neural networks and develop their own artificial neural network software. Here, training of the operation principle of artificial neural networks starts with the generation of training data and the assignment of correct answer labels. Then, the output value calculated from the firing and activation function of the artificial neuron, the parameters of the input layer, hidden layer, and output layer is learned. Finally, learning the process of calculating the error between the correct label of each initially defined training data and the output value calculated by the artificial neural network, and learning the process of calculating the parameters of the input layer, hidden layer, and output layer that minimize the total sum of squared errors. Training on the operation principles of artificial neural networks using a spreadsheet was conducted for undergraduate non-major students. And image training data and basic artificial neural network development results were collected. In this paper, we analyzed the results of collecting two types of training data and the corresponding artificial neural network SW with small 12-pixel images, and presented methods and execution results of using the collected training data for Orange machine learning model learning and analysis tools.

• Korean Journal of Human Ecology
• /
• v.14 no.6
• /
• pp.1017-1025
• /
• 2005

The purpose of this paper was to investigate the relationship between customer orientation and sales training. Data were obtained from 297 apparel salespeople working at six department stores in Daegu. Statistics used for data analysis were frequency, factor analysis, correlation, and t-test. The respondents were classified into 3 groups; high, medium, and low customer-oriented groups based on the mean score of customer orientation, and the high and the low were compared in training contents and educational methods. Based on factor analysis, four factors were extracted from 27 items of training content. Two of four factors were significantly correlated with customer orientation. The regression analysis showed that customer service and duration of work had significant effects on customer orientation. Also, the results were found that there were significant differences between the high and the low customer-oriented group in training contents which salespeople want to have in the future. However, there were not significant differences between the two groups in educational methods.

• Health Policy and Management
• /
• v.9 no.3
• /
• pp.21-32
• /
• 1999

The purpose of this study is to assess individual decisive factors for distribution of medical specialists in Korea. A data set was constructed using several published data sources. including the Korean Medical Association's physician master file as a principal source for physician information. Linear logistic regression analysis was performed to assess the relationship between the location of private specialist clinic for practice with six variables related with individual characteristics: age. sex. location of postgraduate training hospital. location of medical school graduated, size of hospital for training, and specialty. Analysis showed that location of practice. classified into urban and rural areas, was significantly associated with the variables of sex. location of postgraduate training hospital. location of medical school. In addition, significant association was found between the location of practice which was categorized into "near-Seoul area" and others, and sex, location of postgraduate training hospital. and location of medical school. We could conclude that to improve area maldistribution of physicians locations of hospitals for training and medical schools have to have the highest priority in the policymaking.icymaking.

• Journal of Labour Economics
• /
• v.23 no.2
• /
• pp.107-126
• /
• 2000

In this study I am concerned with the impact of training for the unemployed on reemployment in Korea. The data is based on the survey that was conducted on those who participated in training programs in 1998 and those who did not. The matching criteria was the length of the spell of nonemployment that preceded entry to training programs. This data design allows to apply the quasi-experimental evaluation method. My estimation results indicate that the participation in training raises the hazard rate into reemployment, but training characteristics such as training contents, agencies do not affect the hazard rate significantly. This results imply that training participation increases reemployment possibility by preventing withdrawal of participants from the labor market, but training programs make little contribution to improving skills.

• Journal of Internet Computing and Services
• /
• v.18 no.4
• /
• pp.19-25
• /
• 2017

In this paper, we designed and implemented IoT sensors based badminton motion analysis and training system that can be readily used by badminton players with PC. Unlike the traditional badminton training system which uses signals of the flags by coach, the proposed electronic training system used IoT sensors to automatically detect and analysis the motions for badminton players. The proposed badminton motion analysis and training system has the advantage with low power, because it communicates with the program through BLE communication. The badminton motion analysis system automatically measures the training time according to the player's movement, so it is possible to collect objective result data with less errors than the conventional flag signal based method by coach. In this paper, training data of 5 athletes were collected and it provides the feedback function through the visualization of each section of the training results by the players which can enable the effective training. For the weakness section of each player, the coach and the player can selectively and repeatedly perform the training function with the proposed training system. Based on this, it is possible to perform the repeated training on weakness sections and they can improve the response speed for these sections. Continuous research is expected to be able to compare more various players' agility and physical fitness.