• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.1176-1179
• /
• 2020

Vessel path prediction methods generally predict the latitude and longitude of a future location directly. However, in the case of direct prediction, errors could be large since the possible output range is too broad. In addition, error accumulation could occur since recurrent neural networks-based methods employ previous predicted data to forecast future data. In this paper, we propose a vessel path prediction method that does not directly predict the longitude and latitude. Instead, the proposed method predicts the acceleration of the vessel. Then the acceleration is employed to generate the velocity and direction, and the values decide the longitude and latitude of the future location. In the experiment, we show that the proposed method makes smaller errors than the direct prediction method, while both methods employ the same model.

• The Journal of the Korea Contents Association
• /
• v.21 no.1
• /
• pp.119-128
• /
• 2021

The most typical method to generate traffic information is installing vehicle detectors and collecting various traffic variables. The information collection accuracy of a vehicle detector affects the reliability of the generated traffic information. The most universal vehicle detector is an image detector. This study installed a magnetic detector in the same position as an image detector and evaluated the accuracy of traffic volume and speed data depending on a variety of environment. Based on the evaluation, more errors occurred as the image detector was placed farther from the camera, whereas more errors were found to occur during the night rather than the day. Although rainfall did not affect the collection of traffic volume, it negatively affected speed data collection. Therefore, an analysis of the camera's view angle and its optimization depending on the camera installation position and height are required to enhance the currently operated image detector performance. It is judged that a separate performance evaluation criterion should be prepared in a bad weather environment.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.5
• /
• pp.628-636
• /
• 2021

This paper contains the implementation and performance evaluation results of a system that collects environmental data at the time when the wild animal intrusion occurred at farms and then predicts future wild animal intrusions through a machine learning-based Vector Autoregression(VAR) model. To collect the data for intrusion prediction, an IoT-based hardware prototype was developed, which was installed on a small farm located near the school and simulated over a long period to generate intrusion events. The intrusion prediction service based on the implemented VAR model provides the date and time when intrusion is likely to occur over the next 30 days. In addition, the proposed system includes the function of providing real-time notifications to the farmers mobile device when wild animals intrusion occurs in the farm, and performance evaluation was conducted to confirm that the average response time was 7.89 seconds.

• The Journal of Korean Institute of Information Technology
• /
• v.17 no.12
• /
• pp.21-28
• /
• 2019

Side effects of drugs mean harmful and unintended effects resulting from drugs used to prevent, diagnose, or treat diseases. These side effects can lead to patients' death and are the main causes of drug developmental failures. Thus, various methods have been tried to identify side effects. These can be divided into biological and systems biology approaches. In this study, we use systems biology approach and focus on using various phenotypic information in addition to the chemical structure and target proteins. First, we collect datasets that are used in this study, and calculate similarities individually. Second, we generate a set of features using the similarities for each drug-side effect pair. Finally, we confirm the results by AUC(Area Under the ROC Curve), and showed the significance of this study through a comparison experiment.

• Design & Manufacturing
• /
• v.15 no.1
• /
• pp.48-56
• /
• 2021

The quality of products produced by injection molding process is greatly influenced by the process variables set on the injection molding machine during manufacturing. It is very difficult to predict the quality of injection molded product considering the stochastic nature of manufacturing process, because the process variables complexly affect the quality of the injection molded product. In the present study we predicted the quality of injection molded product using Artificial Neural Network (ANN) method specifically from Multiple Input Single Output (MISO) and Multiple Input Multiple Output (MIMO) perspectives. In order to train the ANN model a systematic plan was prepared based on a combination of orthogonal sampling and random sampling methods to represent various and robust patterns with small number of experiments. According to the plan the injection molding experiments were conducted to generate data that was separated into training, validation and test data groups to optimize the parameters of the ANN model and evaluate predicting performance of 4 structures (MISO1-2, MIMO1-2). Based on the predicting performance test, it was confirmed that as the number of output variables were decreased, the predicting performance was improved. The results indicated that it is effective to use single output model when we need to predict the quality of injection molded product with high accuracy.

• Journal of Advanced Navigation Technology
• /
• v.24 no.6
• /
• pp.533-539
• /
• 2020

Real-time service (RTS) provided by IGS provides correction for GNSS orbit and clock via internet, so it is widely used in fields that require real-time precise positioning. However, the RTS signal may be lost due to an unstable Internet environment. When signal disconnection occurs, signal prediction can be performed using polynomial models. However, the RTS changes rapidly after the GNSS navigation message issue of data (IOD) changes, so it is difficult to predict when signal loss occurs at that point. In this study, we proposed an algorithm to generate continuous RTS correction information by applying the difference in navigation trajectory according to IOD change. The use of this algorithm can improve the accuracy of RTS prediction at IOD changes. After performing optimization studies to improve RTS prediction performance, the predicted RTS trajectory information was applied to precision positioning (PPP). Compared to the conventional method, the position error is significantly reduced, and the error increase along with the signal loss interval increase is reduced.

• Journal of Broadcast Engineering
• /
• v.27 no.2
• /
• pp.174-184
• /
• 2022

3D human pose estimation models can be classified into a multi-view model and a single-view model. In general, the multi-view model shows superior pose estimation performance compared to the single-view model. In the case of the single-view model, the improvement of the 3D pose estimation performance requires a large amount of training data. However, it is not easy to obtain annotations for training 3D pose estimation models. To address this problem, we propose a method to generate pseudo ground-truths of multi-view human pose data from a multi-view model and exploit the resultant pseudo ground-truths to train a single-view model. In addition, we propose a multi-view consistency loss function that considers the consistency of poses estimated from multi-view images, showing that the proposed loss helps the effective training of single-view models. Experiments using Human3.6M and MPI-INF-3DHP datasets show that the proposed method is effective for training single-view 3D human pose estimation models.

• The Korean Journal of Franchise Management
• /
• v.13 no.4
• /
• pp.39-48
• /
• 2022

Purpose: COVID-19 has negatively influenced the financial performance of restaurant firms. Previous literature suggests that the franchising strategy effectively helps restaurant firms recover from difficult business conditions through various methods for expanding business size and enhancing business efficiency. According to risk-sharing theory, restaurant franchisors may minimize operational risks by sharing the risks with their franchisees. For instance, restaurant franchisors could generate more stable cash flow using franchise fees from their franchisees. However, research on the effect of franchise's risk reduction factor on business performance during pandemic is scarce. Thus, this study aims to examine the positive moderating effect of franchising between COVID-19 and restaurants' financial performance. Research design, data, and methodology: Panel data including financial information and franchising status of restaurant firms were collected for analysis. In order to control for unobserved firm-specific factors, generalized least squared estimation in fixed effects model was conducted. Huber-White robust standard errors were used to deal with heteroscedasticity issues. Results: It was found that COVID-19 pandemic has a negative effect on the restaurants' financial performance such as ROA (return on assets), ROE (return on equity), and PM (profit margins), which confirms the findings from existing literature. More importantly, results show that the degree of franchising has a positive moderating effect on the relationship between COVID-19 and financial performance of restaurant firms. This suggests that more active engagement in franchising may decrease negative impacts of COVID-19 on the restaurants' financial performance. Conclusions: The study supports existing literature related to risk-sharing theory, by confirming that pandemics, such as COVID-19, negatively affect financial performance of the restaurants. Furthermore, it was found that franchising strategy can help lessen negative impacts of pandemics on the firm performance. These findings can contribute to the franchise and restaurant management literature by suggesting the role of franchising in reducing business risks, thereby positively affecting financial performance. Moreover, this study offers business managers of franchisors and franchisees insights for utilizing franchising in restaurant risk management. Policymakers may also gain information on aiding restaurant firms during global crisis, such as COVID-19.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.6
• /
• pp.83-89
• /
• 2022

Internet of Things technology is rapidly developing due to the recent development of information and communication technology. IoT technology utilizes various sensors to generate unique data from each sensor, enabling diagnosis of system status. However, the equipment management system currently in effect is a post-preservation concept in which administrators must deal with the problem after the problem occurs, which could mean system reliability and availability problems due to system errors, and could result in economic losses due to negative productivity disruptions. Therefore, this study confirmed that edge controller control decision algorithms for more efficient operation of rectifiers in the factory by applying intelligent IoT (AIoT) technology and domain knowledge-based modeling for each sensor data collected based on this, outputting appropriate status messages for each scenario.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.8
• /
• pp.1136-1141
• /
• 2022

Researches of face recognition on masked faces have been increasingly important due to the COVID-19 pandemic. To realize a stable and practical recognition performance, large amount of facial image data should be acquired for the purpose of training. However, it is difficult for the researchers to obtain masked face images for each human subject. This paper proposes a novel method to synthesize a face image and a virtual mask pattern. In this method, a pair of masked face image and unmasked face image, that are from a single human subject, is fed into a convolutional autoencoder as training data. This allows learning the geometric relationship between face and mask. In the inference step, for a unseen face image, the learned convolutional autoencoder generates a synthetic face image with a mask pattern. The proposed method is able to rapidly generate realistic masked face images. Also, it could be practical when compared to methods which rely on facial feature point detection.