• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.05a
• /
• pp.453-456
• /
• 2016

Recently, IoT technologies attract much attenction in medical area. Previous medical IoT had focused mainly on chronological diseases or fitness for particular users. Contrarily, medical use of the IoT technologies is now extended for medical institutes and hospitals to care intensively in-house patients, which requires typically more strict and reliable data delivery and security, authentication and authorization. This study defines scenario of the medical IoT for the intensive care and proposes an architecture of the medical IoT services. We implement a testbed using commerical sensors and Arduino board together with a Web-based platform. Experiment results on the testbed show that our approach can be feasible for the medical system in terms of latency and accuracy in medical data delivery.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.33 no.2
• /
• pp.154-161
• /
• 2010

The conning display which is located in the ship bridge shows the various important information such as ship position, ship speed, track data, rate of turn, thruster rpm so on, and is one of the IBSs(Integrated Bridge Systems). In this study, the survey was conducted for ten officers to find the importance and using frequency of the information which were displayed in the conning display. The results showed that the information of drift speed, ship speed, wind direction and wind force, rate of turn, sea water depth, ship position, heading, thrust rpm, alarm, rudder command and angle got high scores and it meant that these information were very important and high frequency of use during the navigation. The optimized contents arrangement in conning display was suggested based on importance and using frequency of information. The experiment using eye-tracking system was conducted to compare the performance time and error rate of nine different scenarios for suggested arrangement display and three other existing displays. The results showed that the suggested arrangement was the best in performance time and error rate. The scenario concerning the direction and speed of wind showed faster performance time and lower error rate than other scenarios. The movement of subject's eye tended to search from the center and to avoid the comer, called 'the comer effect.' It is expected that the results of this study could help for the bridge staff to grasp the sailing information easily and to cope with the given situations promptly.

• Korean Journal of Soil Science and Fertilizer
• /
• v.49 no.3
• /
• pp.242-250
• /
• 2016

Veterinary antibiotics (VAs) has been used to treat animal disease and to increase animal weight as growth promoter. However, abused usage of VAs can cause production of antibiotic resistance genes (ARGs) in the environment and additionally, residual of VAs in soil can be transferred into crops. Therefore, main objective of this research was to examine bioaccumulation of VAs in sprouts (red cabbage, Brassica Olearacea L. var. Capitata f. rubra and red radish, Raphanus sativus) with hydroponic method. Total of 7 VAs in 2 different classes of VAs (tetracyclcines: tetracycline, oxytetracycline, chlortetracycline, sulfonamides: sulfamethoxazole, sulfamethazine, sulfamethiazole, macrolides: tylosin) were evaluated and experiment was conducted with solid phase extraction (SPE)/high performance liquid chromatography tandem mass spectrometry (HPLC/MS/MS). Initial spiked concentration of 7 VAs was $5mg\;L^{-1}$ and cultivation period was 8 days. Result showed that growth of sprouts was inhibited about 23-27% when VAs was introduced. Amount of bioaccumulated VAs was also differed depending on class of VAs. The highest amount of bioaccmulated VAs was tetracycline and sulfamethoxazole in each class with a concentration of 4.05, $7.73mg\;kg^{-1}$ respectively. Calculated transfer ratio of VAs into crops was also ranged 0.38-54.27%. Overall, bioaccumulation of VAs in crops can be varied depending on crop species and class of VAs. However, further research should be conducted to verify bioaccumulation of VAs in crops in the soil environment.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.3 s.35
• /
• pp.89-99
• /
• 2005

The testing time and cost of developed software can be reduced by automation of the whole or part of testing process. Since the testing automation tools to be used currently have been developed without their interoperability, test case information such as test procedures, test data, and expectation of test results generated at the stage of test execution cannot be shared and reused in other testing automation tools, In order to reduce testing time and cost, in this thesis, we have proposed a software testing automation framework which makes it possible to share and reuse the test case information generated in testing process. To prove the availability and effectiveness of proposed testing automation framework, three testing automation tools that are available in current market were experimented by the test scenario. As a result of experiment, the testing time and cost could be reduced by sharing and reusing the test case information in software testing automation framework.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.3
• /
• pp.209-218
• /
• 2013

Early researches in human action recognition have focused on tracking and classifying articulated body motions. Such methods required accurate segmentation of body parts, which is a sticky task, particularly under realistic imaging conditions. Recent trends of work have become popular towards the use of more and low-level appearance features such as spatio-temporal interest points. Given the great progress in pose estimation over the past few years, redefined views about pose-based approach are needed. This paper addresses the issues of whether it is sufficient to train a classifier only on low-level appearance features in appearance approach and proposes effective pose-based approach with pose estimation for emotional action recognition. In order for these questions to be solved, we compare the performance of pose-based, appearance-based and its combination-based features respectively with respect to scenario of various emotional action recognition. The experiment results show that pose-based features outperform low-level appearance-based approach of features, even when heavily spoiled by noise, suggesting that pose-based approach with pose estimation is beneficial for the emotional action recognition.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.8
• /
• pp.923-929
• /
• 2019

The stability of the propulsion system is crucial for the autonomous vessel. Multiple power generation and propulsion systems should be provided for the stability of the propulsion system. High power generation capacity is calculated for stability, resulting in economical decline due to low load operation. To solve this problem, we need to optimize the power system. In this paper, an OPMS for electric propulsion ship is constructed. The OPMS consists of a hybrid power generation system, an energy storage system, and a control load system. The power generation system consists of a dual fuel engine, the energy storage system is a battery, and the control load system consists of the propulsion load, continuous load, intermittent load, cargo part load and deck machine load. The power system was constructed by modeling the characteristics of each system. For the experiment, a scenario based on ship operation was prepared and the stability and economical efficiency were compared with existing electric propulsion ships.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.6
• /
• pp.1031-1041
• /
• 2020

Cloud computing technology plays an important role in the deep learning industry as deep learning services are deployed frequently on top of cloud infrastructures. In such cloud environment, virtualization technology provides logically independent and isolated computing space for each tenant. However, recent studies demonstrate that by leveraging vulnerabilities of virtualization techniques and shared processor architectures in the cloud system, various side-channels can be established between cloud tenants. In this paper, we propose a novel attack scenario that can steal internal information of deep learning models by exploiting the Meltdown vulnerability in a multi-tenant system environment. On the basis of our experiment, the proposed attack method could extract internal information of a TensorFlow deep-learning service with 92.875% accuracy and 1.325kB/s extraction speed.

• Journal of the Korea Society for Simulation
• /
• v.30 no.1
• /
• pp.139-152
• /
• 2021

Simulations enable virtually experiencing rare events as well as analytically analyzing such events. Defense modeling and simulation research and develops the virtual and the constructive simulations to support these utilizations. These virtual and constructive(VC) simulations can interoperate to simultaneously virtual combat experience as well as evaluations on tactics and intelligence of combat entities. Moreover, recently, for artificial intelligence researches, it is necessary to retrieve human behavior data to proceed the imitation learning and the inverse reinforcement learning. The presented work illustrates a case study of VC interoperations in the aircombat scenario, and the work analyze the collected human behavior data from the VC interoperations. Through this case study, we discuss how to build the VC simulation in the aircombat area and how to utilize the collected human behavior data.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.1
• /
• pp.551-563
• /
• 2022

This study aims to identify the effect on the user experiences when the embodied agent in a self-driving car interacts with emotional expressions by using 'interjection'. An experimental study was designed with two conditions: the inclusion of injections in the agent's conversation feedbacks (with interjections vs. without interjections) and the type of conversation (task-oriented conversation vs. social-oriented conversation). The online experiment was conducted with the four video clips of conversation scenario treatments and measured intimacy, likability, trust, social presence, perceived anthropomorphism, and future intention to use. The result showed that when the agent used interjection, the main effect on social presence was found in both conversation types. When the agent did not use interjection in the task-oriented conversation, trust and future intention to use were higher than when the agent talked with emotional expressions. In the context of the conversation with the AI agent in a self-driving car, we found only the effect of adding emotional expression by using interjection on the enhancing social presence, but no effect on the other user experience factors.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.4
• /
• pp.179-188
• /
• 2022

There have been many successful researches on a bearing fault diagnosis based on Deep Learning, but there is still a critical issue of the data distribution difference between training data and test data from their different working conditions causing performance degradation in applying those methods to the machines in the field. As a solution, a data adaptation method has been proposed and showed a good result, but each and every approach is strictly limited to a specific applying scenario or presupposition, which makes it still difficult to be used as a real-world application. Therefore, in this study, we have proposed a method that, using a data transformation with MFCCs and a simple CNN architecture, can perform a robust diagnosis on a target domain data without an additional learning or tuning on the model generated from a source domain data and conducted an experiment and analysis on the proposed method with the CWRU bearing dataset, which is one of the representative datasests for bearing fault diagnosis. The experimental results showed that our method achieved an equal performance to those of transfer learning based methods and a better performance by at least 15% compared to that of an input transformation based baseline method.