• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.10
• /
• pp.179-186
• /
• 2019

Based on the characteristics of the weapon system, a government-funded research institute or a defense industry company is selected as the R&D supervision institution. On the other hand, research for the selection of a R&D supervision institution has not been conducted actively. This paper proposes a methodology for selecting R&D supervision institutions, such as procedure and indices. First, candidates of the index were obtained using data investigation and consulting, and five indices were deduced using Delphi. The weight of the indices was set using AHP. The high element consisted of 'Technical elements' and 'Business element'. The low element of 'Technical elements' consisted of 'Possession and readiness of critical technology' and 'Experience of similar R&D'. The low element of 'Business element' consisted of 'Base circumstance of the project', 'Risk management', and 'Will for the project'. The total weights of the indices were 'Possession and readiness of critical technology' 0.405, 'Experience of similar R&D' 0.297, 'Base circumstance of the project' 0.124, 'Risk management' 0.127, and 'Will for the project' 0.047. The indices were applied to the 00 weapon system and the result was deduced.

• Journal of Internet Computing and Services
• /
• v.22 no.1
• /
• pp.23-30
• /
• 2021

This paper proposes a noise-tolerant image classification system using multiple autoencoders. The development of deep learning technology has dramatically improved the performance of image classifiers. However, if the images are contaminated by noise, the performance degrades rapidly. Noise added to the image is inevitably generated in the process of obtaining and transmitting the image. Therefore, in order to use the classifier in a real environment, we have to deal with the noise. On the other hand, the autoencoder is an artificial neural network model that is trained to have similar input and output values. If the input data is similar to the training data, the error between the input data and output data of the autoencoder will be small. However, if the input data is not similar to the training data, the error will be large. The proposed system uses the relationship between the input data and the output data of the autoencoder, and it has two phases to classify the images. In the first phase, the classes with the highest likelihood of classification are selected and subject to the procedure again in the second phase. For the performance analysis of the proposed system, classification accuracy was tested on a Gaussian noise-contaminated MNIST dataset. As a result of the experiment, it was confirmed that the proposed system in the noisy environment has higher accuracy than the CNN-based classification technique.

• Journal of Korea Multimedia Society
• /
• v.15 no.1
• /
• pp.148-165
• /
• 2012

Applying maintenance simulation using virtual mockup in product design phase enables maintainability test before prototyping physical products, and it is expected to reduce product development costs. The simulation results can be reused as contents of service manuals and RAM (Reliability, Availability, Maintainability) analysis data. Maintenance simulation should provide realistic representation of physical property of virtual product, assembly relation between parts and manipulation process to verify feasibility of product design. The simulation system should be extended to collaborative virtual environment to perform collaborative maintenance procedures. In this paper, the three layered system architecture and the physics based collaborative interaction technique are proposed to extend current maintenance simulation into collaborative virtual environment. The proposed system was implemented as ViMMS (Virtual Mockup Maintenance Simulation system), and compared with case study results of VADE (Virtual Assembly Design Environment). As a result, the ViMMS encompassed broader range of maintenance tasks by using physics based collaborative interaction technique.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.3C
• /
• pp.125-132
• /
• 2005

This paper describes a computer vision system based on active IR illumination for real-time gaze discrimination system. Unlike most of the existing gaze discrimination techniques, which often require assuming a static head to work well and require a cumbersome calibration process for each person, our gaze discrimination system can perform robust and accurate gaze estimation without calibration and under rather significant head movement. This is made possible by a new gaze calibration procedure that identifies the mapping from pupil parameters to screen coordinates using generalized regression neural networks (GRNNs). With GRNNs, the mapping does not have to be an analytical function and head movement is explicitly accounted for by the gaze mapping function. Futhermore, the mapping function can generalize to other individuals not used in the training. To further improve the gaze estimation accuracy, we employ a reclassification scheme that deals with the classes that tend to be misclassified. This leads to a 10% improvement in classification error. The angular gaze accuracy is about 5°horizontally and 8°vertically. The effectiveness of our gaze tracker is demonstrated by experiments that involve gaze-contingent interactive graphic display.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.1
• /
• pp.81-94
• /
• 1995

By intergrating technologies for protable computers, radio communications, and computer networking, the Land Mobile Computer Network(LMCN) is supposed to overcome the time and space limitation of present computer communication network. However, because the MHs(mobile hosts) do move any time, the network connectivity is often changed causing temporarily duplicated lost, or out-of-sequenced message delivery as well as frequent communications for maintainning the network connectivity. Previous works solved the problems through message flooding, used also in multicasting, resulting in high communication cost(i.e., number of communication messages). In this paper, new protocols for efficient mobility management and multicasting are proposed. The protocols define location update, handover, and multicasting procedure of a MH over a hierarchical LMCN architecture. The protocol specification is presented, and it is shown that the communication cost of the new protocols is superior to the existing ones in terms of the communication cost.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.3 no.1
• /
• pp.38-46
• /
• 2010

This paper describes a computer vision system based on active IR illumination for real-time gaze discrimination system. Unlike most of the existing gaze discrimination techniques, which often require assuming a static head to work well and require a cumbersome calibration process for each person, our gaze discrimination system can perform robust and accurate gaze estimation without calibration and under rather significant head movement. This is made possible by a new gaze calibration procedure that identifies the mapping from pupil parameters to screen coordinates using generalized regression neural networks (GRNNs). With GRNNs, the mapping does not have to be an analytical function and head movement is explicitly accounted for by the gaze mapping function. Furthermore, the mapping function can generalize to other individuals not used in the training. To further improve the gaze estimation accuracy, we employ a reclassification scheme that deals with the classes that tend to be misclassified. This leads to a 10% improvement in classification error. The angular gaze accuracy is about $5^{\circ}$horizontally and $8^{\circ}$vertically. The effectiveness of our gaze tracker is demonstrated by experiments that involve gaze-contingent interactive graphic display.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.24 no.4
• /
• pp.50-56
• /
• 2010

Supervisory Control and Data Acquisition Systems (SCADAs) is real-time monitor and control systems. SCADA systems are used to monitor or control chemical and transportation processes, in municipal water supply systems, electric power generation, transmission and distribution, gas and oil pipelines, and other distributed processes. SCADA refers to a large-scale distributed system. The supervisory control system is placed on top of a real time control system to control external processes. Emerging security technologies and security devices are decreasing the vulnerability of the power system against cyber threats. Dealing with these threats and analyzing vulnerabilities is an important task for equipment such as RTU, IED and FEP. To reduce such risks, we develop such a SCADA testbed. This paper presents the development of a testbed designed to assess the vulnerabilities SCADA networks(including serial communication).

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.9 no.2
• /
• pp.322-329
• /
• 2005

This paper describes a computer vision system based on active IR illumination for real-time gaze discrimination system. Unlike most of the existing gaze discrimination techniques, which often require assuming a static head to work well and require a cumbersome calibration process for each person, our gaze discrimination system can perform robust and accurate gaze estimation without calibration and under rather significant head movement. This is made possible by a new gaze calibration procedure that identifies the mapping from pupil parameters to screen coordinates using generalized regression neural networks (GRNNS). With GRNNS, the mapping does not have to be an analytical function and head movement is explicitly accounted for by the gaze mapping function. furthermore, the mapping function can generalize to other individuals not used in the training. To further improve the gaze estimation accuracy, we employ a reclassification scheme that deals with the classes that tend to be misclassified. This leads to a 10$\%$ improvement in classification error. The angular gaze accuracy is about $5^{circ}$horizontally and $8^{circ}$vertically. The effectiveness of our gaze tracker is demonstrated by experiments that involve gaze-contingent interactive graphic display.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.1
• /
• pp.61-71
• /
• 2017

ARP Spoofing is a basic and core hacking technology for almost all sniffing. It makes change the flow of packets by faking the 2nd layer MAC address. In this paper we suggested an efficient hacking technology for sniffing remote servers in the switched network environment. The suggested 'Faked ARP Reply Unicast Spoofing' makes the bidirectional packets sniffing possible between the client and server, and it makes simplify the procedures for ARP sniffing and hacking program. In this paper we researched the network hacking and implementation technologies based on the suggested ARP spoofing. And we researched various types of servers hacking such as Root ID and PW of Telnet/FTP server, Root ID and PW of MySQL DB server, ID and PW of Web Portal Server, and account information and transaction history of Web Banking Server. And also we researched the implementation techniques of core hacking programs for the ARP Spoofing.

• The KIPS Transactions:PartC
• /
• v.14C no.1 s.111
• /
• pp.65-72
• /
• 2007

The existing RBAC models are not sufficient for managing delegations or separation of roles. Researches have been done on RBDM(Role Based Delegation Model) that deal with delegating role or permission to other users. In this paper, we divide the delegated roles into two groups: periodic and temporary delegation roles. When a role is delegated, a time period is assigned together, which is used to revoke the permission of delegated role automatically. In our model, the role of monotonic delegation by an original user can be revoked at any time in case of malicious use by the delegated user. The contribution of our model is that the malicious use of delegated role can be prohibited and security vulnerability in the role hierarchy due to role delegations can be alleviated. The proposed model, T RBDM(Time out Based RBDM) is analyzed and compared with the conventional models, such as RBDM0, RBDM1 and PBDM. Our model shows an advantage over other models in terms of security robustness.