• 디지털콘텐츠학회 논문지
• /
• 제14권2호
• /
• pp.179-189
• /
• 2013

21세기가 되면서 컴퓨터 및 인터넷 등을 포함한 정보통신기술의 발전으로 다양한 기기에서 네트워크를 이용한 컴퓨팅 환경이 제공되어 지고 있다. 사이버 공간에서 사용자 인증방식은 텍스트 기반의 패스워드 인증방식을 사용하고 있다. 정보시스템의 비인가된 접근과 노출은 사용자, 공급자 모두에게 큰 피해를 입힐 수 있다. 이러한 인증방식은 기술적인 문제뿐만 아니라 사람들의 행동학적인 문제를 가지고 있다. 연구결과에 따르면 사용자들 대부분이 다양한 사이트를 이용하지만 사용하는 비밀번호개수는 그보다 훨씬 적었다. 또한 오랜 기간 한 가지 비밀번호를 사용하는 사용자가 많았으며 변경 시에도 기존의 비밀번호를 이용하여 최소한의 변경을 원하였다. 이에 정보보안의 차원에서 사람들의 전반적인 비밀번호 선택과 사용에 있어서 영향을 미치는 요인을 통계분석을 통해 알아보고자 한다.

• 한국전자거래학회지
• /
• 제5권2호
• /
• pp.27-47
• /
• 2000

Recently firms which have been using traditional VAN-EDI are adapting Web-EDI at very rapid pace. Based on the Internet, Web-EDI is easier to install, is cheaper and more accessible than VAN-EDI. Web-EDI establishes an electronic hierarchy as a type of Inter-Organization Information System. An electronic hierarchy connects two legally separate firms in an electronically mediated relationship. The primary reasons for establishing an electronic hierarchy is to improve the flow of materials and information between firms. This paper suggests two hypotheses that describe the possible impact of Web-EDI on firms in the Web-EDI network, and empirically tests them using data from a case study based on a discount store. The result is as follows. First, the more a firm utilizes information shared through Web-EDI, the higher the inventory turns. Second, a higher level of information sharing does not necessarily increase the inventory turns of Web-EDI adopters who are coerced to implement the electronic networks by the champion. These findings imply that the operational performance of Web-EDI adopters can be improved only when the firms intensively utilize the information shared through Web-EDI for their business processes.

• 대한교통학회지
• /
• 제10권3호
• /
• pp.7-20
• /
• 1992

The common cycle time for the linded signals is usually determined for the critical intersecion, just because the cpacity of a signalized intersection depends on the cycle time. This may not be optimal since the interactions between the flow and the spatial structure of the route or the area are disregarded in this case. It is common to separate the total delay incurred at signals into two parts, a deterministic or uniform delay and a stochastic or random delay. The deterministic delays and the stochastic delays on the artery particularly related to signal cycle time. For this purpose a microscopic simulation technique is used to evaluate deterministic delays, and a macroscopic simulation technique based on the principles of Markov chains is used to evaluate stochastic delays with over flow queue. As a result of investigating the relations between deterministic delays and cycle time in the various circumstances of spacing of signals and traffic volume. As for stochastic delays the resalts of comparisons of the macroscopic simulation and Newell's approximation with the microscopic simulation indicate that the former is valid for the degree of saturation less than 0.95 and the latter is for that above 0.95. Newell's argument that the total stochastic delay on an arterial is dominated by that at or caused by critical intersection is certified by the simulation experiments. The comprehensive analyses of the values of optimal cycle time with various conditions lead to a model. The cycle time determined by this model shows to be approximately 70% of that calculated by Webster's.

• 천문학회지
• /
• 제52권5호
• /
• pp.207-216
• /
• 2019

In this paper we introduce a software package, the Very long baseline interferometry Network SIMulator (VNSIM), which provides an integrated platform assisting radio astronomers to design Very Long Baseline Interferometry (VLBI) experiments and evaluate the network performance, with a user-friendly interface. Though VNSIM is primarily motivated by the East Asia VLBI Network, it can also be used for other VLBI networks and generic interferometers. The software package not only integrates the functionality of plotting (u, v) coverage, scheduling the observation, and displaying the dirty and CLEAN images, but also adds new features including sensitivity calculations for a given VLBI network. VNSIM provides flexible interactions on both command line and graphical user interface and offers friendly support for log reports and database management. Multi-processing acceleration is also supported, enabling users to handle large survey data. To facilitate future developments and updates, all simulation functions are encapsulated in separate Python modules, allowing independent invoking and testing. In order to verify the performance of VNSIM, we performed simulations and compared the results with other simulation tools, showing good agreement.

• 대한조선학회논문집
• /
• 제47권5호
• /
• pp.751-756
• /
• 2010

The rapid growth of IT technology has enabled ship automation systems to gain better functionality and safety with reduced costs and crew numbers. Nowadays, the remote maintenance services for the systems are required because a ship may be located in a very remote area. To provide the remote maintenance services, some issues such as how to collect the ship automation systems data, how to monitor a ship's data from onshore offices, how to get support from experts while sailing, and how to reduce the maintenance costs, should be addressed. In this paper, we propose a remote ship maintenance system for remote monitoring and diagnostics of ship automation systems, which is based on both a ship area network to integrate separate system networks and a ship-shore communication infrastructure to support a remote access using satellite communications. Finally, we present the function test to verify the applicability of the proposed system.

• 한국멀티미디어학회논문지
• /
• 제22권9호
• /
• pp.1000-1010
• /
• 2019

Image semantic segmentation and dehazing are key tasks in the computer vision. In recent years, researches in both tasks have achieved substantial improvements in performance with the development of Convolutional Neural Network (CNN). However, most of the previous works for semantic segmentation assume the images are captured in clear weather and show degraded performance under hazy images with low contrast and faded color. Meanwhile, dehazing aims to recover clear image given observed hazy image, which is an ill-posed problem and can be alleviated with additional information about the image. In this work, we propose a deep multi-task network for simultaneous semantic segmentation and dehazing. The proposed network takes single haze image as input and predicts dense semantic segmentation map and clear image. The visual information getting refined during the dehazing process can help the recognition task of semantic segmentation. On the other hand, semantic features obtained during the semantic segmentation process can provide cues for color priors for objects, which can help dehazing process. Experimental results demonstrate the effectiveness of the proposed multi-task approach, showing improved performance compared to the separate networks.

• 대한임베디드공학회논문지
• /
• 제16권1호
• /
• pp.9-15
• /
• 2021

Recently, it is increasingly necessary to run high-end neural network applications with huge computation overhead on top of resource-constrained embedded systems, such as wearable devices. While the huge computational overhead can be alleviated by distributed neural networks running on multiple separate devices, existing distributed neural network techniques suffer from a large traffic between the devices; thus are very vulnerable to communication failures. These drawbacks make the distributed neural network techniques inapplicable to wearable devices, which are connected with each other through unstable and low data rate communication medium like human body communication. Therefore, in this paper, we propose a distributed neural network partitioning technique that is resilient to communication failures. Furthermore, we show that the proposed technique also improves the inference accuracy even in case of no communication failure, thanks to the improved network partitioning. We verify through comparative experiments with a real-life neural network application that the proposed technique outperforms the existing state-of-the-art distributed neural network technique in terms of accuracy and resiliency to communication failures.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권7호
• /
• pp.2547-2567
• /
• 2021

It has been widely acknowledged that occlusion impairments adversely distress many face recognition algorithms' performance. Therefore, it is crucial to solving the problem of face image occlusion in face recognition. To solve the image occlusion problem in face recognition, this paper aims to automatically de-occlude the human face majority or discriminative regions to improve face recognition performance. To achieve this, we decompose the generative process into two key stages and employ a separate generative adversarial network (GAN)-based network in both stages. The first stage generates an initial coarse face image without an occlusion mask. The second stage refines the result from the first stage by forcing it closer to real face images or ground truth. To increase the performance and minimize the artifacts in the generated result, a new refine loss (e.g., reconstruction loss, perceptual loss, and adversarial loss) is used to determine all differences between the generated de-occluded face image and ground truth. Furthermore, we build occluded face images and corresponding occlusion-free face images dataset. We trained our model on this new dataset and later tested it on real-world face images. The experiment results (qualitative and quantitative) and the comparative study confirm the robustness and effectiveness of the proposed work in removing challenging occlusion masks with various structures, sizes, shapes, types, and positions.

• Nuclear Engineering and Technology
• /
• 제55권3호
• /
• pp.814-826
• /
• 2023

Sensor faults in nuclear power plant instrumentation have the potential to spread negative effects from wrong signals that can cause an accident misdiagnosis by plant operators. To detect sensor faults and make accurate accident diagnoses, prior studies have developed a supervised learning-based sensor fault detection model and an accident diagnosis model with faulty sensor isolation. Even though the developed neural network models demonstrated satisfactory performance, their diagnosis performance should be reevaluated considering real-time connection. When operating in real-time, the diagnosis model is expected to indiscriminately accept fault data before receiving delayed fault information transferred from the previous fault detection model. The uncertainty of neural networks can also have a significant impact following the sensor fault features. In the present work, a pilot study was conducted to connect two models and observe actual outcomes from a real-time application with an integrated system. While the initial results showed an overall successful diagnosis, some issues were observed. To recover the diagnosis performance degradations, additive logics were applied to minimize the diagnosis failures that were not observed in the previous validations of the separate models. The results of a case study were then analyzed in terms of the real-time diagnosis outputs that plant operators would actually face in an emergency situation.

• 전자공학회논문지CI
• /
• 제46권4호
• /
• pp.94-101
• /
• 2009

본 논문은 냉각형 적외선(infrared focal plane array; IRFPA) 영상시스템에서 하드웨어 뉴럴 네트워크를 이용한 실시간 고정패턴잡음 제어를 위해 고속 DSP & FPGA 기반의 H/W 설계 방법을 제안한다. 고정패턴잡음은 검출기의 불균일 보정처리후에도 관측영상의 온도분포 변화에 의해 발생한다. 이것은 열상 화질의 저하뿐만 아니라 다른 응용에도 문제되는 중요한 요소이다. 냉각형 적외선 영상시스템의 신호처리구조는 저온, 상온, 고온의 3개 테이블을 기준으로 이득(gain) 값과 편차(offset) 값을 연산한다. 제안된 방법은 3개 편차 테이블에서 각각 교차되는 영역을 세분화하여 가상의 테이블을 만들고, 입력 영상의 구분된 3개 영역에서 영상의 평균값으로 하드웨어 뉴럴 네트워크의 가중치 값을 조정하여 최적의 온도구간을 선정한다. 이와 같은 방법은 영상의 평균값으로부터 저온, 상온, 혹은 고온의 이득, 편차 테이블을 연산하고, 운용 중에 지속적으로 편차 보상을 적용하지 않아도 된다. 따라서 제안된 방법은 실시간 처리로 관측영상의 온도분포 변화에 의해 발생하는 고정패턴잡음을 제어하여 영상화질의 개선된 결과를 보였다.