• Journal of the Korea Computer Graphics Society
• /
• v.24 no.2
• /
• pp.1-9
• /
• 2018

In this paper, we propose a method to restore corrupted black and white facial images to color. Previous studies have shown that when coloring damaged black and white photographs, such as old ID photographs, the area around the damaged area is often incorrectly colored. To solve this problem, this paper proposes a method of restoring the damaged area of input photo first and then performing colorization based on the result. The proposed method consists of two steps: BEGAN (Boundary Equivalent Generative Adversarial Networks) model based restoration and CNN (Convolutional Neural Network) based coloring. Our method uses the BEGAN model, which enables a clearer and higher resolution image restoration than the existing methods using the DCGAN (Deep Convolutional Generative Adversarial Networks) model for image restoration, and performs colorization based on the restored black and white image. Finally, we confirmed that the experimental results of various types of facial images and masks can show realistic color restoration results in many cases compared with the previous studies.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.13 no.4
• /
• pp.125-139
• /
• 2017

In this paper, we developed a framework to detect and predict insider information leakage by collecting and restoring network traffic. For automated behavior analysis, many meta information and behavior information obtained using network traffic collection are used as machine learning features. By these features, we created and learned behavior model, network model and protocol-specific models. In addition, the ensemble model was developed by digitizing and summing the results of various models. We developed a function to present information leakage candidates and view meta information and behavior information from various perspectives using the visual analysis. This supports to rule-based threat detection and machine learning based threat detection. In the future, we plan to make an ensemble model that applies a regression model to the results of the models, and plan to develop a model with deep learning technology.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.17 no.6
• /
• pp.39-50
• /
• 2014

This study was conducted to establish inventory and propose conservation strategy of 'village wetlands' in Cheonan. As results, the village wetlands are defined as such places as palustrine wetland, village embankment, agricultural reservoir or small reservoirs located in or near the village and related to everyday life or farming. Firstly 791 provisional village wetlands were identified in Cheonan by using Arc-GIS 10.1, then 104 wetlands were defined as village wetlands and listed the inventory of Cheonan Village Wetlands after being validated through their area (greater than $1,000m^2$), satellite images, Korea Land Information System, land use map, land coverage map and field survey. Finally the 49 wetlands were selected for detailed surveying, and function assessment. As the result of the wetland function assessments, 11 wetlands were found to have 'high' wetland function (conservation) 30 wetlands were 'average' (enhancement) and 8 wetlands were 'low' (restoration or enhancement). Enhancing biodiversity and ecosystem services through ecological management of wetlands in Cheonan and connecting with an ecological network were proposed.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.5B
• /
• pp.889-898
• /
• 2000

In order to support various broadband multimedia servies in the future, we designed a well balanced WDM backbone network. In Korean network traffic environment, six regional centers are selected, link capacities between the regional centers are estimated from the PDI traffic model, and the overall network configuration is designed for the all-optical backbone network. Also, we designed a basic configuration to be able to protect minimum communication capability against link failure. A simulation study is carried out to verify the desired performance of the designed WDM backbone network. Simulation results show that performance of the backbone network is well balanced to support various communication services in Korea in the mid 2000s

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.547-551
• /
• 2009

In recent years, a film restoration has gained increasing attention by many researchers, to emergence of variety multimedia and to importance of video preservation. Blotch is the most frequent degradation in old film. This paper presents a blotch detection method using color and shape feature. The proposed method is two major modules: a SROD detector using impulsive feature and NN-based detector using shape feature. To assess the validity of the proposed method, the experiments have been performed on several old films.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.26 no.6
• /
• pp.47-52
• /
• 2012

DAS(Distribution Automation System) is equipped with several software applications such as service restoration, loss minimization, and protective relay coordination. The software applications of DAS are very sensitive to the amount of section load being carried by a particular section of distribution lines. Moreover, each software application requires a different parameter of the section load according to its purpose. Therefore, This paper proposes a new section load management method using real-time measurement data of the distribution lines. In order to provide accurate data to DAS applications, this method considers section loads in terms of the relationship of power versus time. In order to establish that the proposed method is feasible, a performance-testing simulator was developed, and case studies were conducted for a modified real distribution network.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.49 no.1
• /
• pp.1-5
• /
• 2000

This paper presents a new algorithm to restore the de-energized loads in electric power distribution systems for restoration by utilizing object-based structure and heuristic rules without branch and feeder overloading. In order to cope with a complex network configuration and frequent breaker switching operations in distribution systems, the component of power system is described as objects and the information of system configuration is constructed by connecting objects each other. Such a strategy is developed mainly based on the object search algorithm to increase computation efficiency. As a result, the proposed system has been implemented to efficiently deal with large distribution systems by reducing computational burden remarkably compared with the conventional methods.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.65 no.1
• /
• pp.18-24
• /
• 2016

In order to service restoration and enhance power system reliability, a number of impedance based fault location algorithms have been developed for fault locating in a transmission line. This paper presents an advanced impedance-based fault location algorithms in a two-ended sources transmission line to reduce the DC offset error effects. This fault location algorithm uses of the GPS time synchronized voltage and current signals from the local and remote terminal. The algorithm uses an advanced DC offset removal filter. A series of test results using ATPdraw simulation data show the performance effectiveness of the proposed algorithm. The proposed algorithm is valid for a two-end sources transmission network.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.10b
• /
• pp.1107-1110
• /
• 2000

망 복구에 있어서 주요한 과제는 복구의 대상으로 하고 있는 작업 경로(working path)에 대한 복구경로(restoration path) 흑은 대체 경로(backup path)를 설정하는 것이다. 대체 경로를 설정할 때 작업경로의 대역을 보장할 수 있는 대체 경로를 설정해야 한다. 따라서 이 과제는 대역의 보장을 고려하여 경로를 설정하는 QoS 라우팅의 해결책이 요구된다. 본 논문에서는 동일 링크의 작업 경로가 아닐 경우 대체 경로의 대역을 공유함으로써 최대한 대역의 소비를 피하면서, 작업 경로의 대역을 보장하는 대체 경로를 설정하는 disjoint 경로를 계산하는 알고리즘을 제시하고 시뮬레이션을 통해서 대체 경로로 인한 대역의 소비(bandwidth consumption) 효과를 분석하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2016.01a
• /
• pp.65-66
• /
• 2016

최근 과학의 발달로 사물에 대한 다양한 센서와 통신 기능을 부여하여 사람 대 사람 또는 사물 대 사물 서로 교류하는 개념인 IoT(Internet of Things)가 화제이다. IoT를 구성하는 요소 중에 하나인 Wireless Sensor Networks(WSN)에 대한 연구를 한다. WSN은 전쟁터의 정찰과 같은 가혹한 환경에 적용되며 이를 구성하는 센서는 제한된 자원을 가지고 있기 때문에 고장에 취약한 특성이 있다. 따라서 상호 통신이 외부 요인에 의해 단절된 네트워크의 연결 복구에 대한 연구가 지속적으로 이루어지고 있다. 본 논문에서는 이러한 WSN 네트워크의 연결 복구 방식, 연결 복구기법 및 알고리즘의 연구 동향에 대해서 서술한다.