• Journal of Korean Society of Transportation
• /
• v.35 no.3
• /
• pp.220-233
• /
• 2017

Traffic volumes are fundamental data widely used in various traffic analysis, such as origin-and-destination establishment, total traveled kilometer distance calculation, congestion evaluation, and so on. The low number of links collecting the traffic-volume data in a large urban highway network has weakened the quality of the analyses in practice. This study proposes a method to estimate the traffic volume data on a highway link where no collection device is available by introducing a spatial statistic technique with (1) the traffic-volume data from TOPIS, and National Transport Information Center in the Ministry of Land, Infrastructure, and (2) the navigation data from private navigation. Two different component models were prepared for the interrupted and the uninterrupted flows respectively, due to their different traffic-flow characteristics: the piecewise constant function and the regression kriging. The comparison of the traffic volumes estimated by the proposed method against the ones counted in the field showed that the level of error includes 6.26% in MAPE and 5,410 in RMSE, and thus the prediction error is 20.3% in MAPE.

• Journal of Wetlands Research
• /
• v.15 no.3
• /
• pp.413-422
• /
• 2013

In recent, the rainfall is showing different properties in space and time but the ground rain gauge only can observe rainfall at a point. This means the ground rain gauge has the limitations in spatial and temporal resolutions to measure rainfall and so there is a need to utilize radar rainfall which can consider spatial distribution of rainfall This study tried to apply radar rainfall for runoff simulation on an urban drainage system. The study area is Guro-gu, Seoul and we divided study area into subbasins based on rain gauge network of AWS(Automatic Weather station). Then the radar rainfalls were adjusted using rainfall data of rain gauge stations the areal rainfalls were obtained. The runoffs were simulated by using XP-SWMM model in subbasins of an urban drainage system. As the results, the adjusted radar rainfalls were underestimated in the range of 60 to 95% of rain gauge rainfalls and so the simulated runoffs from the adjusted radar and gauge rainfalls also showed the differences. The runoff peak time from radar rainfall was occurred more fast than that from gauge rainfall.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.8
• /
• pp.3389-3404
• /
• 2012

Based on the case study methods, the study analyzes emergence and ensuing typology of global ebook platforms such as Google eBook, Amazon Kindle, iBooks Store. Global ebook platforms show adaptation process responding to rapidly changing digital technological envirment and it's fitness landscape. The critical elements in its emerging process are the distinct selection criteria, the degree of resource abundance and the search process based on open innovation. Based on these critical elements, the global platforms show speciation process, so called niche creation and are evolving into a variety of the typology based on the initial condition of key resource which makes the platform emerge and grow. Each global ebook platforms is evolving into open platform, hybrid platform, closed platform. Google eBook has openness and extensibility due to a variety of devices based on Android and a direct involvement of actors. Amazon Kindle has developed from a online bookstore and into the hybrid platform which have not only closed quality but also openness with ebook devices and mobile network. iBooks Store has developed into the closed platform through the agency model based on competitive hardwares and closed quality with iphone and ipad.

• Journal of KIISE:Software and Applications
• /
• v.26 no.11
• /
• pp.1261-1269
• /
• 1999

다층 퍼셉트론은 다양한 응용 분야에 성공적으로 적용되고 있는 대표적인 신경회로망 모델이다. 그러나 다층 퍼셉트론의 학습에서 나타나는 플라토에 기인한 느린 학습 속도와 지역 극소는 실제 응용문제에 적용함에 있어서 가장 큰 문제로 지적되어왔다. 이 문제를 해결하기 위해 여러 가지 다양한 학습알고리즘들이 개발되어 왔으나, 계산의 비효율성으로 인해 실제 문제에는 적용하기 힘든 예가 많은 등, 현재까지 만족할 만한 해결책은 제시되지 못하고 있다. 본 논문에서는 다층퍼셉트론의 베이시스 함수로 사용되는 시그모이드 함수를 보다 일반화된 형태로 정의하여 사용함으로써 학습에 있어서의 플라토를 완화하고, 지역극소에 빠지는 것을 줄이는 접근방법을 소개한다. 본 방법은 기존의 변형된 가중치 수정식을 사용한 학습 속도 향상의 방법들과는 다른 접근 방법을 택함으로써 기존의 방법들과 함께 사용하는 것이 가능하다는 특징을 갖고 있다. 제안하는 방법의 성능을 확인하기 위하여 간단한 패턴 인식 문제들에의 적용 실험 및 기존의 학습 속도 향상 방법을 함께 사용하여 시계열 예측 문제에 적용한 실험을 수행하였고, 그 결과로부터 제안안 방법의 효율성을 확인할 수 있었다. Abstract A multilayer perceptron is the most well-known neural network model which has been successfully applied to various fields of application. Its slow learning caused by plateau and local minima of gradient descent learning, however, have been pointed as the biggest problems in its practical use. To solve such a problem, a number of researches on learning algorithms have been conducted, but it can be said that none of satisfying solutions have been presented so far because the problems such as computational inefficiency have still been existed in these algorithms. In this paper, we propose a new learning approach to minimize the effect of plateau and reduce the possibility of getting trapped in local minima by generalizing the sigmoidal function which is used as the basis function of a multilayer perceptron. Adapting a new approach that differs from the conventional methods with revised updating equation, the proposed method can be used together with the existing methods to improve the learning performance. We conducted some experiments to test the proposed method on simple problems of pattern recognition and a problem of time series prediction, compared our results with the results of the existing methods, and confirmed that the proposed method is efficient enough to apply to the real problems.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.7 no.1
• /
• pp.62-74
• /
• 2004

Baekdu-daegan is the greatest mountain chain as well as the major ecological axis of the Korean Peninsula. In recent year, however, this area is faced with the various kinds of developmental urge. To cope adequately with these problems, this study was executed to prepare synthetic and systematic management with conservation-oriented strategy for Baekdu-daegan and to suggest spatially definite zoning for the managerial area. This study is to take into consideration the traditional concepts of stream and watershed as well as the actual disturbance on Baekdu-daegan area. The study area is selected with semi-mountainous type, from Namdeokyusan to Sosagogae. To propose the process for reasonably establishing the managerial boundary adjacent to the Ridges, the analysis was carried out that ArcGIS was mainly used for its analysis with digital maps, Landsat TM image and ArcGIS Hydro Model. Landsat TM image was classified by 5 land use types such as cultivated land, urban area, barren area, water body and forest. Based on these analyses results, the managerial boundaries as alternatives from the Ridges were produced by watershed expansion process, and used for tracing the changes of areal ratio of various land use types to the relevant watersheds to search out the adequate managerial boundary. The results show that watershed expansion process could be effective tool for establishing the managerial boundary, and eighth expanded watershed toward Muju-Gun(west) and fifth expanded watershed toward Geochang-Gun(east) might be included for the adequate managerial boundary of the case site.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.8
• /
• pp.1456-1464
• /
• 2007

The research of skin-tone detection has been conducting continuously to enlarge the importance in security, surveillance and administration of the information and 'Password Control System' for using face and skin recognition in airports, harbors and general companies. As well as tile rapid diffusion of the application range in image communications and an electron transaction using wide range of communication network, the importance of the accurate detection of skin color has been augmenting recently. In this paper, it will set up the boundaries of skin colors using the information of Cb and Cr in YCbCr color model of human skin color which is from hundreds compiled portrait images for each race, and suggest a efficient yet simple structure about the skin detection which has been followed by whether the comprehension of the boundaries of skin or not with adaptive skin-range set. With the possibility of the 1D Processes which does not use any memory, it is able to be applied to relatively small-sized hardware and system such as mobile apparatuses. To add the selective mode, it is not only available the improvement of tie skin detection, but also showing the correspondent results about previous face recognition technologies using complicated algorithm.

• Journal of Korea Water Resources Association
• /
• v.54 no.9
• /
• pp.667-680
• /
• 2021

We investigated changes in frost days and frost-free periods and to comparatively assess frost event prediction models developed using logistic regression (LR), random forest (RF), and long short-term memory (LSTM) networks. The meteorological variables for the model development were collected from the Suwon, Cheongju, and Gwangju stations for the period of 1973-2019 for spring (March - May) and fall (September - November). The developed models were then evaluated by Precision, Recall, and f-1 score and graphical evaluation methods such as AUC and reliability diagram. The results showed that significant decreases (significance level of 0.01) in the frequencies of frost days were at the three stations in both spring and fall. Overall, the evaluation metrics showed that the performance of RF was highest, while that of LSTM was lowest. Despite higher AUC values (above 0.9) were found at the three stations, reliability diagrams showed inconsistent reliability. A further study is suggested on the improvement of the predictability of both frost events and the first and last frost days by the frost event prediction models and reliability of the models. It would be beneficial to replicate this study at more stations in other regions.

• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.7
• /
• pp.191-198
• /
• 2021

Since the size of training dataset become large and the model is getting deeper to achieve high accuracy in deep learning, the deep neural network training requires a lot of computation and it takes too much time with a single node. Therefore, distributed deep learning is proposed to reduce the training time by distributing computation across multiple nodes. In this study, we propose hybrid allreduce strategy that considers the characteristics of each layer and communication and computational overlapping technique for synchronization of distributed deep learning. Since the convolution layer has fewer parameters than the fully-connected layer as well as it is located at the upper, only short overlapping time is allowed. Thus, butterfly allreduce is used to synchronize the convolution layer. On the other hand, fully-connecter layer is synchronized using ring all-reduce. The empirical experiment results on PyTorch with our proposed scheme shows that the proposed method reduced the training time by up to 33% compared to the baseline PyTorch.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.3
• /
• pp.629-643
• /
• 2019

In the era of the 4th industrial revolution, countries around the world are conducting projects to rapidly expand smart factory to secure competitiveness in manufacturing industries. However, unlike existing factories where the network environment was closed, smart factories can be vulnerable because internal and external objects are interconnected and various ICT technologies are used. And smart factories are likely to be the subject of cyber-attacks that are designed to cause monetary damage to certain targets because economic damage is so serious when an accident occurs. Therefore, it is necessary to study and apply security for smart factories, but there is no specific smart factory system architecture, so there is no establish for smart factory security requirements. In order to solve these problems, this paper derives the smart factory architecture that can extract and reflect the main characteristics of a smart factory based on the domestic and foreign reference model of smart factories. And this paper identifies the security threats based on the derived smart factory architecture and present the security requirements to cope with them for contributing to the improvement of the security of the smart factory.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.3
• /
• pp.491-501
• /
• 2019

In side-channel analysis, which exploit physical leakage from a cryptographic device, deep learning based attack has been significantly interested in recent years. However, most of the state-of-the-art methods have been focused on classifying side-channel information in a profiled scenario where attackers can obtain label of training data. In this paper, we propose a new method based on deep learning to improve non-profiling side-channel attack such as Differential Power Analysis and Correlation Power Analysis. The proposed method is a signal preprocessing technique that reduces the noise in a trace by modifying Auto-Encoder framework to the context of side-channel analysis. Previous work on Denoising Auto-Encoder was trained through randomly added noise by an attacker. In this paper, the proposed model trains Auto-Encoder through the noise from real data using the noise-reduced-label. Also, the proposed method permits to perform non-profiled attack by training only a single neural network. We validate the performance of the noise reduction of the proposed method on real traces collected from ChipWhisperer board. We demonstrate that the proposed method outperforms classic preprocessing methods such as Principal Component Analysis and Linear Discriminant Analysis.