• The KIPS Transactions:PartC
• /
• v.15C no.2
• /
• pp.93-102
• /
• 2008

The network weather map and bandwidth time-series graph are popularly used to understand the current and past traffic condition of NSP, ISP, and enterprise networks. These systems collect traffic performance data from a SNMP agent running on the network devices such as routers and switches, store the gathered information into a DB, and display the network performance status in the form of a time-series graph or a network weather map using Web user interface. Most of current enterprise networks are constructed in the form of a hierarchical tree-like structure with multi-Gbps Ethernet links, which is quietly different from the national or world-wide backbone network structure. This paper focuses on the network weather map for current enterprise network. We start with the considering points in developing a network weather map system suitable for enterprise network. Based on these considerings, this paper proposes the best way of using SNMP in constructing a network weather map system. To prove our idea, we designed and developed a network weather map system for our campus network, which is also described in detail.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.12
• /
• pp.113-121
• /
• 2013

In this paper, we design and implement a network weather map system, which provides a macroscopic view on the whole network topology as well as the network link status and utilization. The proposed system also provides distributed NetFlow-based database facility and Web-based query interface, through which network operators can check the detailed network router or link status as well as submit predefined queries to easily find out and locate heavy hitters and/or their usage. We believe that our develop system will be a useful tool for small-to-mid-scale ISPs or network operators, in managing their own networks in a cost-effective way.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.3
• /
• pp.203-211
• /
• 2006

Today, Internet covers a world wide range and most appliances of our life are linked to network from enterprise server to household electric appliance. Therefore, the importances of administrable framework that can grasp network state by real-time is increasing day by day. Our objective in this paper is to describe a network weather report framework that monitors network traffic and performance state to report a network situation including traffic status in real-time. We also describe a mobile agent architecture that collects state information in each network segment. The framework could inform a network manager of the network situation. Through the framework. network manager accumulates network data and increases network operating efficiency.

• Journal of Climate Change Research
• /
• v.6 no.1
• /
• pp.55-60
• /
• 2015

Agro-Meteorological Information Service (AMIS) network has been established since 2001 by Rural Development Administration (RDA) in Korea, and has provided access to current and historical weather data with useful information for agricultural activities. AMIS network includes 158 automated weather stations located mostly in farm region, with planning to increase by 200 stations until 2017. Agrometeorological information is disseminated via the web site (http://weather.rda.go.kr) to growers, researchers, and extension service officials. Our services will give enhanced information from observation data (temperature, precipitation, etc.) to application information, such as drought index, agro-climatic map, and early warning service. AMIS network of RDA will help the implementation of an early warning service for weather risk management.

• Journal of Internet Computing and Services
• /
• v.19 no.5
• /
• pp.43-54
• /
• 2018

Images taken in haze weather are characteristic of low contrast and poor visibility. The process of reconstructing clear-weather image from a hazy image is called dehazing. The main challenge of image dehazing is to estimate the transmission map or depth map for an input hazy image. In this paper, we propose a single image dehazing method by utilizing the Generative Adversarial Network(GAN) for accurate depth map estimation. The proposed GAN model is trained to learn a nonlinear mapping between the input hazy image and corresponding depth map. With the trained model, first the depth map of the input hazy image is estimated and used to compute the transmission map. Then a guided filter is utilized to preserve the important edge information of the hazy image, thus obtaining a refined transmission map. Finally, the haze-free image is recovered via atmospheric scattering model. Although the proposed GAN model is trained on synthetic indoor images, it can be applied to real hazy images. The experimental results demonstrate that the proposed method achieves superior dehazing results against the state-of-the-art algorithms on both the real hazy images and the synthetic hazy images, in terms of quantitative performance and visual performance.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.123-123
• /
• 2020

The present study is aimed to correcting radar-based mean areal precipitation forecasts to improve urban flood predictions and uncertainty analysis of water levels contributed at each stage in the process. For this reason, a long short-term memory (LSTM) network is used to reproduce three-hour mean areal precipitation (MAP) forecasts from the quantitative precipitation forecasts (QPFs) of the McGill Algorithm for Precipitation nowcasting by Lagrangian Extrapolation (MAPLE). The Gangnam urban catchment located in Seoul, South Korea, was selected as a case study for the purpose. A database was established based on 24 heavy rainfall events, 22 grid points from the MAPLE system and the observed MAP values estimated from five ground rain gauges of KMA Automatic Weather System. The corrected MAP forecasts were input into the developed coupled 1D/2D model to predict water levels and relevant inundation areas. The results indicate the viability of the proposed framework for generating three-hour MAP forecasts and urban flooding predictions. For the analysis uncertainty contributions of the source related to the process, the Bayesian Markov Chain Monte Carlo (MCMC) using delayed rejection and adaptive metropolis algorithm is applied. For this purpose, the uncertainty contributions of the stages such as QPE input, QPF MAP source LSTM-corrected source, and MAP input and the coupled model is discussed.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.41 no.6
• /
• pp.33-43
• /
• 1999

Drought is a serious diaster in agriculutre, especially to upland crops. Hence, the Agricultural Drought Analysis Model (ADAM) that is integratable with GIS was applied to analyae agriculture drought in upland. ADAM is composed of two sub-models , one is a Soil Water Balance Model (SWBM) and the other is a Drougth Analysis Model (DAM) that is based on the Runs theory. The ADAM needs weather data, rainfall data and soil physical characteristics data as input and calculates daily soil moisture contents. GIS was integrated to the ADAM for the calculation of regional soil moisture using digitized landuse map, detaile dsoil map, thiessen network and district boundary . For the agriculutral drought analysis, the ADAM adapt the Runs theory for analyzing drought duration, severity and magnitude . Log-Pearson Type-III probability distribution function and Kolmogorov-Smirnov test were used to test the fitness of good of the model. The integration of ADAM with GIS was successfully implemented and would be operated effectively for the regional drought analysis.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.10
• /
• pp.455-464
• /
• 2015

As the number of overseas travelers with smartphones increases, there is a growing interest in smartphone applications, which can assist traveling. Typically, travelers need applications to obtain basic information, such as weather, map, currency, etc. However, existing smartphone applications are not suitable to do so because they require network connection that is expensive and unstable in overseas. For example, one of the most frequently used smartphone map application requires a network connection, and much battery to download images. Since travelers spend most of their time outside, there is no chance to charge the battery. In this paper, we propose a study on implementation of a smartphone application for overseas travelers, called Travel Manager, which aims to reduce usages of network connection and battery. Travel Manager first checks whether smartphone is connected to the network, and then synchronizes the travel information. It also automatically calculates traveling expenses by considering currency rate. That is, the proposed smartphone application can be used regardless of the network connection and minimizes the battery usage.

• Journal of Korea Multimedia Society
• /
• v.19 no.2
• /
• pp.428-435
• /
• 2016

Hundreds of millions of new posts and information are being uploaded and propagated everyday on Online Social Networks(OSN) like Twitter, Facebook, or Instagram. This paper proposes and implements a GPS-location based SNS data mapping, analysis, and visualization system, called Smart SNS Map, which collects SNS data from Twitter and Instagram using hundreds of PlanetLab nodes distributed across the globe. Like no other previous systems, our system uniquely supports a variety of functions, including GPS-location based mapping of collected tweets and Instagram photos, keyword-based tweet or photo searching, real-time heat-map visualization of tweets and instagram photos, sentiment analysis, word cloud visualization, etc. Overall, a system like this, admittedly still in a prototype phase though, is expected to serve a role as a sort of social weather station sooner or later, which will help people understand what are happening around the SNS users, systems, society, and how they feel about them, as well as how they change over time and/or space.

• Journal of Korea Multimedia Society
• /
• v.22 no.9
• /
• 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.