• Journal of Korea Society of Industrial Information Systems
• /
• v.7 no.5
• /
• pp.190-194
• /
• 2002

The aim of development is system having convenient, cheap, reliable and better good measurements. The cozy environments in auto and architects is essentiable requirements. This study focuses on system developments of test equipment for better performance of noise protection in automobiles. Sound cards using makes easily to acquire the sound in each room, especially reduced the complexities of the circuits. The developed system have easily been measured and have characteristics of rapid mecanical response. The menu of rough scan method makes efficiently to test the shielding performance of auto rubber in whole frequency scales.

• Journal of Environmental Health Sciences
• /
• v.37 no.1
• /
• pp.1-11
• /
• 2011

Recent advances in Geographic Information Systems and spatial statistical and analytical methods, along with the availability of spatially referenced health and environmental data, have created unique opportunities to investigate spatial associations between environment exposures and health outcomes at multiple spatial scales and resolutions. However, the increased use of spatial data also faces challenges, one of which is to ensure certainty and accuracy of locational data that meets the needs of a study. This article critically reviews the use of spatially referenced data in epidemiologic studies, focusing on the issue of locational uncertainty generated from the process of geocoding health and environmental data. Primarily, major issues involving the use of spatially referenced data are addressed, including completeness and positional accuracy, potential source of bias and exposure misclassification, and implications for epidemiologic studies. The need for critical assessment and caution in designing and conducting spatial epidemiology studies is briefly discussed.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.4 no.2
• /
• pp.138-153
• /
• 2010

The interactions between people who do not know each other have been greatly increased with the on-going increase of people's cyberspace activities. In this situation, there exist potential risk factors such as the possibility of fraud, so we need a method to reduce or eliminate those risk factors. Concerning this necessity, rating systems are widely used, and many trust metrics calculated from rate values that people give to each other are proposed to help them make decisions. However, the trust metrics decrease the accuracy, and this is caused by the different rating scales and ranges of each person. So, we propose a fuzzy adjustment method to solve this problem. It is possible to catch the exact meaning of the trust value that each person selects through applying fuzzy sets, which improve the accuracy of the trust metric calculated from the trust values. We have applied our fuzzy adjustment method to the TidalTrust algorithm, a representative algorithm for calculating the local scalar trust metric, and we performed an experimental evaluation with four data sets and three evaluation methods.

• Journal of Korea Society of Industrial Information Systems
• /
• v.23 no.5
• /
• pp.33-41
• /
• 2018

Most important thing when designing and implementing a supply chain network is to consider various problems which may occur in real world situation. In this paper, we propose a supply chain network considering two problems (environmental factor and transportation types) under real world situation. CO2 emission amount as environmental factor is considered since it is usually generated from production and transportation processes. Normal delivery, direct delivery and direct shipment as transportation types are also considered since many customers ask various transportation types for delivery or shipment of their products under on-line or off-line purchase environment. The proposed supply chain network considering environmental factor and transportation types is represented in a mathematical formulation and implemented using hybrid genetic algorithm (HGA) approach. In numerical experiments, several scales of supply chain networks are presented and implemented using HGA approach. The performance of the HGA approach is compared with those of some conventional approaches under various measures of performance. Finally, it is proved that the performance of the HGA approach is superior to those of the others.

• Journal of Korea Multimedia Society
• /
• v.25 no.5
• /
• pp.685-694
• /
• 2022

The goal of few-shot semantic segmentation is to build a network that quickly adapts to novel classes with extreme data shortage regimes. Most existing few-shot segmentation methods leverage single or multiple prototypes from extracted support features. Although there have been promising results for natural images, these methods are not directly applicable to the aerial image domain. A key factor in few-shot segmentation on aerial images is to effectively exploit information that is robust against extreme changes in background and object scales. In this paper, we propose a Mask-Guided Attention module to extract more comprehensive support features for few-shot segmentation in aerial images. Taking advantage of the support ground-truth masks, the area correlated to the foreground object is highlighted and enables the support encoder to extract comprehensive support features with contextual information. To facilitate reproducible studies of the task of few-shot semantic segmentation in aerial images, we further present the few-shot segmentation benchmark iSAID-, which is constructed from a large-scale iSAID dataset. Extensive experimental results including comparisons with the state-of-the-art methods and ablation studies demonstrate the effectiveness of the proposed method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.8
• /
• pp.2881-2894
• /
• 2014

In complicated environment, context information plays an important role in image segmentation/labeling. The recently proposed auto-context algorithm is one of the effective context-based methods. However, the standard auto-context approach samples the context locations utilizing a fixed radius sequence, which is sensitive to large scale-change of objects. In this paper, we present a scale invariant auto-context (SIAC) algorithm which is an improved version of the auto-context algorithm. In order to achieve scale-invariance, we try to approximate the optimal scale for the image in an iterative way and adopt the corresponding optimal radius sequence for context location sampling, both in training and testing. In each iteration of the proposed SIAC algorithm, we use the current classification map to estimate the image scale, and the corresponding radius sequence is then used for choosing context locations. The algorithm iteratively updates the classification maps, as well as the image scales, until convergence. We demonstrate the SIAC algorithm on several image segmentation/labeling tasks. The results demonstrate improvement over the standard auto-context algorithm when large scale-change of objects exists.

• Journal of Information Processing Systems
• /
• v.18 no.3
• /
• pp.411-427
• /
• 2022

Digging historical and cultural information from seals in ancient books is of great significance. However, ancient Chinese seal samples are scarce and carving methods are diverse, and traditional digital image processing methods based on greyscale have difficulty achieving superior segmentation and recognition performance. Recently, some deep learning algorithms have been proposed to address this problem; however, current neural networks are difficult to train owing to the lack of datasets. To solve the afore-mentioned problems, we proposed an SEL-RefineMask which combines selector of feature pyramid network (SEL-FPN) with RefineMask to segment and recognize seals. We designed an SEL-FPN to intelligently select a specific layer which represents different scales in the FPN and reduces the number of anchor frames. We performed experiments on some instance segmentation networks as the baseline method, and the top-1 segmentation result of 64.93% is 5.73% higher than that of humans. The top-1 result of the SEL-RefineMask network reached 67.96% which surpassed the baseline results. After segmentation, a vision transformer was used to recognize the segmentation output, and the accuracy reached 91%. Furthermore, a dataset of seals in ancient Chinese books (SACB) for segmentation and small seal font (SSF) for recognition were established which are publicly available on the website.

• Atmosphere
• /
• v.30 no.2
• /
• pp.141-154
• /
• 2020

Seasonal forecast is growing in demand, as it provides valuable information for decision making and potential to reduce impact on weather events. This study examines how operational climate prediction systems can be reliable, producing the probability forecast in seasonal scale. A reliability diagram was used, which is a tool for the reliability by comparing probabilities with the corresponding observed frequency. It is proposed for a method grading scales of 1-5 based on the reliability diagram to quantify the reliability. Probabilities are derived from ensemble members using hindcast data. The analysis is focused on skill for 2 m temperature and precipitation from climate prediction systems in KMA, UKMO, and ECMWF, NCEP and JMA. Five categorizations are found depending on variables, seasons and regions. The probability forecast for 2 m temperature can be relied on while that for precipitation is reliable only in few regions. The probabilistic skill in KMA and UKMO is comparable with ECMWF, and the reliabilities tend to increase as the ensemble size and hindcast period increasing.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.63 no.3
• /
• pp.13-26
• /
• 2021

Recent drought events in the South Korea and the magnitude of drought losses indicate the continuing vulnerability of the agricultural drought. Various studies have been performed on drought hazard assessment at the regional scales, but until recently, drought management has been response oriented with little attention to mitigation and preparedness. A vulnerability assessment is introduced in order to preemptively respond to agricultural drought and to predict the occurrence of drought. This paper presents a method for spatial, Geographic Information Systems-based assessment of agricultural drought vulnerability in South Korea. It was hypothesized that the key 14 items that define agricultural drought vulnerability were meteorological, agricultural reservoir, social, and adaptability factors. Also, this study is to analyze agricultural drought vulnerability by comparing vulnerability assessment according to weighting method. The weight of the evaluation elements is expressed through the Analytic Hierarchy Process (AHP), which includes subjective elements such as surveys, and the Entropy method using attribute information of the evaluation items. The agricultural drought vulnerability map was created through development of a numerical weighting scheme to evaluate the drought potential of the classes within each factor. This vulnerability assessment is calculated the vulnerability index based on the weight, and analyze the vulnerable map from 2015 to 2019. The identification of agricultural drought vulnerability is an essential step in addressing the issue of drought vulnerability in the South Korea and can lead to mitigation-oriented drought management and supports government policymaking.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.3
• /
• pp.1477-1491
• /
• 2017

Inspired by the multi-scale nature of hippocampal place cells, a biologically inspired model based on a multi-scale spatial representation for goal-directed navigation is proposed in order to achieve robotic spatial cognition and autonomous navigation. First, a map of the place cells is constructed in different scales, which is used for encoding the spatial environment. Then, the firing rate of the place cells in each layer is calculated by the Gaussian function as the input of the Q-learning process. The robot decides on its next direction for movement through several candidate actions according to the rules of action selection. After several training trials, the robot can accumulate experiential knowledge and thus learn an appropriate navigation policy to find its goal. The results in simulation show that, in contrast to the other two methods(G-Q, S-Q), the multi-scale model presented in this paper is not only in line with the multi-scale nature of place cells, but also has a faster learning potential to find the optimized path to the goal. Additionally, this method also has a good ability to complete the goal-directed navigation task in large space and in the environments with obstacles.