• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.1
• /
• pp.272-287
• /
• 2016

Recently, various and effective contrast based salient object detection models to focus on a single target have been proposed. However, there is a lack of research on detection of multiple objects, and also it is a more challenging task than single target process. In the multiple target problem, we are confronted by new difficulties caused by distinct difference between properties of objects. The characteristic of existing models depending on the global maximum distribution of data point would become a drawback for detection of multiple objects. In this paper, by analyzing limitations of the existing methods, we have devised three main processes to detect multiple salient objects. In the first stage, regional features are extracted from over-segmented regions. In the second stage, the regional features are categorized into homogeneous cluster using the mean-shift algorithm with the kernel function having various sizes. In the final stage, we compute saliency scores of the categorized regions using only spatial features without the contrast features, and then all scores are integrated for the final salient regions. In the experimental results, the scheme achieved superior detection accuracy for the SED2 and MSRA-ASD benchmarks with both a higher precision and better recall than state-of-the-art approaches. Especially, given multiple objects having different properties, our model significantly outperforms all existing models.

• The Bulletin of The Korean Astronomical Society
• /
• v.37 no.1
• /
• pp.56.2-56.2
• /
• 2012

We present a study of the asymptotic giant branch (AGB) contribution to the total Ks band luminosity of star clusters in the Large Magellanic Cloud (LMC) as a function of age. AGB stars, a representative intermediate-age population, are a strong source of NIR to MIR emission so that they are a critical component for understanding the near-to-mid infrared observation of galaxies. Current calibration of IR emission in evolutionary population synthesis (EPS) models for galaxies is mainly based on a small number of LMC star clusters. However, each LMC star cluster with intermediate age contains only a few AGB stars so that it suffers from a stochastic effect. Therefore a large number of them are needed for solid calibration of the EPS models. We study physical properties of a large number of LMC star clusters to estimate the Ks band luminosity fraction of AGB stars in star clusters as a function of age. We discuss the stochastic effect in calibrating models, and the importance of this calibration for studying the evolution of not only nearby galaxies but also of high-z galaxies.

• Health Policy and Management
• /
• v.6 no.2
• /
• pp.58-90
• /
• 1996

This study provides an empirical analysis of location competition for demand maximization by central place theory among physicians in nonmetropolitan areas of Korea. The results show that the primary care physicians distribute themselves evenly from urban communities to rural communities. The subspecialists, however, cluster together in major cities rather than decentralize themselves in rural counties. This study establishes the three statistical models : Primary care physicians, subspecialist physicians, and total physicians. Two models of primary care physicians and total physicians have a strong significance in multiple regression analysis (p=-.0001). The primary care model explains approximately 45% of the variation and the total physicians model explains approximately 70% of the variation in physician/1,000 population ratios across national counties. The subspecialist physicians model analysze the tobit regreassion because of the left consored and truncated values(57 cases = 0). In all three models, analysis of the coefficiencts for physician centralization degrees in the 0- to 5- and 5- to 10-km rings around the core county reveals each a positive and negative association betwee these degrees and the physician/1,000 population ratios in the core county. Also, the results provide moderate evidence that the relationship between clinic physicians and community hospitals is competitive, and the relationship between clinic physicians and pharmaceutists is synergistic. This suggests that public policy makers and local self-governing bodies must take an active role to ensure procider availability and the regional health planning in all nonmetropolitan areas of Korea.

• Earthquakes and Structures
• /
• v.9 no.5
• /
• pp.957-981
• /
• 2015

Earthquakes occur as a cluster in many regions around the world where complex fault systems exist. The repeated shaking usually induces accumulative damage to affected structures. Damage accumulation in structural systems increases their level of degradation in stiffness and also reduces their strength. Many existing analytical tools of modeling RC structures lack the salient damage features that account for stiffness and strength degradation resulting from repeated earthquake loading. Therefore, these tools are inadequate to study the response of structures in regions prone to multiple earthquakes hazard. The objective of this paper is twofold: (a) develop a tool that contains appropriate damage features for the numerical analysis of RC structures subjected to more than one earthquake; and (b) conduct a parametric study that investigates the effects of multiple earthquakes on the response of RC moment resisting frame systems. For this purpose, macroscopic constitutive models of concrete and steel materials that contain the aforementioned damage features and are capable of accurately capturing materials degrading behavior, are selected and implemented into fiber-based finite element software. Furthermore, finite element models that utilize the implemented concrete and steel stress-strain hysteresis are developed. The models are then subjected to selected sets of earthquake sequences. The results presented in this study clearly indicate that the response of degrading structural systems is appreciably influenced by strong-motion sequences in a manner that cannot be predicted from simple analysis. It also confirms that the effects of multiple earthquakes on earthquake safety can be very considerable.

• Wind and Structures
• /
• v.29 no.2
• /
• pp.139-147
• /
• 2019

One of the most important factors in tall buildings design in urban spaces is wind. The present study aims to investigate the aerodynamic behavior in the square and triangular footprint forms through aerodynamic modifications including rounded corners, chamfered corners and recessed corners in order to reduce the length of tall buildings wake region. The method used was similar to wind tunnel numerical simulation conducted on 16 building models through Autodesk Flow Design 2014 software. The findings revealed that in order to design tall 50 story buildings with a height of about 150 meters, the model in triangular footprint with aerodynamic modification of chamfered corner facing wind direction came out to have the best aerodynamic behavior comparing the other models. In comparison to the related reference model (i.e., the triangular footprint with sharp corners and no aerodynamic modification), it could reduce the length of the wake region about 50% in general. Also, the model with square footprint and aerodynamic modification of chamfered corner with the corner facing the wind could present favorable aerodynamic behavior comparing the other models of the same cluster. In comparison to the related reference model (i.e., the square footprint with sharp corners and no aerodynamic modification), it could decrease the wake region up to 30% lengthwise.

• The Bulletin of The Korean Astronomical Society
• /
• v.43 no.2
• /
• pp.40.4-41
• /
• 2018

The current 'standard model' of cosmology provides a minimal theoretical framework that can explain the gaussian, nearly scale-invariant density perturbations observed in the CMB to the late time clustering of galaxies. However accepting this framework, requires that we include within our cosmic inventory a vacuum energy that is ~122 orders of magnitude lower than Quantum Mechanical predictions, or alternatively a new scalar field (dark energy) that has negative pressure. An alternative approach to adding extra components to the Universe would be to modify the equations of Gravity. Although GR is supported by many current observations there are still alternative models that can be considered. Recently there have been many works attempting to test for modified gravity using the large scale clustering of galaxies, ISW, cluster abundance, RSD, 21cm observations, and weak lensing. In this work, we compare various modified gravity models using cosmic shear data from the Deep Lens Survey as well as data from CMB, SNe Ia, and BAO. We use the Bayesian Evidence to quantify the comparison robustly, which naturally penalizes complex models with weak data support. In this talk we present our methodology and preliminary results that show f(R) gravity is mildly disfavoured by the data.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.19 no.10
• /
• pp.947-952
• /
• 2006

It has been confirmed that the inner defect of insulator and the perfect diagnosis for aging are closely related to safe electric power transmission system and that the detection of accident and diagnosis technique turn out to be very important issues. But perfect diagnosis is difficult because discharge pattern is irregular. Thus, we investigated discharge pattern using the new distribution statistical models with cross-inked polyethylene(XLPE) specimens. Voltage was applied to power frequency by step method, and calibration of discharge was set to 50 pC. After the voltage was applied, it measured the discharge occurring during 10s. We investigated discharge pattern using the K-means analysis and Weibull function. We also investigated variation of centroid and shape parameter due to variation of voltage. As a result of analyzing K-means, it was confirmed that cluster including many object numbers was formed by the presence of void. And result of Weibull distribution, it was confirmed that shape parameter of discharge varied from 1.28 to 1.62 in no void specimens, and that shape parameter of discharge number varied from 1.28 to 1.62. In the void, shape parameter of discharge varied from 5.66 to 6.43, and shape parameter of discharge number varied from 5.05 to 5.08.

• The Bulletin of The Korean Astronomical Society
• /
• v.35 no.2
• /
• pp.83.1-83.1
• /
• 2010

Recent observations of Sloan Digital Sky Survey (SDSS) for the ugriz filters have provided the largest and the most homogeneous photometric database. Especially, the ugriz systems have been used in many telescopes, such as SDSS, CFHT, and Gemini, and will be applied next generation large telescopes. In addition, many stellar evolution models, like as PADOVA, BaSTI and DSEP, have introduced theoretical isochrones in ugriz system, to apply the isochrones into the photometry of stars in globular clusters. In this study, we transformed Yonsei-Yale($Y^2$) isochrones to ugriz photometric system and fit the isochrones to the (g-r, r), (g-i, r), and (r-i, r) CMDs of 13 globular clusters (Chun et al. 2009, Clem et al. 2008, An et al. 2008). We found that the derived ages from $Y^2$-isochrones are 1~3Gyr younger and the estimated distance moduli are 0.3mag larger than the values inferred from the other isochrones in r-band. Also, the E(B-V) are 0.02~0.04mag less than those estimated from the other models. The ages of each globular cluster estimated from $Y^2$-isochrones are 12~14Gyr and distance moduli show good agreement with previous studies. From this result, we confirmed the availability of the $Y^2$-isochrones in gri filters. However, it is a problem which will improve that E(B-V) values obtained from $Y^2$-isochrones are too small. Finally, the result of this study is expected to be used research of globular cluster in ugirz photometric system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.5
• /
• pp.1963-1978
• /
• 2015

With the advancement of mobile web environments, identification and analysis of the user behavior play a significant role and remains a challenging task to implement with variations observed in the model. This paper presents an efficient method for mining optimized user behavior prediction model using genetic algorithm on mobile web structure. The framework of optimized user behavior prediction model integrates the temporary and permanent register information and is stored immediately in the form of integrated logs which have higher precision and minimize the time for determining user behavior. Then by applying the temporal characteristics, suitable time interval table is obtained by segmenting the logs. The suitable time interval table that split the huge data logs is obtained using genetic algorithm. Existing cluster based temporal mobile sequential arrangement provide efficiency without bringing down the accuracy but compromise precision during the prediction of user behavior. To efficiently discover the mobile users' behavior, prediction model is associated with region and requested services, a method called optimized user behavior Prediction Model using Genetic Algorithm (PM-GA) on mobile web structure is introduced. This paper also provides a technique called MAA during the increase in the number of models related to the region and requested services are observed. Based on our analysis, we content that PM-GA provides improved performance in terms of precision, number of mobile models generated, execution time and increasing the prediction accuracy. Experiments are conducted with different parameter on real dataset in mobile web environment. Analytical and empirical result offers an efficient and effective mining and prediction of user behavior prediction model on mobile web structure.

• The KIPS Transactions:PartC
• /
• v.16C no.2
• /
• pp.143-150
• /
• 2009

In trust-based networks, it is very important how to decide a node is trustworthy when one node performs communications with other node. In current networks based on trust, a node creates a new trust value from observation and then establishes an intercommunication path through the process of evaluating a targeted communication node. In our paper, we propose four trust models in which a new node enters a cluster and finds a target node to create a communication in ad-hoc networks. The proposed models have been classified according to the existence of reputation server and the trust evaluation functions. Through the proposed model, we found that new node finds target node more quickly in which there exists a reputation server in the cluster and considers neighbor node's recommendation as well as own experience information when calculates trust values. As our performance analysis, we focus the communication delay time to generate a final trust value for each trust model.