• The Journal of Korea Robotics Society
• /
• v.17 no.4
• /
• pp.425-430
• /
• 2022

As robots are considered one of the mainstream digital transformations, robots with machine vision becomes a main area of study providing the ability to check what robots watch and make decisions based on it. However, it is difficult to find a small object in the image mainly due to the flaw of the most of visual recognition networks. Because visual recognition networks are mostly convolution neural network which usually consider local features. So, we make a model considering not only local feature, but also global feature. In this paper, we propose a detection method of a small marker on the object using deep learning and an algorithm that considers global features by combining Transformer's self-attention technique with a convolutional neural network. We suggest a self-attention model with new definition of Query, Key and Value for model to learn global feature and simplified equation by getting rid of position vector and classification token which cause the model to be heavy and slow. Finally, we show that our model achieves higher mAP than state of the art model YOLOr.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.30 no.1
• /
• pp.81-90
• /
• 1988

A physically-based, distributed hydrologic model ANSWERS is described and its test results with two small watersheds near the Banweol reservoir are presented. The sizes of the watersheds are 270 ha and 477 ha, respectively. The smaller one has a mild topography with 40 percent forest area. The other has a steeper slope and 87 percent forest area. The model parameters were calibrated using observed runoff data and used for storm runoff simulation. The simulated peak discharges were in good agreement with the data. The model underestimated the recession parts of hydrographa as compared to the observed ones.

• Journal of Korean Society for Quality Management
• /
• v.32 no.3
• /
• pp.94-108
• /
• 2004

TQM has been applied to the industry area regardless of the size of company. BNQA model should be identically applied to large companies and small & medium-sized company. However, there are many evidences that TQM model should be modified according to the size of company. In this study, we analyzed the practice of TQM in Korean small & medium-sized Company. Also, we tried to give some suggestions for improving the competitiveness of the small & medium-sized company. The result of this study is that leadership, work system, education and strategy development should be improved in domestic companies, and work system, education and strategy deployment should be improved in foreign companies.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2022.04a
• /
• pp.206-207
• /
• 2022

Currently, safety accidents in construction area are managed regardless of the size. Therefore, the objective of this study is to conduct for developing the quantitative risk assessment according to large and small and medium-sized construction sites. The scope of this study is limited to the fall accidents which is the biggest accidents in the construction sites. the regression analysis was conducted based on the collected data. As a result, it was confirmed that there was a statistically significant difference between larce and small and medium-sized construction sites. This study is expected to be used as basic data for research on the development of a risk quantitative model for small and medium-sized construction sites in the future.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.6
• /
• pp.1621-1629
• /
• 2016

This paper studies Bayesian pooling for analysis of categorical data from small areas. Many surveys consist of categorical data collected on a contingency table in each area. Statistical inference for small areas requires considerable care because the subpopulation sample sizes are usually very small. Typically we use the hierarchical Bayesian model for pooling subpopulation data. However, the customary hierarchical Bayesian models may specify more exchangeability than warranted. We, therefore, investigate the effects of pooling in hierarchical Bayesian modeling for the contingency table from small areas. In specific, this paper focuses on the methods of direct or indirect pooling of categorical data collected on a contingency table in each area through Dirichlet priors. We compare the pooling effects of hierarchical Bayesian models by fitting the simulated data. The analysis is carried out using Markov chain Monte Carlo methods.

• Journal of Korean Society on Water Environment
• /
• v.31 no.2
• /
• pp.142-150
• /
• 2015

The relationship between the number of benthic macroinvertebrate species and the accumulated survey area were investigated in a clean stream and an impaired stream of Korea. Five models to characterize species-area functions were compared, and the Weibull model fitted species-area data well. The other models (Arrhenius, Romell-Gleason, Kylin, Lognormal model) had small or notable bias. The maximum number of species and half-saturation area derived from the Weibull model may be used as the indicators of the carrying capacity and the habitat complexity respectively.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2004.05b
• /
• pp.646-650
• /
• 2004

The purpose of this study is to determine the optimal groundwater yield for a small rural watershed, which is necessary for effective groundwater management. The study area$(3.89km^2)$ is located in Kyungpook Sangju Yangchon-dong and hourly groundwater level in an observation well is observed and the data are used to verify the visual MODFLOW model. The groundwater model is applied in the same area to obtain optimal yield for 1992 and 1994, 1982. The optimal yield in this experimental watershed ranged $12.5\%\;to\;14.0\%$ of the annual infiltration rate.

• Journal of the Korean Data and Information Science Society
• /
• v.26 no.3
• /
• pp.755-762
• /
• 2015

One of the main objectives of the U.S. Census Bureau is the proper estimation of median household income for small areas. These estimates have an important role in the formulation of various governmental decisions and policies. Since direct survey estimates are available annually for each state or county, it is desirable to exploit the longitudinal trend in income observations in the estimation procedure. In this study, we consider Fay-Herriot type small area models which include time-specific random effect to accommodate any unspecified time varying income pattern. Analysis is carried out in a hierarchical Bayesian framework using Markov chain Monte Carlo methodology. We have evaluated our estimates by comparing those with the corresponding census estimates of 1999 using some commonly used comparison measures. It turns out that among three types of time-specific random effects the small area model with a time series random walk component provides estimates which are superior to both direct estimates and the Census Bureau estimates.

• Journal of Navigation and Port Research
• /
• v.31 no.7
• /
• pp.579-587
• /
• 2007

Samchunpo(Sin Hyang) Harbor is located in the bay of Sa Chun, the central south coast of Korean peninsula. The harbor and coastal boundaries have been protecting by natural coastal islands and shoals. Currently, The Sin Hyang harbor needs maintenance and renovation of the sheltered structures against the weather deterioration and typhoon damages. Consequently to support this, the calculation of accurate design wave through the typhoon wave attack is necessary. In this study, calculation of incident wave condition is simulated using steady state spectrum energy wave model(wide area wave model) from 50 years return wave condition. And this simulation results in wide offshore area were used for the input of the extended mild slope wave model at the narrow coastal area. Finally, the calculation of design wave at Sin Hyang harbor entrance was induced by Boussinesq wave model(detail area wave model) simulation. The numerical model system was able to simulate wave transformations from generation scale to shoreline or harbor impact. We hope these results will be helpful to the engineers doing placement, design, orientation, and evaluation of a wide range of potential solutions in this area.

• Journal of Korea Water Resources Association
• /
• v.36 no.1
• /
• pp.87-104
• /
• 2003

A numerical model of surface runoff and river flow has been applied to small- and large-size catchment areas in order to investigate the physical characteristics of river flow during flood period. Several refinements are made on the existing model SIRG-RS for the ways of rainfall input through surface runoff, river junction treatment and the computation of river flow on steep slope. For the computation of frictional forces, employed is the power law of friction factor which is a function of Reynolds number and relative roughness height. The empirical equation of friction factor is developed using recent field data as well as laboratory data. The refined model has been applied to small-size catchment area as well as large-size catchment area, and the computation results are found in good agreement with the observations in both cases.