• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.6
• /
• pp.26-36
• /
• 2021

In this study, a deep learning model for predicting the queue length was developed using the information collected from the image detector. Then, a multiple regression analysis model, a statistical technique, was derived and compared using two indices of mean absolute error(MAE) and root mean square error(RMSE). From the results of multiple regression analysis, time, day of the week, occupancy, and bus traffic were found to be statistically significant variables. Occupancy showed the most strong impact on the queue length among the variables. For the optimal deep learning model, 4 hidden layers and 6 lookback were determined, and MAE and RMSE were 6.34 and 8.99. As a result of evaluating the two models, the MAE of the multiple regression model and the deep learning model were 13.65 and 6.44, respectively, and the RMSE were 19.10 and 9.11, respectively. The deep learning model reduced the MAE by 52.8% and the RMSE by 52.3% compared to the multiple regression model.

• Journal of the Korean Society for Marine Environment & Energy
• /
• v.14 no.2
• /
• pp.107-113
• /
• 2011

To evaluate the performance of the hydrodynamic model, we introduced 10 skill assessments that are assorted by two groups: quantitative skill assessments (Absolute Average Error or AAE, Root Mean Squared Error or RMSE, Relative Absolute Average Error or RAAE, Percentage Model Error or PME) and qualitative skill assessments (Correlation Coefficient or CC, Reliability Index or RI, Index of Agreement or IA, Modeling Efficiency or MEF, Cost Function or CF, Coefficient of Residual Mass or CRM). These skill assessments were applied and calculated to evaluate the hydrodynamic modeling at one of Florida estuaries for water level, current, and salinity as comparing measured and simulated values. We found that AAE, RMSE, RAAE, CC, IA, MEF, CF, and CRM are suitable for the error assessment of water level and current, and AAE, RMSE, RAAE, PME, CC, RI, IA, CF, and CRM are good at the salinity error assessment. Quantitative and qualitative skill assessments showed the similar trend in terms of the classification for good and bad performance of model. Furthermore, this paper suggested the criteria of the "good" model performance for water level, current, and salinity. The criteria are RAAE < 10%, CC > 0.95, IA > 0.98, MEF > 0.93, CF < 0.21 for water level, RAAE < 20%, CC > 0.7, IA > 0.8, MEF > 0.5, CF < 0.5 for current, and RAAE < 10%, PME < 10%, CC > 0.9, RI < 1.15, CF < 0.1 for salinity.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.26 no.2
• /
• pp.139-147
• /
• 2008

The positioning accuracies tend to deteriorate as the distance between the rover and the reference station increases in the Real-Time Kinematic (RTK) surveys using Global Positioning System (GPS). To solve this problem, the National Geographic Information Institute (NGII) of Korea has installed Virtual Reference System (VRS), which is one of the network-based RTK systems. In this study, we conducted the accuracy tests of the VRS-RTK surveys. We surveyed 50 control points inside the NGII's GPS Continuously Operating Reference Stations (CORS) network using the VRS-RTK system, and compared the results with the published coordinates to verify the positioning accuracies. We also conducted the general RTK surveys at the same control points. The results showed that the positioning accuracy of the VRS-RTK was comparable to that of the general RTK, because the horizontal positioning accuracy was 3.1 cm while that of general RTK was 2.0 cm. Also the vertical positioning accuracy of VRS-RTK was 6.8 cm.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.1
• /
• pp.46-55
• /
• 2007

The Newton-Raphson's algorithm for finding a floating point reciprocal and inverse square root calculates the result by performing a fixed number of multiplications. In this paper, an improved Newton-Raphson's algorithm is proposed, that performs multiplications a variable number. Since the number of multiplications performed by the proposed algorithm is dependent on the input values, the average number of multiplications per an operation is derived from many reciprocal and inverse square tables with varying sizes. The superiority of this algorithm is proved by comparing this average number with the fixed number of multiplications of the conventional algorithm. Since the proposed algorithm only performs the multiplications until the error gets smaller than a given value, it can be used to improve the performance of a reciprocal and inverse square root unit. Also, it can be used to construct optimized approximate tables. The results of this paper can be applied to many areas that utilize floating point numbers, such as digital signal processing, computer graphics, multimedia, scientific computing, etc.

• Journal of the Korea Convergence Society
• /
• v.7 no.5
• /
• pp.175-180
• /
• 2016

The demand for accurate freeform apsheric surface is increasing to satisfy the optical performance. In this paper, we develop the algorithm for opto-mechatronics convergence, that reconstruct the surface 3D profiles from the curvarure data along two orthogonal directions. A synthetic freeform surface with 8.4 m diameter was simulated for the testing. The simulation results show that the reconstruction error is 0.065 nm PV(Peak-to-valley) and 0.013 nm RMS(Root mean square) residual difference. Finally the sensitivity to noise is diagnosed for probe position error, the simulation results proving that the suggested method is robust to position error.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.444-444
• /
• 2022

하천 수심 계측은 수심을 사람이 직접 계측하거나 초음파 기반 유속계 (ADCP) 등 최신 계측기기를 이용하여 간접적으로 계측을 실시하고 있다. 하지만 사람이 직접 하천에서 수심을 측정하는 것은 위험이 동반되고, 수심자료의 측정오차가 크게 발생한다. 따라서 수심측정에서 직접 측정 방식의 한계를 극복하기 위해, 초분광 영상의 반사도와 수심이 높은 상관관계를 보이는 것을 활용하여, 초분광 영상 기반 수심 산정 기법을 개발하였다. 초분광 영상 기반 수심 산정 기법은 복수의 파장이 존재하는 초분광영상으로부터 두 개의 파장대의 밴드를 추출하여 모든 경우의 수에 대해 밴드비를 산정한 후, 실측수심과 밴드비 간의 회귀분석을 실시하여 상관계수가 가장 높은 회귀식을 찾아내는 방식이 최적 밴드비 분석법에 기반한다. 최적 밴드비 분석법을 통해 획득된 높은 상관성의 밴드비-수심 관계식을 이용하여 수심을 추정할 수 있다. 이러한 방법은 직접 수심 측정 방식에 비해, 높은 해상도와 밀도, 양질의 데이터를 수집할 수 있는 장점이 있다. 과거 연구에 따르면 저수심부에서의 높은 정확도의 수심추정 결과를 보였지만, 고수심부에서는 실측수심과의 오차도 높아지는 등 정확성이 떨어지는 경향을 보인다. 따라서 본 연구에서는 보다 효율적인 수심계측을 할 수 있도록 최적 밴드비 분석법을 활용한 수심추정에서 신뢰성 있는 수심의 범위를 파악할 수 있는 방법을 제시하고자 한다. 본 연구에서는 대상지역으로 낙동강 본류와 황강 지류 합류부로 선정하였고, 초음파 기반 유속계(ADCP)와 드론을 활용하여 실측수심과 초분광 영상을 취득하였다. 민감도 분석을 위한 수심자료를 0.5m 단위로 분할하였으며, 구간별로 최적 밴드비 분석을 실시하였다. 그 결과, 구간별로 산정된 상관계수와 평균제곱근오차 (RMSE)를 통해 정확도가 높은 구간을 구별할 수 있었다. 또한 해당 구간을 초과하는 수심은 초분광 영상을 통해 추정이 어려운 것으로 판단되며, 분석한 구간까지를 최대 추정 가능 수심으로 정의하였다. 마지막으로 검증을 위해 최대추정가능수심으로 판단된 구간까지의 데이터만 활용하여 최적 밴드비 분석법을 적용하여 상관계수나 평균제곱근오차 결과의 개선여부 확인을 통해, 본 연구에서 제시한 방법이 정확한 최대추정가능수심 구간을 산정할 수 있는지 확인하였다.

• Journal of Korean Society of Disaster and Security
• /
• v.13 no.2
• /
• pp.27-38
• /
• 2020

Climate variations have become worse and diversified recently, which caused catastrophic disasters for our communities and ecosystem including economic property damages in Korea. Heat wave of summer season is one of causes for such damages of which outbreak tends to increase recently. Related short-term forecasting information has been provided by the Korea Meteorological Administration based on results from numerical forecasting model. As the study area, the ◯◯ province was selected because of the highest mortality rate in Korea for the past 15 years (1998~2012). When comparing the forecasted temperatures with field measurements, it showed RMSE of 1.57℃ and RMSE of 1.96℃ was calculated when only comparing the data corresponding to the observed value of 33℃ or higher. The forecasting process would take at least about 3~4 hours to provide the 4 hours advanced forecasting information. Therefore, this study proposes a methodology for temperature prediction using LSTM considering the short prediction time and the adequate accuracy. As a result of 4 hour temperature prediction using this approach, RMSE of 1.71℃ was occurred. When comparing only the observed value of 33℃ or higher, RMSE of 1.39℃ was obtained. Even the numerical prediction model of the whole range of errors is relatively smaller, but the accuracy of prediction of the machine learning model is higher for above 33℃. In addition, it took an average of 9 minutes and 26 seconds to provide temperature information using this approach. It would be necessary to study for wider spatial range or different province with proper data set in near future.

• The Korean Journal of Applied Statistics
• /
• v.19 no.2
• /
• pp.217-230
• /
• 2006

e discuss proper confidence intervals for interval estimation of a low binomial proportion. A large sample surveys are practically executed to find rates of rare diseases, specified industrial disaster, and parasitic infection. Under the conditions of 0 < p ${\leq}$ 0.1 and large n, we compared 6 confidence intervals with mean coverage probability, root mean square error and mean expected widths to search a good one for interval estimation of population proportion p. As a result of comparisons, Mid-p confidence interval is best and AC, score and Jeffreys confidence intervals are next.

• The Korean Journal of Applied Statistics
• /
• v.26 no.6
• /
• pp.943-952
• /
• 2013

Wald, Agresti-Coull, Jeffreys, and Bayes-Laplace methods are commonly used for confidence interval of binomial proportion are applied for prediction intervals. We used coverage probability, mean coverage probability, root mean squared error, and mean expected width for numerical comparisons. From the comparisons, we found that Wald is not proper as for confidence interval and Agresti-Coull is too conservative to differ from confidence interval. However, Jeffrey and Bayes-Laplace are good for prediction interval and Jeffrey is especially desirable as for confidence interval.

• The Korean Journal of Applied Statistics
• /
• v.34 no.4
• /
• pp.579-588
• /
• 2021

Wald, Agresti-Coull, Jeffreys, and Bayes-Laplace methods are commonly used for confidence interval of binomial proportion are applied for prediction intervals. We used coverage probability, mean coverage probability, root mean squared error, and mean expected width for numerical comparisons. From the comparisons, we found that Wald is not proper as for confidence interval and Agresti-Coull is too conservative to differ from confidence interval. However, Jeffrey and Bayes-Laplace are good for prediction interval and Jeffrey is especially desirable as for confidence interval.