• Journal of KIISE:Computer Systems and Theory
• /
• v.30 no.11
• /
• pp.610-620
• /
• 2003

Various shadows are one of main factors that cause errors in vision based vehicle detection. In this paper, two simple methods, land mark based method and BS & Edge method, are proposed for vehicle detection and shadow rejection. In the experiments, the accuracy of vehicle detection is higher than 96%, during which the shadows arisen from roadside buildings grew considerably. Based on these two methods, vehicle counting, tracking, classification, and speed estimation are achieved so that real-time traffic parameters concerning traffic flow can be extracted to describe the load of each lane.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.55 no.1
• /
• pp.9-15
• /
• 2013

This study is to evaluate the parameter behavior of VfloTM distributed rainfall-runoff model by applying 3 kinds of rainfall interpolation methods viz. Inverse Distance Weighting (IDW), Kriging (KRI), and Thiessen network (THI). For the 1,544 $km^2$ Dongcheon watershed of Nakdong river, the model was calibrated using 4 storm events in 2007 and 2009, and validated using 2 storm events in 2010. The model was calibrated with Nash-Sutcliffe model efficiency of 0.97 for IDW, 0.94 for KRI, and 0.95 for THI respectively. For the sensitive parameters, the saturated hydraulic conductivity ($K_{sat}$) for IDW, KRI, and THI were 0.33, 0.31, and 0.43 cm/hr, and the soil suction head at the wetting front (${\Psi}_f$) were 4.10, 3.96, and 5.19 cm $H_2O$ respectively. These parameters affected the infiltration process by the spatial distribution of antecedent moisture condition before a storm.

• Journal of the Korean Institute of Telematics and Electronics T
• /
• v.35T no.3
• /
• pp.87-95
• /
• 1998

This paper presents the method for the automatic tuning of a design weighting polynomial parameter of a generalized minimum-variance stochastic self tuning controller which adapts to changes in the system parameters with time delays and noises. The self tuning effect is achieved through the recursive least square algorithm at the parameter estimation stage and also through the Robbins-Monro algorithm at the stage of optimizing a design weighting polynomial parameters. The proposed self tuning method is simple and effective compared with other existing self tuning methods. The computer simulation results are presented to illustrate the procedure and to show the performance of the control system.

• Computational Structural Engineering
• /
• v.10 no.2
• /
• pp.149-160
• /
• 1997

Mechanical structures are composed of substructures connected by joints and boundary elements. While the finite element representation of plain substructures is well developed and reliable, joints have a lot of uncertainties in being accurately modelled and affect dynamic behavior of a total system. In order to improve the accuracy of a finite element model, a new method is proposed, in which reduced finite element model is combined with a system identification technique. After substructures except joints are modelled with finite element method and joint properties are represented by parameter states, non-linear state equation is derived in which parameter states are multiplied by physical states such as displacements and velocities. So the joint parameter identification is transformed into non-linear state estimation problem. The methods are tested and discussed numerically and the feasibility for physical application has been demonstrated through two example structures.

• Communications for Statistical Applications and Methods
• /
• v.28 no.1
• /
• pp.1-19
• /
• 2021

Heavy tailed distributions are useful for modeling actuarial and financial risk management problems. Actuaries often search for finding distributions that provide the best fit to heavy tailed data sets. In the present work, we introduce a new class of heavy tailed distributions of a special sub-model of the proposed family, called a new extended alpha power transformed Weibull distribution, useful for modeling heavy tailed data sets. Mathematical properties along with certain characterizations of the proposed distribution are presented. Maximum likelihood estimates of the model parameters are obtained. A simulation study is provided to evaluate the performance of the maximum likelihood estimators. Actuarial measures such as Value at Risk and Tail Value at Risk are also calculated. Further, a simulation study based on the actuarial measures is done. Finally, an application of the proposed model to a heavy tailed data set is presented. The proposed distribution is compared with some well-known (i) two-parameter models, (ii) three-parameter models and (iii) four-parameter models.

• Journal of applied mathematics & informatics
• /
• v.27 no.1_2
• /
• pp.315-326
• /
• 2009

The unknown parameters in regression models are usually estimated by using various existing methods. There are several existing methods, such as the least squares method, which is the most common one, the least absolute deviation method, the regression quantile method, and the asymmetric least squares method. For the nonlinear time series regression models, which do not satisfy the general conditions, we will compare them in two ways: 1) a theoretical comparison in the asymptotic sense and 2) an empirical comparison using Monte Carlo simulation for a small sample size.

• Communications for Statistical Applications and Methods
• /
• v.26 no.6
• /
• pp.591-610
• /
• 2019

Deep Learning is one of the machine learning methods to find features from a huge data using non-linear transformation. It is now commonly used for supervised learning in many fields. In particular, Convolutional Neural Network (CNN) is the best technique for the image classification since 2012. For users who consider deep learning models for real-world applications, Keras is a popular API for neural networks written in Python and also can be used in R. We try examine the parameter estimation procedures of Deep Neural Network and structures of CNN models from basics to advanced techniques. We also try to figure out some crucial steps in CNN that can improve image classification performance in the CIFAR10 dataset using Keras. We found that several stacks of convolutional layers and batch normalization could improve prediction performance. We also compared image classification performances with other machine learning methods, including K-Nearest Neighbors (K-NN), Random Forest, and XGBoost, in both MNIST and CIFAR10 dataset.

• Journal of the military operations research society of Korea
• /
• v.27 no.1
• /
• pp.101-113
• /
• 2001

Taguchi and VTA(variation Transmission Analysis) methods have been widely used recently as new methods for product design. In this study, Taguchi method using analysis of variance and VTA method using regression analysis are reviewed and compared with each other in terms of parameter design and tolerance design. In analysis of variance, variation of quality characteristics arises from noise factors, therefore the optimal levels of design factors are selected to minimize the effect of noise factors. n regression analysis, variation of quality characteristics arises from variation of each own design factors. As a method to reduce variation of these quality characteristics, sensitivity analysis was performed for each design factors. An example of calculating tolerance interval for the given defect rate in PPM is also introduced. Especially, the new method is suggested to increase the estimation accuracy of variation of quality characteristics through regression analysis.

• The Journal of Engineering Geology
• /
• v.18 no.4
• /
• pp.359-369
• /
• 2008

For the design and effective management of underground storage caverns, preliminary investigation on the hydrogeologic parameters around caverns and analysis on the groundwater flow must be carried out. The data collection is very imporatnat task for the hydrogeologic design so various hydraulic tests have been performed. When analyzing the injection/fall off test data, existing graphical method to estimate the parameters in Theis' equation is widely used. However this method has some sources of error when estimating parameters by means of human faults. Therefore the method of estimating parameters by means of statistical methods such as regression type is evaluated as a useful tool. In this study, nonlinear regression analysis for the Theis' equation is suggested and applied to the estimation of parameters for the real field interference data around underground storage caverns. Damping parameter which reduce the iteration numbers and inhance the convergence is also introduced.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.4
• /
• pp.349-355
• /
• 2015

Underwater robots generally show better performances for tasks than humans under certain underwater constraints such as. high pressure, limited light, etc. To properly diagnose in an underwater environment using remotely operated underwater vehicles, it is important to keep autonomously its own position and orientation in order to avoid additional control efforts. In this paper, we propose an efficient method to assist in the operation for the various disturbances of a remotely operated vehicle for the diagnosis of underwater structures. The conventional AHRS-based bearing estimation system did not work well due to incorrect measurements caused by the hard-iron effect when the robot is approaching a ferromagnetic structure. To overcome this drawback, we propose a sensor fusion algorithm with the camera and AHRS for estimating the pose of the ROV. However, the image information in the underwater environment is often unreliable and blurred by turbidity or suspended solids. Thus, we suggest an efficient method for fusing the vision sensor and the AHRS with a criterion which is the amount of blur in the image. To evaluate the amount of blur, we adopt two methods: one is the quantification of high frequency components using the power spectrum density analysis of 2D discrete Fourier transformed image, and the other is identifying the blur parameter based on cepstrum analysis. We evaluate the performance of the robustness of the visual odometry and blur estimation methods according to the change of light and distance. We verify that the blur estimation method based on cepstrum analysis shows a better performance through the experiments.