• Journal of KIISE
• /
• v.41 no.12
• /
• pp.1081-1089
• /
• 2014

Context-sensitive spelling-error correction methods are largely classified into rule-based methods and statistical data-based methods, the latter of which is often preferred in research. Statistical error correction methods consider context-sensitive spelling error problems as word-sense disambiguation problems. The method divides a vocabulary pair, for correction, which consists of a correction target vocabulary and a replacement candidate vocabulary, according to the context. The present paper proposes a method that integrates a word-phrase n-gram model into a conventional model in order to improve the performance of the probability model by using a correction vocabulary pair, which was a result of a previous study performed by this research team. The integrated model suggested in this paper includes a method used to interpolate the probability of a sentence calculated through each model and a method used to apply the models, when both methods are sequentially applied. Both aforementioned types of integrated models exhibit relatively high accuracy and reproducibility when compared to conventional models or to a model that uses only an n-gram.

• Atmosphere
• /
• v.25 no.2
• /
• pp.271-281
• /
• 2015

The objectives of this study are to develop a hydrological drought outlook system using GloSea5 (Global Seasonal forecasting system 5) which has recently been used by KMA (Korea Meteorological Association) and to evaluate the forecasting capability. For drought analysis, the bilinear interpolation method was applied to spatially downscale the low-resolution outputs of GloSea5 and PR (Predicted Runoff) was produced for different lead times (i.e., 1-, 2-, 3-month) running LSM (Land Surface Model). The behavior of PR anomaly was similar to that of HR (Historical Runoff) and the estimated values were negative up to lead times of 1- and 2-month. For the evaluation of drought outlook, SRI (Standardized Runoff Index) was selected and PR_SRI estimated using PR. ROC score was 0.83, 0.71, 0.60 for 1-, 2- and 3-month lead times, respectively. It also showed the hit rate is high and false alarm rate is low as shorter lead time. The temporal Correlation Coefficient (CC) was 0.82, 0.60, 0.31 and Root Mean Square Error (RMSE) was 0.52, 0.86, 1.20 for 1-, 2-, 3-month lead time, respectively. The accuracy of PR_SRI was high up to 1- and 2-month lead time on local regions except the Gyeonggi and Gangwon province. It can be concluded that GloSea5 has high applicability for hydrological drought outlook.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.12 no.4
• /
• pp.71-79
• /
• 1992

In general, the accelerogram of earthquake ground motion or the accelerogram obtained from dynamic tests contain various errors. In these errors of the accelerograms, there are instrumental errors(magnitude and phase distortion) due to the response characteristics of accelerometer and the digitizing error concentrated in low and high frequency components and random errors. Then, these errors may be detrimental to the results of data processing and dynamic analysis. An efficient method which can correct the errors of the accelerogram is proposed in this study. The correction of errors can be accomplished through four steps as followes ; 1) using an interpolation method a data form appropriate to the error correction is prepared, 2) low and high frequency errors of the accelerogram are removed by band-pass filter between prescribed frequency limits, 3) instrumental errors are corrected using dynamic equilibrium equation of the accelerometer, 4) velocity and displacement are obtained by integrating corrected accelerogram. Presently, infinite impulse response(IIR) filter and finite impulse response (FIR) filter are generally used as band-pass filter. In the proposed error correction procedure, the deficiencies of FIR filter and IIR filter are reduced and, using the properties of the differentiation and the integration of Fourier transform, the accuracy of instrument correction and integration is improved.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.10 no.4
• /
• pp.550-563
• /
• 1999

A wavelet-Galerkin scheme based on the time-dependent Maxwell's equations is presented. Daubechies wavelet with two vanishing wavelet moments is expanded for basis function in spatial domain and Yee's leap-frog approach is applied. The shifted interpolation property of Daubechies wavelet family leads to the simplified formulations for inhomogeneous media without the additional matrices for the integral or material operator. The stability condition is formulated. The dispersion characteristics are analyzed and compared with those of finite difference time domain and multiresolution time domain methods. The analyses show the excellent trade-off between the regularity and the support width of the basis function. Although the basis function has only two vanishing wavelet moments, it is enough to provide negligible dispersive error in the numerical analysis and its compact support enables only several involved terms per nodes. The storage effectiveness, execution time reduction and accuracy of this scheme are demonstrated by calculating the resonant frequencies of the homogeneous and inhomogeneous cavities.

• Journal of Computational Design and Engineering
• /
• v.2 no.4
• /
• pp.206-217
• /
• 2015

This paper presents a complete S-shape feed rate scheduling approach (CSFA) with confined jerk, acceleration and command feed rate for parametric tool path. For a Non-Uniform Rational B-Spline (NURBS) tool path, the critical points of the tool path where the radius of curvature reaches extreme values are found firstly. Then, the NURBS curve is split into several NURBS sub-curves or blocks by the critical points. A bidirectional scanning strategy with the limitations of chord error, normal/tangential acceleration/jerk and command feed rate is employed to make the feed rate at the junctions between different NURBS blocks continuous. To improve the efficiency of the feed rate scheduling, the NURBS block is classified into three types: short block, medium block and long block. The feed rate profile corresponding to each NURBS block is generated according to the start/end feed rates and the arc length of the block and the limitations of tangential acceleration/jerk. In addition, two compensation strategies are proposed to make the feed rate more continuous and the arc increment more precise. Once the feed rate profile is determined, a second-order Taylor's expansion interpolation method is applied to generate the position commands. Finally, experiments with two free-form NURBS curves are conducted to verify the applicability and accuracy of the proposed method.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.537-540
• /
• 1992

This paper presents Samsung CNC (Computer Numerical Controller) system with an intel 80486/487 as the main CPU and a 32 bit floating point DSP(Digital Signal Processor) TMS320C30 as the motion control CPU. The Samsung CNC system diverse user-frienly characteristics such as multi-tasking, powerful menu system, internal PLC system, and 2/3 dimensional graphics in wire and solid mode. The main CPU executes central processing program, user interface program, interpreter, BMI, etc while the motion control CPU carries out some interpolations, acceleration/deceleration, and PID control algorithm with feedforward terms. Complex interpolations except linear and circular ones are performed on the main control CPU. The experimental results for the circular interpolation under linear acceleration/deceleration shows that the proposed CNC system can be widely used in controlling machining centers with good machining accuracy.

• Structural Engineering and Mechanics
• /
• v.45 no.5
• /
• pp.677-691
• /
• 2013

A semi-analytical finite strip method is developed for analyzing the post-buckling behavior of rectangular composite laminated plates of arbitrary lay-up subjected to progressive end-shortening in their plane and to normal pressure loading. In this method, all the displacements are postulated by the appropriate harmonic shape functions in the longitudinal direction and polynomial interpolation functions in the transverse direction. Thin or thick plates are assumed and correspondingly the Classical Plate Theory (CPT) or Higher Order Plate Theory (HOPT) is applied. The in-plane transverse deflection is allowed at the loaded ends of the plate, whilst the same deflection at the unloaded edges is either allowed to occur or completely restrained. Geometric non-linearity is introduced in the strain-displacement equations in the manner of the von-Karman assumptions. The formulations of the finite strip methods are based on the concept of the principle of the minimum potential energy. The Newton-Raphson method is used to solve the non-linear equilibrium equations. A number of applications involving isotropic plates, symmetric and unsymmetric cross-ply laminates are described to investigate the through-thickness shearing effects as well as the effect of pressure loading, end-shortening and boundary conditions. The study of the results has revealed that the response of the composite laminated plates is particularly influenced by the application of the Higher Order Plate Theory (HOPT) and normal pressure loading. In the relatively thick plates, the HOPT results have more accuracy than CPT.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.321-325
• /
• 2009

Sub-pel level motion estimation contributes to significant increase in R-D performance for H.264|MPEG 4 Part 10 AVC. However, several supplements, such as interpolation, block matching, and Hadamard transform which entails large computational complexity of encoding process, are essential to find best matching block in sub-pel level motion estimation and compensation. In this paper, a fast motion estimation scheme in sub-pel accuracy is proposed based on a parabolic model of SAD to avoid such computational complexity. In the proposed scheme, motion estimation (ME) is only performed in integer-pel levels and the following sub-pel level motion vectors are found from the parametric SAD model for which the model parameters are estimated from the SAD values obtained in the integer-pel levels. Fall-back check is performed to ensure the validity of the parabolic SAD model with the estimated parameters. The experiment result shows that the proposed scheme can reduce the motion estimation time up to about 30% of the total ME times in average with negligible amount of PSNR drops (0.14dB in maximum) and bit increments (2.54%in maximum).

• Journal of Broadcast Engineering
• /
• v.13 no.3
• /
• pp.339-349
• /
• 2008

We present a novel method to estimate the light source direction and relight a face texture image of a single 3D model under arbitrary unknown illumination conditions. We create a virtual irradiance sphere to detect the light source direction from a given illuminated texture image using both normal vector mapping and weighted bilinear interpolation. We then induce a relighting equation with estimated ambient and diffuse coefficients. We provide the result of a series of experiments on light source estimation, relighting and face recognition to show the efficiency and accuracy of the proposed method in restoring the shading and shadows areas of a face texture image. Our approach for face relighting can be used for not only illuminant invariant face recognition applications but also reducing visual load and Improving visual performance in tasks using 3D displays.

• The KIPS Transactions:PartB
• /
• v.14B no.4
• /
• pp.255-262
• /
• 2007

This paper proposes an enhanced skin color-based detection method to find a region of human face in color images. The proposed detection method combines three color spaces, RGB, $YC_bC_r$, YIQ and builds color union histograms of luminance and chrominance components respectively. Combined color union histograms are then fed in to the back-propagation neural network for training and Levenberg-Marquadt algorithm is applied to the iteration process of training. Proposed method with Levenberg-Marquadt algorithm applied to training process of neural network contributes to solve a local minimum problem of back-propagation neural network, one of common methods of training for face detection, and lead to make lower a detection error rate. Further, proposed color-based detection method using combined color union histograms which give emphasis to chrominance components divided from luminance components inputs more confident values at the neural network and shows higher detection accuracy in comparison to the histogram of single color space. The experiments show that these approaches perform a good capability for face region detection, and these are robust to illumination conditions.