• The Journal of the Acoustical Society of Korea
• /
• v.28 no.8
• /
• pp.832-842
• /
• 2009

Automatic speech recognition (ASR) has been successfully applied to many real human computer interaction (HCI) applications; however, its performance tends to be significantly decreased under noisy environments. The invention of audio visual speech recognition (AVSR) using an acoustic signal and lip motion has recently attracted more attention due to its noise-robustness characteristic. In this paper, we describe our novel integration scheme for AVSR based on a late integration approach. Firstly, we introduce the robust reliability measurement for audio and visual modalities using model based information and signal based information. The model based sources measure the confusability of vocabulary while the signal is used to estimate the noise level. Secondly, the output probabilities of audio and visual speech recognizers are normalized respectively before applying the final integration step using normalized output space and estimated weights. We evaluate the performance of our proposed method via Korean isolated word recognition system. The experimental results demonstrate the effectiveness and feasibility of our proposed system compared to the conventional systems.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.94-94
• /
• 2000

In this paper, the performance of INS/GPS integration is analyzed. Generally the integration method is classified into two method. One is loosely coupled integration, Another is tightly coupled integration. For the simulation, the error model of Kalman fitter is defined and the result of simulation is analyzed. When the number of visible satellites is less than four, the error of tightly coupled integration is smaller than that of loosely coupled integration. To very the result of simulation, van test that use LP-81 IMU and Millen3151R GPS receiver is carried out.

• Structural Engineering and Mechanics
• /
• v.69 no.3
• /
• pp.243-255
• /
• 2019

Utilization of fiber beam-column element has gained considerable attention in recent years due mainly to its ability to model distributed plasticity over the length of the element through a number of integration points. However, the relatively high sensitivity of the method to modeling parameters as well as material behavior models can pose a significant challenge. Residual drift is one of the seismic demands which is highly sensitive to modeling parameters and material behavior models. Permanent deformations play a prominent role in the post-earthquake evaluation of serviceability of bridges affected by a near-fault ground shaking. In this research, the influence of distributed plasticity modeling parameters using both force-based and displacement-based fiber elements in the prediction of internal forces obtained from the nonlinear static analysis is studied. Having chosen suitable type and size of elements and number of integration points, the authors take the next step by investigating the influence of material behavioral model employed for the prediction of permanent deformations in the nonlinear dynamic analysis. The result shows that the choice of element type and size, number of integration points, modification of cyclic concrete behavior model and reloading strain of concrete significantly influence the fidelity of fiber element method for the prediction of permanent deformations.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.27 no.11
• /
• pp.1907-1916
• /
• 2003

During decades, there has been much progress in understanding of the inelastic behavior of the materials and numerous inelastic constitutive equations have been developed. The complexity of these constitutive equations generally requires a stable and accurate numerical method. To obtain the increment of state variable, its evolution laws are linearized by several approximation methods, such as general midpoint rule(GMR) or general trapezoidal rule(GTR). In this investigation, semi-implicit integration schemes using GTR and GMR were developed and implemented into ABAQUS by means of UMAT subroutine. The comparison of integration schemes was conducted on the simple tension case, and simple shear case and nonproportional loading case. The fully implicit integration(FI) was the most stable but amplified the truncation error when the nonlinearity of state variable is strong. The semi-implicit integration using GTR gave the most accurate results at tension and shear problem. The numerical solutions with refined time increment were always placed between results of GTR and those of FI. GTR integration with adjusting midpoint parameter can be recommended as the best integration method for viscoplastic equation considering nonlinear kinematic hardening.

• Proceedings of the KSR Conference
• /
• 2009.05b
• /
• pp.233-237
• /
• 2009

In this study, we suggest that several approaches to evaluate the collision acceleration value of a car in the article 35 and the guideline 16 of Korean rolling stock safety regulation. There are various methods to evaluate collision acceleration such as; a displacement comparison method by the double integration of filtered acceleration data, a velocity comparison method by the integration of filtered acceleration data, an analysis method of time-velocity curve, or a differential method of time-velocity curve. We compared these methods one another using 1D dynamic simulation model composed of nonlinear dampers, springs and bars, and masses. Also, we applied these methods to a hybrid model, which is made of 3D shell element model and 2D collision dynamics model, in order to evaluate whether 1D force-displacement curve modeling for energy absorbing structures have an effect on the collision acceleration levels or not.

• Journal of the Korean Society for Precision Engineering
• /
• v.18 no.4
• /
• pp.129-134
• /
• 2001

It is essential to predict the welding deformation at assembly stage, to increase productivity through mechanization and automation effectively. A practical analysis method appled for production engineering was proposed to simulate the deformation of arc welding, with an analytical model using finite element method solving thermal-elastic-plastic behavior. In this research, for accurate assembling, 3-D thermal-elastic-plastic finite element model is used to simulate the out-of-plane deformation caused by arc welding. Efforts have been made to find out the efficient method to improve the reliability and accuracy of the numerical calculation. Each of theories of small and large deformation is applied in solving 3-D thermal-elastic-plastic problem to compare with their efficiency about calculation imes and solution accuracy. When solid elements are used in a bending problem of a plate, phenomenon that the predictive deformation is more than that of actual survey is observed. To prevent this phenomenon, reduced integration method for element is employed instead of full integration that is generally used in 3-D thermal-elastic-plastic analysis.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 1997.10a
• /
• pp.87-94
• /
• 1997

The circular plate resting on Boussinesq's half-space model under axisymmetric loading is studied by a finite element procedure to evaluate the distribution of contact pressure between plate and elastic half-space. The displacement of half-space due to axisymmetric surface loading can be evaluated by double integration of Boussinesq's solution. On that case the analytical integration can be executed for the radial direction but the analytical integration for the circumferential direction is impossible and the numerical integration should be considered. With the radial integration we can get non-dimensional function. Then the numerical integration for the formula is executed for the circumferential direction and the results are approximated 5th order Polynomials by using the least square method. With these 5th order approximate formula, the flexibility matrix of half-space is constructed as the coefficient matrix of nodal contact pressure by the finite element procedures. Iteration procedures are attempted by using this method to determine the separated region.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.31 no.6_2
• /
• pp.577-583
• /
• 2013

Recently, massive archives of ground information imagery from new sensors have become available. To establish a functional relationship between the image and the ground space, sensor models are required. The rational functional model (RFM), which is used as an alternative to the rigorous sensor model, is an attractive option owing to its generality and simplicity. To determine the rational polynomial coefficients (RPC) in RFM, however, we encounter the problem of obtaining a stable solution. The design matrix for solutions is usually ill-conditioned in the experiments. To solve this unstable solution problem, regularization techniques are generally used. In this paper, we describe the effective determination of the optimal regularization parameter in the regularization technique during RPC derivation. A brief mathematical background of RFM is presented, followed by numerical approaches for effective determination of the optimal regularization parameter using the Euler Method. Experiments are performed assuming that a tilted aerial image is taken with a known rigorous sensor. To show the effectiveness, calculation time and RMSE between L-curve method and proposed method is compared.

• 유체기계공업학회:학술대회논문집
• /
• 2002.12a
• /
• pp.370-377
• /
• 2002

Generally, the system of calculation for the multi-path ultrasonic flow meters can be divided into two methods by how to get the mean velocity, namely, weighting and direct method. Weighting-method derive the mean velocity through modeling in theoretical velocity profile. Direct-method derive the mean velocity though actual flow distribution. The system of calculation varies with maker's transducer configuration and integration method. Each system has merits and demerits. This paper describes the system of integration that calculates line velocity over cross-section of the circular pipe. Flow rate mr discussed in this paper is a difference between theoretical flow rate and integrated flow rate according to values of Reynolds number in symmetric flow field or theoretical flow rate and integrated flow rate according to rotated model in asymmetric flow field.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.1
• /
• pp.49-56
• /
• 2017

In this paper we propose the approach for big data integration so as to analyze, visualize and predict the future of the trend of the market, and that is to get the integration data model using the R language which is the future of the statistics and the Hadoop which is a parallel processing for the data. As four approaching methods using R and Hadoop, ff package in R, R and Streaming as Hadoop utility, and Rhipe and RHadoop as R and Hadoop interface packages are used, and the strength and weakness of four methods are described and analyzed, so Rhipe and RHadoop are proposed as a complete set of data integration model. The integration of R, which is popular for processing statistical algorithm and Hadoop contains Distributed File System and resource management platform and can implement the MapReduce programming model gives us a new environment where in R code can be written and deployed in Hadoop without any data movement. This model allows us to predictive analysis with high performance and deep understand over the big data.