• The Journal of the Acoustical Society of Korea
• /
• v.25 no.1
• /
• pp.14-20
• /
• 2006

In this paper. a novel technique to set up the initial weights in BSS of delayed mixtures is proposed. After analyzing Eigendecomposition for the correlation matrix of mixing data. the initial weights are set from the Eigenvectors ith delay information. The Proposed setting of initial weighting method for conventional FDICA technique improved the separation Performance. The computer simulation shows that the Proposed method achieves the improved SIR and faster convergence speed of learning curve.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.7 no.5
• /
• pp.1033-1039
• /
• 2012

In this paper, a simple and robust algorithm is proposed for detecting each steel plate from a image which contains several steel plates. Steel plate is characterized by line edge, so line detection is a fundamental task for analyzing and understanding of steel plate images. To detect the line edge, the proposed algorithm uses the small eigenvalue analysis. The proposed approach scans an input edge image from the top left corner to the bottom right corner with a moving mask. A covariance matrix of a set of edge pixels over a connected region within the mask is determined and then the statistical and geometrical properties of the small eigenvalue of the matrix are explored for the purpose of straight line detection. Using the detected line edges, each plate is determined based on the directional information and the distance information of the line edges. The results of the experiments emphasize that the proposed algorithm detects each steel plate from a image effectively.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.7
• /
• pp.1475-1479
• /
• 2004

High dimensional data requires efficient feature extraction techliques. Though PCA(Principal Component Analysis) is a famous feature extraction method it requires huge memory space and computational cost is high. In this paper we use incremental PCA for feature extraction on high dimensional data. Through experiment we show that proposed method is superior to APEX model.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.5A
• /
• pp.531-541
• /
• 2009

The research analyzes and investigates conventional concrete, hydration heat, set, and mechanical properties by making high flowing self-compacting concretes of binary blend and ternary blend as one of evaluations about the properties of the hydration heat of high flowing self-compacting concrete with a strength of 30, 50, and 70 MPa. In addition, it estimates concrete adiabatic temperatures by calculating a thermal property value of powder obtained by measuring a heat evolution amount for powder used in concrete, a thermal property value of concrete obtained by conducting a simple adiabatic temperature test, and a normal thermal property value of material used in concrete, using a simple equation. Moreover, it analyzes and investigates the hydration heat property of high flowing self-compacting concrete and the thermal stress caused by hydration heat by conducting a 3D temperature stress analysis for the hydration heat and the adiabatic temperature obtained by temperature analysis, using MIDAS CIVIL 06 program.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.4
• /
• pp.125-132
• /
• 2021

As the layers of artificial neural network deepens, and the dimension of data used as an input increases, there is a problem of high arithmetic operation requiring a lot of arithmetic operation at a high speed in the learning and recognition of the neural network (NN). Thus, this study proposes a data dimensionality reduction method to reduce the dimension of the input data in the NN. The proposed Line-segment Feature Analysis (LFA) algorithm applies a gradient-based edge detection algorithm using median filters to analyze the line-segment features of the objects existing in an image. Concerning the extracted edge image, the eigenvalues corresponding to eight kinds of line-segment are calculated, using 3×3 or 5×5-sized detection filters consisting of the coefficient values, including [0, 1, 2, 4, 8, 16, 32, 64, and 128]. Two one-dimensional 256-sized data are produced, accumulating the same response values from the eigenvalue calculated with each detection filter, and the two data elements are added up. Two LFA256 data are merged to produce 512-sized LAF512 data. For the performance evaluation of the proposed LFA algorithm to reduce the data dimension for the recognition of handwritten numbers, as a result of a comparative experiment, using the PCA technique and AlexNet model, LFA256 and LFA512 showed a recognition performance respectively of 98.7% and 99%.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2007.06a
• /
• pp.597-600
• /
• 2007

Face is considered to be one of the biometrics in person identification. But Face recognition is a high dimensional pattern recognition problem. Even low-resolution face images generate huge dimensional feature space. The aim of this paper is to present a fast feature extraction method for real time human face recognition. first, It compute eigen-vector and eigen-value by Principle component analysis on inputed human face image, and propose method of feature extraction that make feature vector by apply gabor filter to computed eigen-vector. And it compute feature value which multiply by made eigen-value. This study simulations performed using the ORL Database.

• Proceedings of the Korean Association of Geographic Inforamtion Studies Conference
• /
• 2010.06a
• /
• pp.298-302
• /
• 2010

This study proposes a differential method to analyze the properties of the topographic surface driven from lidar imagery. Although airborne lidar imagery provides elevation information rapidly, a sequence of extraction processes are needed to acquire semantic information about objects such as terrain, roads, trees, vegetation, and buildings. For the processes, the properties present in a given lidar data need to be analyzed. In order to investigate the geometric characteristics of the surface, this study employs eigenvalues of the Hessian matrix. For experiments, a lidar image containing university campus buildings with the point density of about 1 meter was processed and the results show that the approach is effective to obtain the properties of each land object Surface.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.13 no.3
• /
• pp.59-67
• /
• 1993

The variations of the natural frequencies and the peak response acceleration at the top of prestressed concrete reactor building due to random variability and/or model uncertainty of structural parameters are studied. The results may be used as essential input parameters in seismic probabilistic risk assessment or seismic margin assessment of the reactor building. The sensitivity test of each structural parameter is first performed to determine the most influential parameter upon the natural frequency of structure model. Then Monte Carlo simulation technique is applied to evaluate the effect of parameter variation on the natural frequencies and the peak response acceleration. The acceleration time history is obtained by direct integration scheme. As the study results, it is found that the fundamental natural frequency and the peak response acceleration at the top of the building are most strongly affected by Young's modulus among the structural parameters, in which the value of mean plus one standard deviation obtained by probabilistic approach deviates up to about (+)12% from the result of deterministic method. Considering the uncertainty of flexural rigidity, the structural responses vary in range of (-)4%~(+)14%.

• Journal of the Korean Geosynthetics Society
• /
• v.18 no.3
• /
• pp.69-77
• /
• 2019

This paper presents a new analytical model for estimating the free vibration of tapered friction piles. The governing differential equation for the free vibration of statically axially-loaded piles embedded in non-homogeneous soil is derived. The equation is numerically integrated by the Runge-Kutta method, and then the eigenvalue of natural frequency is determined by the Regula-Falsi scheme. For a cylindrical non-tapered pile, the computed natural frequencies compare well with the available data from literature. Numerical examples are presented to investigate the effects of the tapering, the skin friction resistance, the end condition of the pile, the vertical compressive loading, and the soil non-homogeneity on the natural frequency and mode shape of tapered friction piles.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.8
• /
• pp.868-876
• /
• 2016

MUSIC (multiple signal classification) algorithm is a representative algorithm estimating the angle of arrival using the eigenvalues and eigenvectors. Generally, the eigenvalues and eigenvectors are obtained through the eigen-analysis, but this analysis requires high computational complexity and late convergence time. For this reason, it is almost impossible to construct the real-time system with low-cost using this approach. Even though QR iteration is considered as the eigen-analysis approach to improve these problems, this is inappropriate to apply to the MUSIC algorithm. In this paper, we analyze the problems of conventional method based on the eigenvalues for convergence decision and propose the improved decision algorithm using the eigenvectors.