• Journal of the Korean Data and Information Science Society
• /
• 제14권2호
• /
• pp.337-343
• /
• 2003

In this paper we present the prediction interval estimation method using bootstrap method for least squares support vector machine(LS-SVM) regression, which allows us to perform even nonlinear regression by constructing a linear regression function in a high dimensional feature space. The bootstrap method is applied to generate the bootstrap sample for estimation of the covariance of the regression parameters consisting of the optimal bias and Lagrange multipliers. Experimental results are then presented which indicate the performance of this algorithm.

• Journal of Multimedia Information System
• /
• 제8권2호
• /
• pp.131-142
• /
• 2021

This paper traces the process of constructing a new one-dimensional chaotic map, and will provide a simple application in color image encryption. The use of Sarkovskii's theorem will make it possible to determine the existence of chaos and restrict all conditions to ensure the existence of this new sequence. In addition, the sensitivity to initial conditions will be proved by Lyapunov's index value. Similarly, the performance of this new chaotic map will be illustrated graphically and compared with other chaotic maps most commonly used in cryptography. Finally, a humble color image encryption application will show the power of this new chaotic map.

• 대한수학회지
• /
• 제58권5호
• /
• pp.1261-1277
• /
• 2021

An inertial manifold is often constructed as a graph of a function from low Fourier modes to high ones and one used to consider backward bounded (in time) solutions for that purpose. We here show that the proof of the uniqueness of such solutions is crucial in the existence theory of inertial manifolds. Avoiding contraction principle, we mainly apply the Arzela-Ascoli theorem and Laplace transform to prove their existence and uniqueness respectively. A non-self adjoint example is included, which is related to a differential system arising after Kwak transform for Navier-Stokes equations.

• 한국운동역학회지
• /
• 제15권2호
• /
• pp.57-67
• /
• 2005

The purpose of this study was to analyze the kinetic factors of the golf driver swing using the Inverse Dynamics function. For this purpose, joint force were calculated. In order to test the possibility of Inverse Dynamics function(motion-dependent interaction), a triple segmental system was set for wrist, left shoulder and lumbar and joint force working on the anatomical joint region was estimated. For this study, 7 professional golfers were sampled, and then, their driver swings were recorded with two high-speed digital video cameras (180 frames/sec.) to be synthesized into 3-dimensional images and coordinated. Then, Eular's equation was used to produce some kinematic data, which were used to calculate joint force and torque with Newton's function. All data were calculated using LabVIEW 6.1 graphic program. The results of this study can be summarized as follows; It was found that the joint force was generated on wrist, shoulder and lumbar joints in the direction of the target, and that the joint force was stronger in the direction of target immediately before impact. The joint force was generated towards the target to activate the nodes, and then, it was generated in the reverse direction to increase the speed during impact.

• 한국생산제조학회지
• /
• 제23권2호
• /
• pp.118-126
• /
• 2014

In this paper, a new method is proposed for the five-degree-of-freedom precision alignment and stitching of three-dimensional surface-profile data sets. The control parameters for correcting thealignment error are calculated from the surface profile data for overlapped areas among the adjacent measuring areas by using the "least squares method" and "maximum lag position of cross correlation function." To ensure the alignment and stitching reliability, the relationships betweenthe alignment uncertainty, overlapped area, and signal-to-noise level of the measured profile data are investigated. Based on the results of this uncertainty analysis, an appropriate size is proposed for the overlapped area according to the specimen's surface texture and noise level.

• 대한조선학회논문집
• /
• 제45권1호
• /
• pp.10-17
• /
• 2008

In this paper an accurate and stable gridless method that can be applied to multi-dimensional convection problems is developed on a flow directional local grid. A two dimensional pure convection problem is calculated and more accurate and stable solution is obtained compared with other schemes in grid method. The tested numerical schemes include 1st-order upwind scheme, 2nd-order Leith scheme, 3rd-order MUSCL, and QUICK scheme. It is seen that more accurate results are expected when the schemes combined with a MMT control limiter.

• 품질경영학회지
• /
• 제22권4호
• /
• pp.79-101
• /
• 1994

In modern manufacturing, a product consists of many components created by different processes. Variations in the individual component dimensions and in the processes may result in unacceptable final assemblies. Thus, engineers have increased pressure to properly set tolerance specifications for individual components and to control manufacturing processes. When a proper variation stack-up analysis is not performed for all of the components in a functional system, all component parts can be within specifications, but the final assembly may not be functional. Thus, in order to improve the performance of the final assembly, a proper variation stack-up analysis is essential for specifying dimensional tolerances and process control. This research provides a detailed case example of the use of variation stack-up analysis using a Monte Carlo simulation method to improve the defect rate of a complex process, which is the commutator brush track undercut process of an armature assembly of a small motor. Variations in individual component dimensions and process mean shifts cause high defect rate, Since some dimensional characteristics have non-normal distributions and the stack-up function is non-linear, the Monte Carlo simulation method is used.

• 한국언어정보학회:학술대회논문집
• /
• 한국언어정보학회 2007년도 정기학술대회
• /
• pp.430-439
• /
• 2007

In this paper, we use non-negative matrix factorization (NMF) to refine the document clustering results. NMF is a dimensional reduction method and effective for document clustering, because a term-document matrix is high-dimensional and sparse. The initial matrix of the NMF algorithm is regarded as a clustering result, therefore we can use NMF as a refinement method. First we perform min-max cut (Mcut), which is a powerful spectral clustering method, and then refine the result via NMF. Finally we should obtain an accurate clustering result. However, NMF often fails to improve the given clustering result. To overcome this problem, we use the Mcut object function to stop the iteration of NMF.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 하계종합학술대회 논문집(4)
• /
• pp.89-92
• /
• 2002

In this paper we present a barcode reader to decode two-dimensional symbology PDF-417 and propose a novel method to extract the bar-space patterns directly from the gray-level barcode image, which employs the location and the distance between extreme points of each row of the barcode image. This algorithm proves to be very robust from the high convolution distortion environments such as defocussing and warping, even under badly illuminating condition. If the scanned barcode image is a result of the convolution of a Gaussian-shaped point spread function with a bi-level image, popular image segmentation methods such as image thresholding can not distinguish between very narrow bar-space patterns with a couple of pixels wide. The proposed algorithm shows improved performance over current barcode readers.

• 전자공학회논문지S
• /
• 제35S권4호
• /
• pp.118-127
• /
• 1998

In this paper, we propose a voice personality transformation method which makes one person's voice sound like another person's voice. In order to transform the voice personality, vocal tract transfer function is used as a transformation parameter. Comparing with previous methods, the proposed method can obtain high-quality transformed speech with low computational complexity. Conversion between the vocal tract transfer functions is implemented by a linear mapping based on soft clustering. In this process, mean LPC cepstrum coefficients and mean removed LPC cepstrum modeled by the low dimensional vector are used as transformation parameters. To evaluate the performance of the proposed method, mapping rules are generated from 61 Korean words uttered by two male and one female speakers. These rules are then applied to 9 sentences uttered by the same persons, and objective evaluation and subjective listening tests for the transformed speech are performed.