• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.6A
• /
• pp.602-609
• /
• 2002

In this paper, we propose an robust invisible watermarking method using wavelet transform and singular value decomposition for the ownership protection. of images. For this method, after we decompose the original image in three level using wavelet transform, we use singular value decomposition based key depended watermark insertion method in the lowest band $LL_3.$ And we also watermark using DCT for extraction of watermark and verification of robustness. In the experiments, we found that it had a good quality and robustness in attack such as compression, image processing, geometric transformation and noises. Especially, we know that this method have very high extraction ratio against nose and JPEG compression. And Digimarc's method can not extract watermark in 80 percent compression ratio of JPEG, but the proposed method can extract well.

• ETRI Journal
• /
• v.37 no.4
• /
• pp.824-832
• /
• 2015

Heart sounds are the main obstacle in lung sound analysis. To tackle this obstacle, we propose a diagnosis algorithm that uses singular spectrum analysis (SSA) and frequency features of heart and lung sounds. In particular, we introduce a frequency coefficient that shows the frequency difference between heart and lung sounds. The proposed algorithm is applied to a synthetic mixture of heart and lung sounds. The results show that heart sounds can be extracted successfully and localizations for the first and second heart sounds are remarkably performed. An error analysis of the localization results shows that the proposed algorithm has fewer errors compared to the SSA method, which is one of the most powerful methods in the localization of heart sounds. The presented algorithm is also applied in the cases of recorded respiratory sounds from the chest walls of five healthy subjects. The efficiency of the algorithm in extracting heart sounds from the recorded breathing sounds is verified with power spectral density evaluations and listening. Most studies have used only normal respiratory sounds, whereas we additionally use abnormal breathing sounds to validate the strength of our achievements.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.9
• /
• pp.1041-1043
• /
• 2016

This letter proposes a new blind fractionally spaced equalization (FSE). The conventional linear program (LP) FSE reduces the degree of freedom (DOF) by abandoning many equalization filter taps, which causes severe performance degradations. We use singular value decomposition (SVD) to obtain the signal subspace and to fully utilize all samples for performance improvement. The proposed scheme has similar performance with the nuclear norm minimization and has as low complexity as the LP equalizer.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.1
• /
• pp.108-113
• /
• 2007

The classification of fingerprint images is to classify fingerprint images into varies fingerprint types, it is very important in automatic fingerprint recognition. In this paper, a new singular points detection technique was presented. A direction uniform measure is defined to describe the changes of direction fields in a certain neighborhood of fingerprint images. Singular points can be detected by adopting the measure. It should be pointed out that singular points in accurate positions would be obtained in this ways. And an improved Poincare exponential algorithm is presented to identify core points and triangle points. In this paper. making use of 102 experimental fingerprint images datas and attained 7.8% classification errors. This was better than experimental result of abstract [9]. It is possible to use on-line fingerprint images classification.

• Journal of Internet Computing and Services
• /
• v.25 no.1
• /
• pp.49-56
• /
• 2024

There have been emerging many use-cases applying recommendation systems especially in online platform. Although the performance of recommendation systems is affected by a variety of factors, selecting appropriate features is difficult since most of recommendation systems have sparse data. Conventional matrix factorization (MF) method is a basic way to handle with problems in the recommendation systems. However, the MF based scheme cannot reflect non-linearity characteristics well. As deep learning technology has been attracted widely, a deep neural network (DNN) framework based collaborative filtering (CF) was introduced to complement the non-linearity issue. However, there is still a problem related to feature embedding for use as input to the DNN. In this paper, we propose an effective method using singular value decomposition (SVD) based feature embedding for improving the DNN performance of recommendation algorithms. We evaluate the performance of recommendation systems using MovieLens dataset and show the proposed scheme outperforms the existing methods. Moreover, we analyze the performance according to the number of latent features in the proposed algorithm. We expect that the proposed scheme can be applied to the generalized recommendation systems.

• Journal of applied mathematics & informatics
• /
• v.31 no.1_2
• /
• pp.165-177
• /
• 2013

In this paper, we apply Homotopy perturbation transform method (HPTM) for solving singular fourth order parabolic partial differential equations with variable coefficients. This method is the combination of the Laplace transform method and Homotopy perturbation method. The nonlinear terms can be easily handled by the use of He's polynomials. The aim of using the Laplace transform is to overcome the deficiency that is mainly caused by unsatisfied conditions in other semi-analytical methods such as Homotopy perturbation method (HPM), Variational iteration method (VIM) and Adomain Decomposition method (ADM). The proposed scheme finds the solutions without any discretization or restrictive assumptions and avoids the round-off errors. The comparison shows a precise agreement between the results and introduces this method as an applicable one which it needs fewer computations and is much easier and more convenient than others, so it can be widely used in engineering too.

• Journal of applied mathematics & informatics
• /
• v.32 no.1_2
• /
• pp.185-201
• /
• 2014

In this paper, we construct a numerical method to solve singularly perturbed one-dimensional parabolic convection-diffusion problems. We use Euler method with uniform step size for temporal discretization and exponential-spline scheme on spatial uniform mesh of Shishkin type for full discretization. We show that the resulting method is uniformly convergent with respect to diffusion parameter. An extensive amount of analysis has been carried out to prove the uniform convergence with respect to the singular perturbation parameter. The obtained numerical results show that the method is efficient, stable and reliable for solving convection-diffusion problem accurately even involving diffusion parameter.

• Honam Mathematical Journal
• /
• v.38 no.3
• /
• pp.661-674
• /
• 2016

In [15] they introduced a new finite element method for accurate numerical solutions of Poisson equations with corner singularities, which is useful for the problem with known stress intensity factor. They consider the Poisson equations with homogeneous Dirichlet boundary condition, compute the finite element solution using standard FEM and use the extraction formula to compute the stress intensity factor, then they pose a PDE with a regular solution by imposing the nonhomogeneous boundary condition using the computed stress intensity factor, which converges with optimal speed. From the solution we could get accurate solution just by adding the singular part. This approach works for the case when we have the accurate stress intensity factor. In this paper we consider Poisson equations with mixed boundary conditions and show the method depends the accrucy of the stress intensity factor by considering two algorithms.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.101.1-101
• /
• 2001

This paper describes a grid mapping for a vision based mobile robot in uncertain indoor environment. The map building is a prerequisite for navigation of a mobile robot and the problem of feature correspondence across two images is well known to be of crucial Importance for vision-based mapping We use a stereo matching algorithm obtained by singular value decomposition of an appropriate correspondence strength matrix. This new correspondence strength means a correlation weight for some local measurements to quantify similarity between features. The visual range data from the reconstructed disparity image form an occupancy grid representation. The occupancy map is a grid-based map in which each cell has some value indicating the probability at that location ...

• Journal of Mechanical Science and Technology
• /
• v.17 no.2
• /
• pp.269-279
• /
• 2003

The anti-plane shear problem of bonded elastic materials containing a crack at an arbitrary angle to the graded interfacial zone is investigated in this paper The interfacial zone is modeled as a nonhomogeneous interlayer of finite thickness with the continuously varying shear modulus between the two dissimilar, homogeneous half-planes. Formulation of the crack problem is based upon the use of the Fourier integral transform method and the coordinate transformations of basic field variables. The resulting Cauchy-type singular integral equation is solved numerically to provide the values of mode 111 stress intensity factors. A comprehensive parametric study is then presented of the influence of crack obliquity on the stress intensity factors for different crack size and locations and for different material combinations, in conjunction with the material nonhomogeneity within the graded interfacial zone.