• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2005년도 한국컴퓨터종합학술대회 논문집 Vol.32 No.1 (B)
• /
• pp.748-750
• /
• 2005

Isomap [1] is a manifold learning algorithm, which extends classical multidimensional scaling (MDS) by considering approximate geodesic distance instead of Euclidean distance. The approximate geodesic distance matrix can be interpreted as a kernel matrix, which implies that Isomap can be solved by a kernel eigenvalue problem. However, the geodesic distance kernel matrix is not guaranteed to be positive semidefinite. In this paper we employ a constant-adding method, which leads to the Mercer kernel-based Isomap algorithm. Numerical experimental results with noisy 'Swiss roll' data, confirm the validity and high performance of our kernel Isomap algorithm.

• Communications for Statistical Applications and Methods
• /
• 제16권2호
• /
• pp.383-388
• /
• 2009

In this paper we propose a variance function estimation method based on kernel trick for replicated data or data consisted of sample variances. Newton-Raphson method is used to obtain associated parameter vector. Furthermore, the generalized approximate cross validation function is introduced to select the hyper-parameters which affect the performance of the proposed variance function estimation method. Experimental results are then presented which illustrate the performance of the proposed procedure.

• 전기의세계
• /
• 제29권10호
• /
• pp.660-666
• /
• 1980

The finite leement method is a effetive analysis technique for obtaining approximate solutions of continuum problems with boundary conditions. This paper deals with the programming for the application of this method and the preciser analysis of the magnetic field in the air gap of the linear stepper motor by the method. The finite element analysis based on the variational principle is adopted and the computer program for reducing input data and a large number of the memory words required by the system matrix is presented. The 2-dimensional analysis of the air gap is made and several cases according to varying the position are considered.

• 산업공학
• /
• 제4권2호
• /
• pp.43-51
• /
• 1991

A line search method is developed to locate all the conerpoints of 2-dimensional polygon images for inspection purposes. This optimization-based method is used to approximate a 2-D curved object by a polygon. This scheme is also developed for inspection of objects in industrial environment. The inspection includes dimensional verification and pattern matching which compares a 2-D image of an object to a pattern image. The method proves to be computationally efficient and accurate for real time application.

• International Journal of CAD/CAM
• /
• 제6권1호
• /
• pp.139-148
• /
• 2006

Catmull-Clark subdivision scheme provides a powerful method for building smooth and complex surfaces. But the number of faces in the uniformly refined meshes increases exponentially with respect to subdivision depth. Adaptive tessellation reduces the number of faces needed to yield a smooth approximation to the limit surface and, consequently, makes the rendering process more efficient. In this paper, we present a new adaptive tessellation method for general Catmull-Clark subdivision surfaces. Different from previous control mesh refinement based approaches, which generate approximate meshes that usually do not interpolate the limit surface, the new method is based on direct evaluation of the limit surface to generate an inscribed polyhedron of the limit surface. With explicit evaluation of general Catmull-Clark subdivision surfaces becoming available, the new adaptive tessellation method can precisely measure error for every point of the limit surface. Hence, it has complete control of the accuracy of the tessellation result. Cracks are avoided by using a recursive color marking process to ensure that adjacent patches or subpatches use the same limit surface points in the construction of the shared boundary. The new method performs limit surface evaluation only at points that are needed for the final rendering process. Therefore it is very fast and memory efficient. The new method is presented for the general Catmull-Clark subdivision scheme. But it can be used for any subdivision scheme that has an explicit evaluation method for its limit surface.

• Journal of Mechanical Science and Technology
• /
• 제20권11호
• /
• pp.1961-1971
• /
• 2006

A soft recovery system (SRS) is a device that stops a high speed projectile without damaging the projectile. The SRS is necessary to verify the shock resistant requirements of microelectronics and electro-optic sensors in smart munitions, where the projectiles experience over 20,000 g acceleration inside the barrel. In this study, a computer code for the performance evaluation of a SRS based on ballistic compression decelerator concept has been developed. It consists of a time accurate compressible one-dimensional Euler code with use of deforming grid and a projectile motion analysis code. The Euler code employs Roe's approximate Riemann solver with a total variation diminishing (TVD) method. A fully implicit dual time stepping method is used to advance the solution in time. In addition, the geometric conservation law (GCL) is applied to predict the solutions accurately on the deforming mesh. The equation of motion for the projectile is solved with the four-stage Runge-Kutta time integration method. A small scale SRS to catch a 20 mm bullet fired at 500 m/s within 1,600 g-limit has been designed with the proposed method.

• 한국컴퓨터정보학회논문지
• /
• 제14권5호
• /
• pp.29-36
• /
• 2009

서열데이터베이스에 있는 자주 발현하는 부분 서열을 패턴으로 찾아내는 순차패턴 탐색은 넓은 응용 분야를 가지는 중요한 데이터 마이닝 문제이다. DNA 서열에서 순차패턴이 모티프가 될 수 있으므로 DNA 서열에서 순차패턴을 찾는 것을 연구하였다. 대부분의 기존 마이닝 방법은 순차패턴의 정의에 따라 정확한 정합에 주력하여 노이즈가 있는 환경이나 실제 문제에서 발생하는 부정확한 데이터에 대하여 제대로 작동하지 않을 수 있다. 이러한 문제가 생물 데이터인 DNA 서열에서 자주 나타난다. 이러한 문제를 다루기 위한 근사 정합 방법을 연구하였다. 본 연구의 아이디어는 자주 발생하는 패턴을 근사 패턴이라 부르는 그룹으로 분류할 수 있다는 관찰에서 기반을 둔다. 기존의 Prefixspan 알고리즘은 주어진 긴 서열에서 순차패턴을 잘 찾을 수 있다. 본 연구는 Prefixspan 알고리즘을 개선하여 유사한 순차패턴을 찾을 수 있게 하였다. 실험 결과는 PreFixSpan보다 제안한 방법이 패턴 길이가 4일 때, 근사 순차패턴의 빈도가 5배 높아짐을 보였다.

• IEIE Transactions on Smart Processing and Computing
• /
• 제6권4호
• /
• pp.229-236
• /
• 2017

This paper presents an efficient object-based color image-retrieval algorithm that is suitable for the classification and retrieval of images from small to mid-scale datasets, such as images in PCs, tablets, phones, and cameras. The proposed method first finds salient regions by using regional feature vectors, and also finds several dominant colors in each region. Then, each salient region is partitioned into small sub-blocks, which are assigned 1 or 0 with respect to the number of pixels corresponding to a dominant color in the sub-block. This gives a binary map for the dominant color, and this process is repeated for the predefined number of dominant colors. Finally, we have several binary maps, each of which corresponds to a dominant color in a salient region. Hence, the binary maps represent the spatial distribution of the dominant colors in the salient region, and the union (OR operation) of the maps can describe the approximate shapes of salient objects. Also proposed in this paper is a matching method that uses these binary maps and which needs very few computations, because most operations are binary. Experiments on widely used color image databases show that the proposed method performs better than state-of-the-art and previous color-based methods.

• Journal of applied mathematics & informatics
• /
• 제8권2호
• /
• pp.305-326
• /
• 2001

We introduce a class of multilevel recursive incomplete LU preconditioning techniques (RILUM) for solving general sparse matrices. This techniques is based on a recursive two by two block incomplete LU factorization on the coefficient martix. The coarse level system is constructed as an (approximate) Schur complement. A dynamic preconditioner is obtained by solving the Schur complement matrix approximately. The novelty of the proposed techniques is to solve the Schur complement matrix by a preconditioned Krylov subspace method. Such a reduction process is repeated to yield a multilevel recursive preconditioner.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 1993년도 봄 학술발표회논문집
• /
• pp.125-133
• /
• 1993

Preflex composit girder is intended for a better use on both steel and concrete by introducting prestress into the lower flange concrete with preflection. In Korea, recently preflex bridges are widely used especially for urban construction but the design method depends on the conventional ASD(Allowable Stress Design). This paper suggests an optimization model for the design of preflex composite bridges based on LIFD(Load Resistance Factor Design). The optimization algorithm adopted for the NLP model proposed in the paper is the FTM(Flexible Tolerance Method) which is very efficient for the approximate continuous optimization. For the discrete optimum results, a pesudo discrete optimization is used for the economical round-up to the available dimensions. The economic effectiveness of optimum design based on the LRFD method is investigation by comparing the results with those of the ASD method. Based on applications to the actual design examples, it may be concluded that the optimization model suggested in the paper provides economical but reliable design. And the suggested in the paper provides economical but reliable design. And the computer code for the automatic optimum design of preflex bridges developed in the paper for a CAD system may be successfully used in practice.