• 한국통신학회논문지
• /
• 제23권9A호
• /
• pp.2333-2344
• /
• 1998

We present a two step approach for estimating the motionand sturcture parameters from region orrespondences in two frames. Given four or more region corresondences on the same planar surface, the motion and planar orientation parameters are first linearly estimated based on second-order approximation of the displacement field of the image plane. Then, using this linear estimate as an initial guess, a nonlinear estimate is obtained by iteratively minimizing an objective function using the exact experession of the displacement field. The objective function involves the centroids of corresponding regions and relationships among low-order moments. Through simulations, we show that the two-step region-based approach gives robust estimates. The performance of nonlinear region-based estimation is compared with that of linear region-based and point-based methods. Experimental results for two image pairs, on esynthetic and one real, ar epresented to show the practical applicability of our approach.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1997년도 하계학술대회 논문집 B
• /
• pp.717-719
• /
• 1997

This paper discusses approximate modelling of discrete-time linear time-varying system(LTVS). The wavelet transform is considered as a tool for representing and approximating a LTVS. The joint time-frequency properties of wave analysis are appropriate for describing the LTVS. Simulatly results is included to illustrate the popential application of the technique.

• 충청수학회지
• /
• 제25권2호
• /
• pp.367-377
• /
• 2012

The goal of the image restoration is to find a good approximation of the original image for the degraded image, the blurring matrix, and the statistics of the noise vector given. Fast truncated Lagrange (FTL) method has been proposed by G. Landi as a image restoration method for large-scale ill-conditioned BTTB linear systems([3]). We implemented FTL method for the image restoration problem with reflective boundary condition which gives better reconstructions of the unknown, the true image.

• 산업기술연구
• /
• 제14권
• /
• pp.19-28
• /
• 1994

일반적으로 2차원 Poisson 방정식을 풀기 위해 유한 차분법을 이용하여 tridiagonal block Toeplitz 선형방정식을 얻는다. 이 선형방정식의 독특한 형태를 활용하기 위해 Lyapunov 방정식으로 변화시킨 다음 이산정현변환(DST)을 이용해서 대각선 행렬로 만들면 계산이 용이해진다. 또 DST는 FFT를 이용해 계산할 수 있으므로 고속 계산이 가능하다. FFT를 병렬로 처리하기 위해 프로세서가 4,096개인 SIMD 컴퓨터 MP-2에서 시뮬레이션했다. 본 논문에서는 알고리즘 유도, 매핑 및 시뮬레이션 결과를 제시했다.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2000년도 춘계학술대회 학술발표 논문집
• /
• pp.130-133
• /
• 2000

In this paper, a new design methodology named FNNN(Fuzzy Polynomial Neural Network) algorithm is proposed to identify the structure and parameters of fuzzy model using PNN(Polynomial Neural Network) structure and a fuzzy inference method. The PNN is the extended structure of the GMDH(Group Method of Data Handling), and uses several types of polynomials such as linear, quadratic and modified quadratic besides the biquadratic polynomial used in the GMDH. The premise of fuzzy inference rules defines by triangular and gaussian type membership function. The fuzzy inference method uses simplified and regression polynomial inference method which is based on the consequence of fuzzy rule expressed with a polynomial such as linear, quadratic and modified quadratic equation are used. Each node of the FPNN is defined as fuzzy rules and its structure is a kind of neuro-fuzzy architecture Several numerical example are used to evaluate the performance of out proposed model. Also we used the training data and testing data set to obtain a balance between the approximation and generalization of proposed model.

• 정보처리학회논문지B
• /
• 제8B권4호
• /
• pp.351-356
• /
• 2001

표준 매개변수 소속 함수(SPMF)에 기반을 둔 구간 선형 변환 방법(PLTM)을 제안한다. 이는 구간 선형 변환 방법을 사용해서 비 매개변수 소속 함수(NPMF)로 표현된 퍼지 집합이 매개변수 소속 함수(PMF)로 표현된 퍼지 집합으로 변환될 수 있다는 생각에서 유래되었다. 이 경우, 이들 매개변수들은 퍼지 집합의 구조를 결정하기 위한 특징점들 이라고 할 수 있다. 결과적으로 구간 선형 변환 방법은 비 매개변수 소속 함수를 매개변수 소속 함수로 변환해 줌으로써 비 매개변수 소속 함수에 기반을 둔 퍼지 시스템과 비교해 볼 때 퍼지 시스템이 상대적으로 빠르게 처리될 수 있게 한다. 한편, 표준 매개변수 소속 함수들의 전형적인 형태가 소개되고 분석된다. 끝으로, PLTM의 전형적인 응용을 제시하고 수치적인 예를 보여준다.

• Nuclear Engineering and Technology
• /
• 제49권1호
• /
• pp.29-42
• /
• 2017

The purpose of the present study is the presentation of the appropriate element and shape function in the solution of the neutron diffusion equation in two-dimensional (2D) geometries. To this end, the multigroup neutron diffusion equation is solved using the Galerkin finite element method in both rectangular and hexagonal reactor cores. The spatial discretization of the equation is performed using unstructured triangular and quadrilateral finite elements. Calculations are performed using both linear and quadratic approximations of shape function in the Galerkin finite element method, based on which results are compared. Using the power iteration method, the neutron flux distributions with the corresponding eigenvalue are obtained. The results are then validated against the valid results for IAEA-2D and BIBLIS-2D benchmark problems. To investigate the dependency of the results to the type and number of the elements, and shape function order, a sensitivity analysis of the calculations to the mentioned parameters is performed. It is shown that the triangular elements and second order of the shape function in each element give the best results in comparison to the other states.

• 한국항만학회지
• /
• 제7권1호
• /
• pp.79-88
• /
• 1993

This study deals with the case of a fixed floating structure(FFS) at the mouth of a rectangular harbor under the action of waves represented by the linear wave theory. Modified forms of the mild-slope equation is applied to the propagation of regular wave over constant water depth. The model is extended to include bottom friction and boundary absorption. A hybrid element approximation is used for calculation of linear wave oscillation in and near coastal harbor. Modification of the model was necessary for the FFS. For the conditions tested, the results of laboratory experiments by Ippen and Goda(1963), and Lee (1969) are compared with the calculated one from this model. The cases of flat cylinderical structures, both fixed and floating, were taken to be in an intermediate water depth.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1994년도 추계학술대회 논문집 학회본부
• /
• pp.356-358
• /
• 1994

In this paper, to alleviate the effect of approximation error and discontinuous variation of the controller parameters, the variable structure control scheme using neural networks is presented. In the proposed method, the variable structure control rules for each local linear models are designed to reject the effect of linearization error caused by linearization of the nonlinear system. And neural network infer approximate controller gains from combination of local linear control gains. The proposed control methods can be used to control nonlinear systems and it has robust characteristic against system parameter variations and external disturbances.

• 한국수학교육학회지시리즈B:순수및응용수학
• /
• 제12권4호
• /
• pp.275-287
• /
• 2005

Geometric invariants are basic tools for geometric processing and computer vision. In this paper, we give a linear approximation for the differentiation of the binormal vector field of a space curve by using the forward and backward differences of discrete binormal vectors. Two kind of discrete torsion, say, back-ward torsion $T_b$ and forward torsion $T_f$ can be defined by the dot product of the (backward and forward) discrete differentiation of binormal vectors that are linear approximations of torsion. Using Frenet formula and Taylor series expansion, we give error estimations for the discrete torsions. We also give numerical tests for a curve. Notably the average of $T_b$ and $T_f$ looks more stable in errors.