• 지구물리와물리탐사
• /
• 제17권2호
• /
• pp.95-103
• /
• 2014

MT 자료에서 3차원 전기비저항 구조와 정적효과를 동시에 구하기 위한 실용적인 역산법을 소개한다. 이 방법은 감도행렬이 필요한 Gauss-Newton법을 기반으로 하고 반복과정에서 Broyden의 방식으로 감도를 수정하는 것을 기본으로 하고 있다. 이 논문에서는 합성 MT 자료에 대한 역산실험을 통해 근사역산법의 성능과 정적효과에 대한 가중치에 대해 검토하였다. 해석적으로 구해지는 초기감도를 Broyden의 방식으로 수정하는 역산법은 초기감도를 끝까지 쓰는 역산법보다 자료오차를 줄이는데 효과적이었다. 그리고 완전한 감도행렬을 반복 중간에서 단 한번만 사용하는 근사역산법으로서는 반복 전반부에서 완전한 감도를 사용할 때 자료오차를 가장 많이 줄이는 것으로 나타났다. 정적효과에 대한 가중치는 어느 특정 한계값 이하로 선택하면 최종 자료오차에는 결정적인 영향을 주지 않는다. 합성 MT 자료에 대한 실험 결과 이 역산법은 정적효과가 포함된 MT 자료로부터 3차원 전기비저항 구조를 재현하는데 효과적임을 확인하였다.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제4권1호
• /
• pp.79-92
• /
• 2000

We investigate numerical properties of Newton's algorithm for improving an eigenpair of a real matrix A using only fixed precision arithmetic. We show that under natural assumptions it produces an eigenpair of a componentwise small relative perturbation of the data matrix A.

• Journal of applied mathematics & informatics
• /
• 제39권3_4호
• /
• pp.295-302
• /
• 2021

In this paper, we propose a new iterative algorithm to automatically prove the existence of solutions for a unilateral boundary value problems for second order equations.

• 한국멀티미디어학회논문지
• /
• 제16권11호
• /
• pp.1338-1347
• /
• 2013

Face recognition is a science of automatically identifying individuals based their unique facial features. In order to avoid overfitting and reduce the computational reduce the computational burden, a new face recognition algorithm using PCA-fisher linear discriminant (PCA-FLD) and fuzzy radial basis function neural network (RBFNN) is proposed in this paper. First, face features are extracted by the principal component analysis (PCA) method. Then, the extracted features are further processed by the Fisher's linear discriminant technique to acquire lower-dimensional discriminant patterns, the processed features will be considered as the input of the fuzzy RBFNN. As a widely applied algorithm in fuzzy RBF neural network, BP learning algorithm has the low rate of convergence, therefore, an improved learning algorithm based on Levenberg-Marquart (L-M) for fuzzy RBF neural network is introduced in this paper, which combined the Gradient Descent algorithm with the Gauss-Newton algorithm. Experimental results on the ORL face database demonstrate that the proposed algorithm has satisfactory performance and high recognition rate.

• 정보처리학회논문지B
• /
• 제11B권3호
• /
• pp.381-386
• /
• 2004

본 논문에서는 첨도가 추가된 뉴우턴법의 고정점 알고리즘에 의한 독립성분분석을 제안하였다. 여기서 첨도의 추가는 유사한 속성을 가지는 성분의 군집화된 분석순서를 얻기 위함이고, 뉴우턴법의 고정점 알고리즘은 성분의 빠른 분석과 우수한 분석성능을 얻기 위함이다. 제안된 독립성분분석을 500개 샘플을 가지는 6개의 혼합신호와 $512\times512$ 픽셀을 가지는 8개의 혼합영상의 분리에 각각 적용하여 실험한 결과, 제안된 기법은 항상 일정한 분석순서를 유지하여 기존의 기법에서 알고리즘의 수행 때마다 랜덤하게 변하는 분석순서의 제약을 해결할 수 있었다. 특히 군집화의 속성을 가진 제안된 독립성분분석은 신호나 영상의 분류나 식별에도 적용할 수 있음을 확인하였다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 하계종합학술대회 논문집 V
• /
• pp.2581-2584
• /
• 2003

In this paper, we propose a self-localization algorithm using vertical line segments. Indoor environment is consist of horizontal and vertical line features such as doors, furniture, and so on. From the input image, vertical line edges are detected by an edge operator, Then, line segments are obtained by projecting edge image vertically and detecting local maximum from the projected histogram. From the relation of horizontal position of line segments and the location of the robot, nonlinear equations are come out Localization is done by solving the equations by using Newton's method. Experimental results show that the proposed algorithm using one camera is simple and applicable to indoor environment.

• Journal of applied mathematics & informatics
• /
• 제3권2호
• /
• pp.253-264
• /
• 1996

The generalized logit model of nominal type with random regressors is studied for bootstrapping. We assess the accuracy of some estimators for our generalized logit model using a Monte Carlo simu-lation. That is we study the finite sample properties containing the consistency and asymptotic normality of the maximum likelihood es-timators. Also we compare Newton Raphson algorithm with BHHH algorithm.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1988년도 추계학술대회 논문집 학회본부
• /
• pp.135-139
• /
• 1988

This paper presents a new algorithm using power flow solution which is given by the polar form Newton-Raphson method in a transient stability analysis. The computation time to solve network equations can be much saved by a decoupled power flow method. In addition, the time is much saved in performing a approximate stability analysis by linearizing the differential equations and using a voltage and angle sensitivity matrix given in network equations.

• Journal of applied mathematics & informatics
• /
• 제20권1_2호
• /
• pp.595-602
• /
• 2006

Determinations of conformal map from the unit disk onto a Jordan region are reduced to solve the Theodorsen equation which is an integral equation for the boundary correspondence function. Among numerical conformal maps the Wegmann's method is well known as a Newton efficient one for solving Theodorsen equation. However this method has not so wide class of convergence. We proposed as an improved method for convergence by applying a low frequency filter to the Wegmann's method. In this paper, we investigate error analysis and propose an automatic algorithm based on this analysis.

• Communications for Statistical Applications and Methods
• /
• 제19권1호
• /
• pp.157-168
• /
• 2012

The EM algorithm is the most important tool to obtain the maximum likelihood estimator in finite mixture models due to its stability and simplicity. However, its convergence rate is often slow because the conventional EM algorithm is based on a large missing data space. Several techniques have been proposed in the literature to reduce the missing data space. In this paper, we review existing methods and propose a new EM algorithm for Gaussian mixtures, which reduces the missing data space while preserving the stability of the conventional EM algorithm. The performance of the proposed method is evaluated with other existing methods via simulation studies.