• 제목/요약/키워드: and regularization

검색결과 458건 처리시간 0.033초

A Simulation Study on Regularization Method for Generating Non-Destructive Depth Profiles from Angle-Resolved XPS Data

  • Ro, Chul-Un
    • 분석과학
    • /
    • 제8권4호
    • /
    • pp.707-714
    • /
    • 1995
  • Two types of regularization method (singular system and HMP approaches) for generating depth-concentration profiles from angle-resolved XPS data were evaluated. Both approaches showed qualitatively similar results although they employed different numerical algorithms. The application of the regularization method to simulated data demonstrates its excellent utility for the complex depth profile system. It includes the stable restoration of the depth-concentration profiles from the data with considerable random error and the self choice of smoothing parameter that is imperative for the successful application of the regularization method. The self choice of smoothing parameter is based on generalized cross-validation method which lets the data themselves choose the optimal value of the parameter.

  • PDF

A Regularization-direct Method to Numerically Solve First Kind Fredholm Integral Equation

  • Masouri, Zahra;Hatamzadeh, Saeed
    • Kyungpook Mathematical Journal
    • /
    • 제60권4호
    • /
    • pp.869-881
    • /
    • 2020
  • Most first kind integral equations are ill-posed, and obtaining their numerical solution often requires solving a linear system of algebraic equations of large condition number, which may be difficult or impossible. This article proposes a regularization-direct method to numerically solve first kind Fredholm integral equations. The vector forms of block-pulse functions and related properties are applied to formulate the direct method and reduce the integral equation to a linear system of algebraic equations. We include a regularization scheme to overcome the ill-posedness of integral equation and obtain a stable numerical solution. Some test problems are solved using the proposed regularization-direct method to illustrate its efficiency for solving first kind Fredholm integral equations.

THE METHOD OF REGULARIZATION RATIOS APPLIED TO RECONSTRUCTIONS OF ELASTIC RIGID OBSTACLES VIA THE FACTORIZATION METHOD

  • Kim, K.;Leem, K.H.;Pelekanos, G.
    • East Asian mathematical journal
    • /
    • 제32권1호
    • /
    • pp.129-138
    • /
    • 2016
  • In this paper, we propose an efficient regularization technique (The Method of Regularized Ratios) for the reconstruction of the shape of a rigid elastic scatterer from far field measurements. The approach used is based on the factorization method and creates via Picard's condition ratios, baptized Regularized Ratios, that serve to effectively remove unwanted singular values that may lead to poor reconstructions. This is achieved through the use of a sophisticated algorithm that progressively adjusts an initially set moderate tolerance. In comparison with the well established Tikhonov-Morozov regularization techniques our new algorithm appears to be more computationally efficient as it doesn't require computation of the regularization parameter for each point in the grid.

식생여과대 유사 저감 효율 산정을 위한 정규화 방안 (A Study on Regularization Methods to Evaluate the Sediment Trapping Efficiency of Vegetative Filter Strips)

  • 배주현;한정호;양재의;김종건;임경재;장원석
    • 한국농공학회논문집
    • /
    • 제61권6호
    • /
    • pp.9-19
    • /
    • 2019
  • Vegetative Filter Strip (VFS) is the best management practice which has been widely used to mitigate water pollutants from agricultural fields by alleviating runoff and sediment. This study was conducted to improve an equation for estimating sediment trapping efficiency of VFS using several different regularization methods (i.e., ordinary least squares analysis, LASSO, ridge regression analysis and elastic net). The four different regularization methods were employed to develop the sediment trapping efficiency equation of VFS. Each regularization method indicated high accuracy in estimating the sediment trapping efficiency of VFS. Among the four regularization methods, the ridge method showed the most accurate results according to $R^2$, RMSE and MAPE which were 0.94, 7.31% and 14.63%, respectively. The equation developed in this study can be applied in watershed-scale hydrological models in order to estimate the sediment trapping efficiency of VFS in agricultural fields for an effective watershed management in Korea.

Impact Force Reconstruction of Composite materials based on Improved Regularization Technology

  • Sun, Yajie;Yin, Tao;Yang, Jian;Cai, Zhiyu;Wu, Shaoen
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • 제15권8호
    • /
    • pp.2718-2731
    • /
    • 2021
  • In the structural health monitoring of composite materials, in order to solve the ill-posed problem of impact force reconstruction, regularization techniques are often used to deal with it. Due to the poor convergence of the traditional Tikhonov regularization method, in order to accurately reconstruct the time history of the impact force, this paper improves Tikhonov regularization method and constructs homotopy function with strong convergence. Since the optimal regularization parameters need to be found in the homotopy function, the Newton downhill method is used to find the optimal parameters and the homotopy function can be calculated, which can accurately reconstruct the time history of the impact force. In order to verify the universality of the method in this paper, impact hammers of different materials were used in the experiment in this paper to study and compare the reconstruction effect of impact time history of different impact hammers.

A Spatial Regularization of LDA for Face Recognition

  • Park, Lae-Jeong
    • International Journal of Fuzzy Logic and Intelligent Systems
    • /
    • 제10권2호
    • /
    • pp.95-100
    • /
    • 2010
  • This paper proposes a new spatial regularization of Fisher linear discriminant analysis (LDA) to reduce the overfitting due to small size sample (SSS) problem in face recognition. Many regularized LDAs have been proposed to alleviate the overfitting by regularizing an estimate of the within-class scatter matrix. Spatial regularization methods have been suggested that make the discriminant vectors spatially smooth, leading to mitigation of the overfitting. As a generalized version of the spatially regularized LDA, the proposed regularized LDA utilizes the non-uniformity of spatial correlation structures in face images in adding a spatial smoothness constraint into an LDA framework. The region-dependent spatial regularization is advantageous for capturing the non-flat spatial correlation structure within face image as well as obtaining a spatially smooth projection of LDA. Experimental results on public face databases such as ORL and CMU PIE show that the proposed regularized LDA performs well especially when the number of training images per individual is quite small, compared with other regularized LDAs.

볼록 규준화 RLS의 규준화 상수를 정하기 위한 두 가지 방법과 희소성 음향 통신 채널 추정 성능 비교 (Two regularization constant selection methods for recursive least squares algorithm with convex regularization and their performance comparison in the sparse acoustic communication channel estimation)

  • 임준석;홍우영
    • 한국음향학회지
    • /
    • 제35권5호
    • /
    • pp.383-388
    • /
    • 2016
  • 본 논문은 볼록 규준화 RLS(Recursive Least Squares)에 쓰이는 규준화를 위한 상수를 정하는 법을 제안한다. Eksioglu와 Tanc는 희소성 음향 채널 추정을 위해서 볼록 규준화 RLS 알고리즘을 구현하였다. 그러나 이 알고리즘은 더 좋은 추정 성능을 위해서 채널의 참 임펄스 응답 정보가 사용된다. 본 논문에서는 이 같은 참 임펄스 응답 정보가 필요 없는 규준화 상수를 위한 두 가지 선정법을 제안한다. 그리고 제안한 방법을 사용했을 때 Eksioglu와 Tanc의 방법에 필적한 추정 성능을 유지함을 보인다.

전기 임피던스 단층촬영법에서 잔류오차 기반의 반복적 조정기법을 이용한 영상 복원 (Image Reconstruction Using Iterative Regularization Scheme Based on Residual Error in Electrical Impedance Tomography)

  • 강숙인;김경연
    • 전기전자학회논문지
    • /
    • 제18권2호
    • /
    • pp.272-281
    • /
    • 2014
  • 전기 임피던스 단층촬영법을 이용한 정적 영상 복원에서 대표적으로 사용되고 있는 복원 알고리즘은 modified Newton-Raphson(mNR) 알고리즘으로 수렴 속도 및 추정 정확도 측면에서 비교적 다른 알고리즘들에 비해 좋은 성능을 나타낸다. mNR 알고리즘에서는 측정 전압과 계산 전압과의 차이, 즉 잔류오차를 최소화하도록 목적함수를 설정하고 이를 반복 연산하여 내부의 저항률 분포를 추정한다. 이때 EIT 역문제의 비정치성을 완화시키기 위해 조정방법을 사용하며 조정인자에 따라 서로 다른 영상 복원 성능을 나타낸다. 기존 기법에서는 반복 연산마다 일정한 상수 값의 조정인자를 사용하기 때문에 대상 물체의 내부 상태가 변하거나 측정 잡음 등이 있는 경우 때때로 조정인자에 따라 영상 복원이 수렴되지 않는다. 따라서 본 논문에서는 영상 복원 수렴 및 성능을 개선하기 위하여 잔류오차에 기반하여 반복 연산마다 자동적으로 조정인자를 수정하는 기법을 제안하였다. 시뮬레이션과 실험을 수행하여 제안된 기법의 영상 복원성능을 평가한 결과 비교적 양호한 성능을 나타내었다.

격자 정방형화 방법을 이용한 박판 성형해석의 효율개선 (Efficiency enhancement of sheet metal forming analysis with a mesh regularization method)

  • 윤종헌;허훈
    • 한국소성가공학회:학술대회논문집
    • /
    • 한국소성가공학회 2003년도 춘계학술대회논문집
    • /
    • pp.339-342
    • /
    • 2003
  • This paper newly proposes a mesh regularization method for the enhancement of the efficiency in sheet metal forming analysis. The regularization method searches for distorted elements with appropriate searching criteria and constructs patches including the elements to be modified. Each patch is then extended to a three-dimensional surface in order to obtain the information of the continuous coordinates. In constructing the surface enclosing each patch, NURBS(Non-Uniform Rational B-Spline) surface is employed to describe a three-dimensional free surface. On the basis of the constructed surface, each node is properly arranged to form unit elements as close as to a square. The analysis results with the proposed method are compared to the results from the direct forming analysis without mesh regularization in order to confirm the validity of the method.

  • PDF

격자 정방형화 방법을 이용한 박판 성형해석의 효율개선 (Efficiency Enhancement in Sheet Metal Forming Analysis with a Mesh Regularization Method)

  • 윤종헌;허훈
    • 소성∙가공
    • /
    • 제12권4호
    • /
    • pp.401-407
    • /
    • 2003
  • This paper newly proposes a mesh regularization method for the enhancement of the efficiency in sheet metal forming analysis. The regularization method searches for distorted elements with appropriate searching criteria and constructs patches including the elements to be modified. Each patch is then extended to a three-dimensional surface in order to obtain the information of the continuous coordinates. In constructing the surface enclosing each patch, NURBS(Non-Uniform Rational B-Spline) surface is employed to describe a three-dimensional free surface. On the basis of the constructed surface, each node is properly arranged to form unit elements as close as to a square. The state variables calculated from its original mesh geometry are mapped into the new mesh geometry for the next stage or incremental step of a forming analysis. The analysis results with the proposed method are compared to the results from the direct forming analysis without mesh regularization in order to confirm the validity of the method.