• 제목/요약/키워드: Normal matrices

검색결과 104건 처리시간 0.022초

자료별 분류분석(DDA)에 의한 특징추출 (Datawise Discriminant Analysis For Feature Extraction)

  • 박명수;최진영
    • 한국지능시스템학회논문지
    • /
    • 제19권1호
    • /
    • pp.90-95
    • /
    • 2009
  • 본 논문은 선형차원감소(Linear Dimensionality Reduction)을 위해 널리 이용되고 있는 특징추출 알고리듬인 선형판별분석(Linear Discriminant Analysis)의 문제점을 해결할 수 있는 새로운 특징추출 알고리듬을 제안한다. 선형판별분석에 포함되는 평균-자료 간 거리 및 평균-평균 간의 거리에 기반한 분산행렬은 역행렬 연산, 계수의 제한 등으로 인하여 계산상의 문제와 추출되는 특징의 수가 제한되는 한계를 가지고 있다. 또한 자료의 집단이 단일 모드의 정규 분포로부터 얻어진 것으로 가정되며 그렇지 않은 경우에 대해서는 적절한 결과를 얻을 수 없다. 본 논문에서는 자료-자료 간의 거리에 기반하고 적절하게 가중치가 추가된 새로운 행렬을 정의하였으며. 이에 기반하여 특징을 추출하는 방법을 제안하였다. 그럼으로써 앞서 선형판별분석의 여러 문제를 해결하고자 시도하였다. 제안된 방법의 성능을 실험을 통해 확인하였다.

가토에서 하악골 신연 양에 따른 하치조신경의 조직학적 변화 (HISTOLOGIC CHANGES OF THE INFERIOR ALVEOLAR NERVE ACCORDING TO THE AMOUNT OF THE MANDIBULAR LENGTHENING IN RABBITS)

  • 김기영;유선열
    • Maxillofacial Plastic and Reconstructive Surgery
    • /
    • 제20권3호
    • /
    • pp.250-255
    • /
    • 1998
  • The purpose of this study was to evaluate the histologic change of the inferior alveolar nerve according to distraction amount following mandibular lengthening. Seven rabbits weighing about 2 kg were used. Corticotomy was performed on the mandibular body anterior to the right first premolar region and unilateral external fixation device was placed. Every effort was made to preserve the inferior alveolar nerve during the corticotomy. The rabbits were then allowed to heal for 7 days without distraction of the device. The mandible was lengthened 0.36 mm/day, 0.76 mm/day, or 1.0 mm/day. Corticotomy and lengthening of mandible were not performed in control group. After the completion of the lengthening process, a 14-day-consolidation period was allowed. After consolidation, rabbits were sacrificed, and histologic examination of the inferior alveolar nerve was performed. The results obtained were as follows : 1. In the control group, normal trifascicular pattern of inferior alveolar nerve was observed. Epineurium, perineurium, endoneurium, and axon with myelin sheath were observed in normal appearance. 2. In 0.36 mm/day distraction group, the trifascicular pattern was normally shown, and there was no destruction in epineurium, perineurium, and endoneurium. The mild changes including myelin attenuation, axoplasmic swelling and darkening were observed. 3. In 0.72 mm/day distraction group, it was possible to differentiate the epineurium from the perineurium. Two normal fascicles and one injuried fascicle were observed with a partially destructed perineurium. Most of the axons had axoplasmic swelling and darkening. 4. In 1 mm/day distraction group, it was difficult to differentiate the nerve structures such as fascicles, epineurium, perineurium, and endoneurium. The axons were severely destroyed, except few which showed decreases in size and changes in shape. Some collagen matrices were observed around the axons. These results suggest that the higher the distraction amount, the more severe the injury to the inferior alveolar nerve, fascicles, axons. Although distraction osteogenesis may be useful, the amount of distraction should be carefully selected.

  • PDF

일변량 공간 연관성 측도의 통계적 검정을 위한 일반화된 고차 적률 추출 절차: 정규성 가정의 경우 (A Generalized Procedure to Extract Higher Order Moments of Univariate Spatial Association Measures for Statistical Testing under the Normality Assumption)

  • 이상일
    • 대한지리학회지
    • /
    • 제43권2호
    • /
    • pp.253-262
    • /
    • 2008
  • 이 논문의 주요 목적은 정규성 가정 하에 일변량 공간 연관성 측도의 첫 번째 네 적률을 구해내는 일반화된 추출 절차를 정식화하고, 그것을 바탕으로 각 측도의 가설 검정을 위해 정규근사가 갖는 가능성과 한계를 평가하는 것이다. 중요 연구 결과는 다음과 같다. 첫째, 이전의 연구에 기반함으로써, 정규성 가정 하에 전역적 측도와 국지적 측도에 모두 적용될 수 있는 일반화된 적률 추출절차가 도출되었다. 개별 공간 연관성 측도를 위한 필수적인 메트릭스가 적절히 정의되었을 때, 일반화된 유의성 검정 방법은 각 공간 연관성 측도의 기대값과 분산은 물론 첨도와 왜도를 효과적으로 산출하였다. 둘째, 첫 번째 두 적률에 근거한 정규근사 방법은 전역적 통계량에 대해서는 유효한 것으로 판명되었지만, 국지적 통계량에 대해서는 매우 높은 왜도와 첨도로 말미암아 그 유효성이 현저히 떨어지는 것으로 드러났다.

측정된 S-파라미터에서 MESFET과 HEMT의 기생 저항을 구하는 새로운 방법 (A New Method for Determination the Parasitic Extrinsic Resistances of MESFETs and HEMTs from the Meaured S-parameters under Active Bias)

  • 임종식;김병성;남상욱
    • 한국전자파학회논문지
    • /
    • 제11권6호
    • /
    • pp.876-885
    • /
    • 2000
  • 추가적인 DC 측정이나 반복 계산법 또는 최적화 방법에 의존하지 않고도, 정상적인 바이어스(Normal activebias) 조건에서 측정하 S-파라미터로부터 MESFET과 HEMT의 외부 기생 저항을 간단히 구할 수 있는 방법이 제시되었다. 이를 위해서 zero 바이어스 조건에서 측정한 Z-파라미터로부터 Rs와 Rd의 차이를 구할수 있다는 사실이 이용된다. 측정한 S-파라미터로부터 외부 기생 인덕터와 캐패시터의 효과를 제거하면, 내부 소자와 외부 기생 저항을 포함한 새로운 소자를 정의할 수 있다. 내부 소자의 Y-파라미터인 Yint,11과 Yint,12의 실수부 값이 이론적으로 0이라는 사실을 이용하여 S-, Y-, Z-파라미터 행렬간의 상화관계를 이용하여 기생 저항 값을 쉽게계산할수 있다. 제시된 방법으로 기생 저항들을 구하고, 이 결과를 이용하여 내부 소자 등가회로를 구한 후에 40GHz까지 S-파라미터를 계산한 결과, 측정된 S-파라미터와 잘 일치하였다.

  • PDF

A PRECONDITIONER FOR THE LSQR ALGORITHM

  • Karimi, Saeed;Salkuyeh, Davod Khojasteh;Toutounian, Faezeh
    • Journal of applied mathematics & informatics
    • /
    • 제26권1_2호
    • /
    • pp.213-222
    • /
    • 2008
  • Iterative methods are often suitable for solving least squares problems min$||Ax-b||_2$, where A $\epsilon\;\mathbb{R}^{m{\times}n}$ is large and sparse. The well known LSQR algorithm is among the iterative methods for solving these problems. A good preconditioner is often needed to speedup the LSQR convergence. In this paper we present the numerical experiments of applying a well known preconditioner for the LSQR algorithm. The preconditioner is based on the $A^T$ A-orthogonalization process which furnishes an incomplete upper-lower factorization of the inverse of the normal matrix $A^T$ A. The main advantage of this preconditioner is that we apply only one of the factors as a right preconditioner for the LSQR algorithm applied to the least squares problem min$||Ax-b||_2$. The preconditioner needs only the sparse matrix-vector product operations and significantly reduces the solution time compared to the unpreconditioned iteration. Finally, some numerical experiments on test matrices from Harwell-Boeing collection are presented to show the robustness and efficiency of this preconditioner.

  • PDF

희박행렬의 기법을 이용한 대규모 측지망의 조정 (Adjustment Program for Large Sparse Geodetic Networks)

  • 이영진
    • 대한토목학회논문집
    • /
    • 제11권4호
    • /
    • pp.143-150
    • /
    • 1991
  • 이 논문은 대규모인 약 2,000점(미지수 약 4,000개)의 평면 측지망을 조정할 수 있는 프로그램을 개발하는데 목적이 있다. 데이터의 저장 및 관리에는 희박행렬(sparse matrix)의 기법이 사용되었으며, 관측방정식에는 RR(C)U (Row-Wise Representation Complete Unodered)방식, 정규방정식에는 RR(U)U(Row-Wise Representation Upper Unodered) 방식을 도입하고 해법에는 수정 Cholesky법을 적용하였다. PC 386에서 개발된 이 프로그램은 정밀 2차 기준점망인 테스트망에 적용되었으며, 2차원 배열을 사용한 Cholesky 분해법 및 직교분해법을 채용한 프로그램과의 상대적인 비교분석이 이루어졌다. 연구의 결과에서는 희박행렬의 기법이 기억용량의 면에서 뿐만 아니라 처리시간에 있어서도 극히 효과적인 기법임을 보여주고 있다.

  • PDF

Benchmark Results on the Linearized Equations of Motion of an Uncontrolled Bicycle

  • Schwab A. L.;Meijaard J. P.;Papadopoulos J. M.
    • Journal of Mechanical Science and Technology
    • /
    • 제19권spc1호
    • /
    • pp.292-304
    • /
    • 2005
  • In this paper we present the linearized equations of motion for a bicycle as a benchmark. The results obtained by pencil-and-paper and two programs are compared. The bicycle model we consider here consists of four rigid bodies, viz. a rear frame, a front frame being the front fork and handlebar assembly, a rear wheel and a front wheel, which are connected by revolute joints. The contact between the knife-edge wheels and the flat level surface is modelled by holonomic constraints in the normal direction and by non-holonomic constraints in the longitudinal and lateral direction. The rider is rigidly attached to the rear frame with hands free from the handlebar. This system has three degrees of freedom, the roll, the steer, and the forward speed. For the benchmark we consider the linearized equations for small perturbations of the upright steady forward motion. The entries of the matrices of these equations form the basis for comparison. Three diffrent kinds of methods to obtain the results are compared : pencil-and-paper, the numeric multibody dynamics program SPACAR, and the symbolic software system Auto Sim. Because the results of the three methods are the same within the machine round-off error, we assume that the results are correct and can be used as a bicycle dynamics benchmark.

Buckling analysis of functionally graded truncated conical shells under external displacement-dependent pressure

  • Khayat, Majid;Poorveis, Davood;Moradi, Shapour
    • Steel and Composite Structures
    • /
    • 제23권1호
    • /
    • pp.1-16
    • /
    • 2017
  • This paper is presented to solve the buckling problem of functionally graded truncated conical shells subjected to displacement-dependent pressure which remains normal to the shell middle surface throughout the deformation process by the semi-analytical finite strip method. Material properties are assumed to be temperature dependent, and varied continuously in the thickness direction according to a simple power law distribution in terms of the volume fraction of a ceramic and metal. The governing equations are derived based on first-order shear deformation theory which accounts for through thickness shear flexibility with Sanders-type of kinematic nonlinearity. The element linear and geometric stiffness matrices are obtained using virtual work expression for functionally graded materials. The load stiffness also called pressure stiffness matrix which accounts for variation of load direction is derived for each strip and after assembling, global load stiffness matrix of the shell which may be un-symmetric is formed. The un-symmetric parts which are due to load non-uniformity and unconstrained boundaries have been separated. A detailed parametric study is carried out to quantify the effects of power-law index of functional graded material and shell geometry variations on the difference between follower and non-follower lateral buckling pressures. The results indicate that considering pressure stiffness which arises from follower action of pressure causes considerable reduction in estimating buckling pressure.

Analysis of laminated composite plates based on different shear deformation plate theories

  • Tanzadeh, Hojat;Amoushahi, Hossein
    • Structural Engineering and Mechanics
    • /
    • 제75권2호
    • /
    • pp.247-269
    • /
    • 2020
  • A finite strip formulation was developed for buckling and free vibration analysis of laminated composite plates based on different shear deformation plate theories. The different shear deformation theories such as Zigzag higher order, Refined Plate Theory (RPT) and other higher order plate theories by variation of transverse shear strains through plate thickness in the parabolic form, sine and exponential were adopted here. The two loaded opposite edges of the plate were assumed to be simply supported and remaining edges were assumed to have arbitrary boundary conditions. The polynomial shape functions are applied to assess the in-plane and out-of-plane deflection and rotation of the normal cross-section of plates in the transverse direction. The finite strip procedure based on the virtual work principle was applied to derive the stiffness, geometric and mass matrices. Numerical results were obtained based on various shear deformation plate theories to verify the proposed formulation. The effects of length to thickness ratios, modulus ratios, boundary conditions, the number of layers and fiber orientation of cross-ply and angle-ply laminates were determined. The additional results on the same effects in the interaction of biaxial in-plane loadings on the critical buckling load were determined as well.

로지스틱 회귀모형에서 이변량 정규분포에 근거한 로그-밀도비 (Log-density Ratio with Two Predictors in a Logistic Regression Model)

  • 강명욱;윤재은
    • 응용통계연구
    • /
    • 제26권1호
    • /
    • pp.141-149
    • /
    • 2013
  • 로지스틱회귀모형에서 두 설명변수의 조건부 분포가 모두 이변량 정규분포라고 할 수 있다면 설명변수들의 함수로 표현되는 로그-밀도비를 통해 모형에 포함시켜야하는 항을 알 수 있다. 두개의 이변량 정규분포에서 분산-공분산행렬이 같은 경우에는 이차항과 교차항 없이 일차항만으로 충분하다. 상관계수가 모두 0이면 교차항은 설명변수의 분산과 관계없이 필요하지 않다. 또한 로지스틱회귀모형에서 로그-밀도비를 통해 이차항과 교차항이 필요하지 않게 되는 다른 조건들도 알아본다.