The Experts below are selected from a list of 45177 Experts worldwide ranked by ideXlab platform
Osamu Yamaguchi - One of the best experts on this subject based on the ideXlab platform.
-
the kernel orthogonal mutual Subspace Method and its application to 3d object recognition
Asian Conference on Computer Vision, 2007Co-Authors: Kazuhiro Fukui, Osamu YamaguchiAbstract:This paper proposes the kernel orthogonal mutual Subspace Method (KOMSM) for 3D object recognition. KOMSM is a kernel-based Method for classifying sets of patterns such as video frames or multiview images. It classifies objects based on the canonical angles between the nonlinear Subspaces, which are generated from the image patterns of each object class by kernel PCA. This Methodology has been introduced in the kernel mutual Subspace Method (KMSM). However, KOMSM is different from KMSM in that nonlinear class Subspaces are orthogonalized based on the framework proposed by Fukunaga and Koontz before calculating the canonical angles. This orthogonalization provides a powerful feature extraction Method for improving the performance of KMSM. The validity of KOMSM is demonstrated through experiments using face images and images from a public database.
-
a framework for 3d object recognition using the kernel constrained mutual Subspace Method
Lecture Notes in Computer Science, 2006Co-Authors: Kazuhiro Fukui, Bjorn Stenger, Osamu YamaguchiAbstract:This paper introduces the kernel constrained mutual Subspace Method (KCMSM) and provides a new framework for 3D object recognition by applying it to multiple view images. KCMSM is a kernel Method for classifying a set of patterns. An input pattern x is mapped into the high-dimensional feature space F via a nonlinear function Φ, and the mapped pattern Φ(x) is projected onto the kernel generalized difference Subspace, which represents the difference among Subspaces in the feature space F. KCMSM classifies an input set based on the canonical angles between the input Subspace and a reference Subspace. This Subspace is generated from the mapped patterns on the kernel generalized difference Subspace, using principal component analysis. This framework is similar to conventional kernel Methods using canonical angles, however, the Method is different in that it includes a powerful feature extraction step for the classification of the Subspaces in the feature space F by projecting the data onto the kernel generalized difference Subspace. The validity of our Method is demonstrated by experiments in a 3D object recognition task using multiview images.
Kazuhiro Fukui - One of the best experts on this subject based on the ideXlab platform.
-
the kernel orthogonal mutual Subspace Method and its application to 3d object recognition
Asian Conference on Computer Vision, 2007Co-Authors: Kazuhiro Fukui, Osamu YamaguchiAbstract:This paper proposes the kernel orthogonal mutual Subspace Method (KOMSM) for 3D object recognition. KOMSM is a kernel-based Method for classifying sets of patterns such as video frames or multiview images. It classifies objects based on the canonical angles between the nonlinear Subspaces, which are generated from the image patterns of each object class by kernel PCA. This Methodology has been introduced in the kernel mutual Subspace Method (KMSM). However, KOMSM is different from KMSM in that nonlinear class Subspaces are orthogonalized based on the framework proposed by Fukunaga and Koontz before calculating the canonical angles. This orthogonalization provides a powerful feature extraction Method for improving the performance of KMSM. The validity of KOMSM is demonstrated through experiments using face images and images from a public database.
-
a framework for 3d object recognition using the kernel constrained mutual Subspace Method
Lecture Notes in Computer Science, 2006Co-Authors: Kazuhiro Fukui, Bjorn Stenger, Osamu YamaguchiAbstract:This paper introduces the kernel constrained mutual Subspace Method (KCMSM) and provides a new framework for 3D object recognition by applying it to multiple view images. KCMSM is a kernel Method for classifying a set of patterns. An input pattern x is mapped into the high-dimensional feature space F via a nonlinear function Φ, and the mapped pattern Φ(x) is projected onto the kernel generalized difference Subspace, which represents the difference among Subspaces in the feature space F. KCMSM classifies an input set based on the canonical angles between the input Subspace and a reference Subspace. This Subspace is generated from the mapped patterns on the kernel generalized difference Subspace, using principal component analysis. This framework is similar to conventional kernel Methods using canonical angles, however, the Method is different in that it includes a powerful feature extraction step for the classification of the Subspaces in the feature space F by projecting the data onto the kernel generalized difference Subspace. The validity of our Method is demonstrated by experiments in a 3D object recognition task using multiview images.
Wentao Liu - One of the best experts on this subject based on the ideXlab platform.
-
3 d full time tem modeling using shift and invert krylov Subspace Method
IEEE Transactions on Geoscience and Remote Sensing, 2020Co-Authors: Jianmei Zhou, Wentao LiuAbstract:The 3-D time-domain electromagnetic (TEM) modeling is computationally expensive. Shift-and-invert (SAI) Krylov Subspace technique has proved efficient for the modeling of OFF-time TEM data. However, the ON-time data, which also involves useful Earth conductivity information, have not been well treated. In this article, we extend the SAI Krylov Subspace Method to deal with the ON-time data and obtain a complete algorithm for modeling of the full-time data. To deal with the ON-time problem, while retaining the whole framework of an SAI-style Method, we have developed a new time integration Method, based on the exponential trapezoidal rule, to help discrete the integral terms into matrix exponential function. Next, the solution of this matrix exponential function is obtained by constructing a new type of SAI Krylov Subspaces. Numerical results for typical transmitting waveforms, such as half-sine, versatile-time-domain-electromagnetic (VTEM), and MULTIPULSE, demonstrate that the novel algorithm is accuracy for full-time TEM modeling.
-
3d transient electromagnetic modeling using a shift and invert krylov Subspace Method
Journal of Geophysics and Engineering, 2018Co-Authors: Jianmei Zhou, Wentao LiuAbstract:We present a fast and efficient algorithm for simulation of a three-dimensional (3D) transient electromagnetic (TEM) response using a modified shift-and-invert Krylov Subspace Method. The mimetic finite volume Method with a staggered grid is carried out for spatial discretization of the time-domain Maxwell's equations. The transient electromagnetic response then can be expressed as a matrix exponential function with an analytic initial magnetic field for a step-off loop source. The shift-and-invert Krylov Subspace Method can be used to solve the matrix exponential function. However, it requires solving dozens of large sparse linear equations at every time point to reconstruct the Krylov Subspace, which makes the conventional shift-and-invert Krylov Subspace Method time-consuming. By analyzing the characteristics of the optimal shift and shift-and-invert Krylov Subspace dimension in detail, we proposed a fast substitute approach to obtain the optimal shift and Subspace order by using single optimal shift and constant Subspace order with a useful stopping criterion, and developed an efficient modified shift-and-invert Krylov Subspace Method. Only one LU factorization of a shift coefficient matrix and hundreds of times backward substitutions are required to obtain the results of the TEM modeling data. Time savings are considerable, and this approach makes it possible to compute the response at any time point in the given time interval within the given residual easily and accurately. This is illustrated by using synthetic examples both in layered models and in a 3D complicated model.
Leader Chen - One of the best experts on this subject based on the ideXlab platform.
-
application of the krylov Subspace Method to numerical heat transfer
Numerical Heat Transfer Part A-applications, 1996Co-Authors: Hsiaowen Lin, Leader ChenAbstract:Five algorithms of the Krylov Subspace Method and three preconditioning techniques are presented. Applications of the Krylov Subspace Method were illustrated in solving four example problems. The matrix inversion example showed that the Krylov Subspace Method can be viewed as a direct Method. It was demonstrated that, in solving an example problem of heat conduction with a constant source term, incomplete lower-upper (LU) (ILU) decomposition and polynomial preconditioning could substantially reduce the number of iterations. Also, linear relationships were observed between the iteration number and the equation number. It was also found that double preconditioning using a fifth-order polynomial and ILU decomposition could further reduce the computing time. The generalized minimal residual (GMRES) Method with double preconditioning was compared with such iterative Methods as alternating direction implicit (ADI) and Gauss-Seidel. The results showed that the GMRES Method only required fractions of the computing time required by ADI or Gauss-Seidel Method. Example problems of heat conduction with an Arrhenius source term and cavity flow were also solved by the GMRES Method with preconditioning. Converged solutions were obtained with one or two iterations for the momentum equation of cavity flow considered, and three to six iterations for the pressure Poisson equation. Furthermore » effort seems to be warranted to explore the implementation of the Krylov Subspace Method for the finite difference modeling of heat transfer and fluid flow problems.« less
Jianmei Zhou - One of the best experts on this subject based on the ideXlab platform.
-
3 d full time tem modeling using shift and invert krylov Subspace Method
IEEE Transactions on Geoscience and Remote Sensing, 2020Co-Authors: Jianmei Zhou, Wentao LiuAbstract:The 3-D time-domain electromagnetic (TEM) modeling is computationally expensive. Shift-and-invert (SAI) Krylov Subspace technique has proved efficient for the modeling of OFF-time TEM data. However, the ON-time data, which also involves useful Earth conductivity information, have not been well treated. In this article, we extend the SAI Krylov Subspace Method to deal with the ON-time data and obtain a complete algorithm for modeling of the full-time data. To deal with the ON-time problem, while retaining the whole framework of an SAI-style Method, we have developed a new time integration Method, based on the exponential trapezoidal rule, to help discrete the integral terms into matrix exponential function. Next, the solution of this matrix exponential function is obtained by constructing a new type of SAI Krylov Subspaces. Numerical results for typical transmitting waveforms, such as half-sine, versatile-time-domain-electromagnetic (VTEM), and MULTIPULSE, demonstrate that the novel algorithm is accuracy for full-time TEM modeling.
-
3d transient electromagnetic modeling using a shift and invert krylov Subspace Method
Journal of Geophysics and Engineering, 2018Co-Authors: Jianmei Zhou, Wentao LiuAbstract:We present a fast and efficient algorithm for simulation of a three-dimensional (3D) transient electromagnetic (TEM) response using a modified shift-and-invert Krylov Subspace Method. The mimetic finite volume Method with a staggered grid is carried out for spatial discretization of the time-domain Maxwell's equations. The transient electromagnetic response then can be expressed as a matrix exponential function with an analytic initial magnetic field for a step-off loop source. The shift-and-invert Krylov Subspace Method can be used to solve the matrix exponential function. However, it requires solving dozens of large sparse linear equations at every time point to reconstruct the Krylov Subspace, which makes the conventional shift-and-invert Krylov Subspace Method time-consuming. By analyzing the characteristics of the optimal shift and shift-and-invert Krylov Subspace dimension in detail, we proposed a fast substitute approach to obtain the optimal shift and Subspace order by using single optimal shift and constant Subspace order with a useful stopping criterion, and developed an efficient modified shift-and-invert Krylov Subspace Method. Only one LU factorization of a shift coefficient matrix and hundreds of times backward substitutions are required to obtain the results of the TEM modeling data. Time savings are considerable, and this approach makes it possible to compute the response at any time point in the given time interval within the given residual easily and accurately. This is illustrated by using synthetic examples both in layered models and in a 3D complicated model.
